I only started reading ACX after the move from SSC. I have gone back and read the top posts from SSC but still felt like I was missing out on some of the old classics. I built a website to resurface old content from blogs and track your progress through the backlog. Let me know if you have any suggestions of other blogs / content you’d like to see, hope you find it helpful!
Calling all research scientists! I'm trying to build a tool to help scientists with their work. I have this survey to gather ideas: https://tally.so/r/mBYBJQ
(To be clear, this is just an amusing story, and not a request for medical advice)
I have just had a message from my doctor asking me to come in for an asthma checkup.
Now, while I have been mildly asthmatic for all my life, I only needed an inhaler for about 6 months starting in 2020. So, I guess I’m going to report that it seems fine now. And yes, I think it is a very very suspicious coincidence that I only had respiratory difficulty during the covid pandemic. Can’t prove that it was Covid, of course, but …
The difficulty in making nuclear weapons isn't so much the theory as it is sourcing the fissile material--as we've seen and are seeing again, nuclear fuel processing facilities are large and specialized; it's not really possible to keep a nuclear program secret, given how tightly the mines are controlled (so tightly that French chemists detected the existence of a natural fission reactor based on the composition of ore from specific sites).
Put in a bulk order for cough medicine, or diesel fuel and fertilizer, and see how that works out--and those things are much more common than yellowcake.
Yes, I agree the idea of any domestic threat from nukes (now generalized to "CBRN") is overblown, and primarily fearmongering to manufacture a pretext for some political goal.
I am less sure about what foreign states are limited by. Is it just the threat of America bombing their facilities when they do some simple things at sufficient scale, or is there some technical capability it'll take them decades to develop on their own?
The UN has inspectors, and yes, the US has enforcers. Refining uranium fuel requires centrifuges and chemical handling equipment--but the main thing is getting the uranium itself. Even if a rogue government managed to build the necessary facilities in secret--not impossible, I guess, but it'd require assembling enough people with the right education who aren't also on some watchlist--they'd have to source the fuel. There aren't that many uranium mines in the world, and they're closely monitored.
Yeah, that's basically the Strangelove line, right? "The whole point of a doomsday machine is lost if you keep it a secret! Why didn't you tell the world, eh?"
If they're trying to use nuclear weapons as a deterrent and bizarrely pretend their weapons don't work, they invite an attack.
Given that Iran has not demonstrated the ability to actually build a bomb, it's probably not that bad. The other nuclear powers would all respond with a "Ha! Nope, wrong, not even close" and nobody would have any idea how seriously to take it.
I think that the number of entities out there with the desire and ability to build nuclear weapons, who haven't already built nuclear weapons, is pretty small. Countries like Germany or Japan or even Taiwan probably could build them, but know they're more trouble than they're worth. Entities like Al Qaeda (or whatever they are these days) would love to build them but are never going to get the money, materials and expertise. Sophisticated well-funded supervillain networks don't actually exist. The only people who could benefit from nuclear weapons are reasonably-sophisticated international pariah states -- Iran, North Korea, and that's about it?
Intuition suggests that the US (and the other nuclear powers) keeping this tech as secret as they do, far beyond regular military secrecy, means THEY think it's still dangerous if it became public.
I have heard it said that nuclear weapons clearance aren’t that hard to get, compared to e.g. codebreaking. We might infer that the US doesn’t try all that hard to keep nuclear technology,ogy secret, compared to other things. They would appear to be relying on the difficulty of getting the fissile material.
On monday there is an AI safety protest in London 5PM at Granary square/Deepmind headquarters (https://lu.ma/bvffgzmb) organized by mostly Pause AI. Come join :).
Even if you don't support the direct message of the protest it might still be worthwhile to attend to overall make AI safety a more visible topic. If you have any questions you could join a meetup on Friday evening (dm me for details) or you can also still apply to https://pausecon.org/.
I have recently criticized rationalists, but please note I do not mean anything wrong, it is 10+ years of frustration bursting out. At any rate please reply:
>Many rationalists are foodies for the mind.
Strangely, I have the complete opposite experience with them. They are the most anti-intellectual people out here. They are not interested in ideas as such. They are interested in evidence, not ideas. That is super anti-intellectual as it means you are not allowed to think unless someone else ran an experiment...
When I tell them "You know maybe a farmer in rural Thailand might be happier than a lawyer in NY, because I know a Westerner lady who used to live there and was happy?" they are NOT interested at all.
They tell me to bring evidence. Ideally a scientific study, or at least anything in NUMBERS. So that they can plug the number into their beloved Bayesian **replacement for actual thinking** and get an automatic result. One of their leaders is literally called PutANumOnIt.
They do not want to think, they want an algorithm, a method to think for them. They are more anti-intellectual than any Facebook guy spreading anti-science crap, because those at least spent at least 30 seconds to use their OWN brains, asking themselves "does this make sense to me?", and not a method or algorithm as a replacement. Yarvin was right at least in this, that they are "automatists" who do not have a concept of "reasonable judgement".
These people seriously think the above question can be solved by putting a number on it. Find some sort of a Happiness Measurement Index and decide it that way. They seriously think when someone goes around asking people "On a scale from 1 to 10, how happy are you?" is anything like a reliable number. So those people who do that find that Finland is the happiest place ever. I talk with actually Finnish people and they laugh and say it is not so. They say the Finnish cultural norm of "sisu", basically stoicism, means that if you do not have a strong reason to complain, you do not complain, stoicism, stiff upper lip, so if a Finnish person says I am 7/10 happy it means they feel they OUGHT to be 7/10 happy because they have no strong reason to complain. They can still be as depressed and miserable as anyone else. Someone using common sense, not numbers, would probably find the beach people in Rio the Janeiro are the happiest. They certainly smile really a lot.
Apologies for being blunt, but I read this and think you simply don't understand much about the world at all. Like this:
"They tell me to bring evidence. "
Obviously. You cannot POSSIBLY form any accurate conclusions about the world without evidence, any more than you can know whether it's safe to cross the street without looking or listening[1]. Your brain is not magic. It can't conjure correct information from nowhere. If you want to know if it's raining out or if your pot of water is boiling or whether your car has been stolen, you have to *go look.* And if you want to know the answer to some less localized question, whose answer is spread among millions of people across millions of square kilometers--like how happy rural Thai farmers are--you have to actually go observe a whole bunch of those rural time farmers *in some fashion.*
Now, you seem to have a particularly large chip on your shoulder about numbers and calculations. Not all useful evidence is numeric, but a lot of it sure is. Sometimes our ape-brains have trouble understanding this, but at its fundamental level reality seems to be really, really, extremely well-suited to being described by numbers. That doesn't mean all attempts to describe pieces of reality by numbers are equally good. Asking people to rate their happiness on a scale of 1-10 is a really, extremely crude and uninformative and low-accuracy way of interrogating reality, as you correctly ascertain. But basing your opinion on the description of a single person in a single set of circumstances is even cruder. Finding a way to answer important questions--like how happy Thai farmers are compared to American lawyers--in a way that *both* captures the richness and detail of the human experience *and* can scale to the size that you need to be any good[2] is quite a difficult challenge. Most rationalists I've read or talked to seem to be rather more aware of this fact that you do. And I'm sure lots of them *would* be interested in the question, they're just not interested in your anecdote because they're well aware of how useless it is by itself.
[1] Which is, in fact, a form of evidence gathering.
[2] A quick bit of googling tells me there 10s of millions of Thai farmers and a million or so American lawyers, so your one anecdote is *really* slim evidence.
"Sometimes our ape-brains have trouble understanding this, but at its fundamental level reality seems to be really, really, extremely well-suited to being described by numbers. "
I think the hard science-soft science distinction exists for a reason. Namely the world of objects is well-suited to that, the world of people less so. The reason for that is simple: measurement. Objects are well measurable, while with people there are always a lot of subjective stuff that is not well measurable.
How old is utilitarianism as a philosophy? At least 200 years and no one ever figured out how to measure utilons...
1. There are no "1.5 kids" meaning that statistics are not always applicable to real life conditions. They are a guide based on observing populations over time, but the reality isn't that: individually it varies and only collectively can it be used as a rule of thumb.
There is also the MMO version, which is "if its not 100% its 0%"
meaning no matter if it says 95% success, you can and will go 1 in 5 on occasion, and if you need it for a specific occasion its unreliable.
2. objective materialist reductionism leads to it; if the culture believes in god it loves cathedrals, in this case it loves quantification. This is not them
as much as the culture, loves its objective science.
3. Control. so much of this is trying to find control or meaning through technique and knowledge. The idea of mystery, fate, and randomness is terrifying, and people always seek ways to gain some measure of power over it.
rationalism loves its one weird trick solutions to common problems as long as they are intellectually elegant, and there is some power in knowing when Jesus may come back. otherwise its terrifying.
I think they fail to appreciate that the evidence can't support the right idea if you don't have the right idea...having as many hypotheses as possible is you'd, because it increases your chances of having the right one. The problem is exacerbated by thinking in terms of *ideal* Bayesianism, where the hypothesis-formation problem has been waved away.
> They are not interested in ideas as such. They are interested in evidence, not ideas. That is super anti-intellectual as it means you are not allowed to think unless someone else ran an experiment...
We need to make reading the Sequences mandatory, and maybe require a re-certification every three years. (Just kidding.) First, there is a difference between "there is evidence against X" and "there is no evidence either for or against X". I think Scott posted an entire article on this topic once. Second, there are different degrees and kinds of evidence (anecdote, survey, experiment, meta review), and although stronger evidence is better, that doesn't mean we should ignore everything else.
> I talk with actually Finnish people and they laugh
See, that's your evidence. They laugh, it means they are happy. :D
>When I tell them "You know maybe a farmer in rural Thailand might be happier than a lawyer in NY, because I know a Westerner lady who used to live there and was happy?" they are NOT interested at all.
Do you apply this logic to other things in your life? If I tell you that on the one hand, I have a scientific study which has a bunch of numbers showing that the flu vaccine reduces your mortality risk by X%, but on the other hand, I know a guy who got the flu shot and had a heart attack the next day, are you going to become an anti-vaxxer because you trust anecdotes more than evidence?
Like, it's one thing if you want to dunk on social science as not being as rigorous as it claims to be - lots of people do that. But your post seems to be dunking on the very idea of gathering evidence in a systematic way!
>They seriously think when someone goes around asking people "On a scale from 1 to 10, how happy are you?" is anything like a reliable number.
But when *you* ask some Westerner lady who used to live in Thailand if she was happy there, that's reliable, right? How is that any different from what the scientists are doing?
I actually would trust anecdotes more than science, because anecdotes come from trusted sources, aren't an attempt to manipulate me and can be personally verified.
"The Science" as exemplified by Doctor Fauci, might be corrupt, has plenty of incentives to lie and is by now very politicized on hot button issues.
"Do you trust the Science"? is not a referendum on the scientific process. It's a referendum on the integrity of the people and institutions running the process, and they have lost all trust through their actions.
If some random dude I know had a heart attack the day after taking a vaccine, that's a data point that absent anything else suggests strong correlation. When I keep hearing those stories, it's more and more data points and since I just don't trust the institutions to be honest on this topic, I simply think it's more likely than not that there are actual problems.
""Do you trust the Science"? is not a referendum on the scientific process. It's a referendum on the integrity of the people and institutions running the process, and they have lost all trust through their actions."
This is an awesome point. The problem is the following. the "anti-establishment" does not seem much more credible either. And I don't have the time to verify everything myself.
"I actually would trust anecdotes more than science, because anecdotes come from trusted sources, aren't an attempt to manipulate me and can be personally verified."
This presumes quite a number of things:
1. You only accept anecdotes from trusted sources (maybe do, but most people don;'t)
2. Every single one of the sources you trust is actually reliable (maybe yours are, but most people's aren't)
3. All of the important anecdotes are both possible and practical to personally verify (maybe some are, but it's highly unlikely that all will be).
Perhaps you're not used to thinking like this, but you CAN, in fact, personally verify the science. There's quite a lot of science that you can verify directly with pretty modest effort. One of my favorite YouTube videos involves the creator stumbling on some flat Earth forum where people were talking about this especially long, straight lake that could be used to show that the Earth was flat. He realize that he lived a 90 minute drive from the exact lake they were talking about, drove out there with a video camera and tripod, and made a video where he lowers the camera and you can see the lower trunk of a tree on the far shore sinking below the horizon in real time. SCIENCE!
Probably you never gave much credence to flat Earthers to begin with[1], so perhaps this seems silly or pointless. But the key is, you can do this all over the place if you want to. Physics, astronomy, biology, chemistry, they all make millions of predictions that you can just...go out there and test. Properties of moving objects, behaviors and life cycles of plants and animals, results of mixing various chemicals together. Some you can do walking down the street. Some you can do with simple equipment on your kitchen table or garage. Some you might need to drive out to a local lake for. And then some require the resources of a large institution. But if you're determined, you can get that. With enough patience and perseverance, there's no conclusion reach by modern science that you can't test (or at least be a part of testing).
The only limit is that there's only one of you. You can test anything, but you can't test everything. Ultimately, you have to pick a lane and rely on other people to test some of the things. But you can read about what they did. You can check their calculations, talk to the people that were involved in their work, read other papers on the same topics--including responses and even refutations sometimes. There are mistakes, for sure and even deliberate deceptions sometimes. Everyone involved is human, after all. But it would be hard to imagine a field of human endeavor that better lends itself to skeptical parties *checking its work* than science[2].
Honestly, if you genuinely don't think you can trust anything you hear except anecdotes from people you consider reliable, then I have to wonder how you can possibly have such confidence that the science you're criticizing *is* untrustworthy, corrupt or even, y'know, wrong. I absolutely guarantee that not all of your evidence *against* it comes from trustworthy anecdotes.
"If some random dude I know had a heart attack the day after taking a vaccine, that's a data point that absent anything else suggests strong correlation. "
This is correct. It is a data point. It is ONE data point. Did you stop to consider how much weight you should put on it? If you know 500 people well enough to hear about something like this, and only one has this happen, then the *absolute highest* your assessment of vaccine related mortality should be is 1/500[3]. Which is the lowest non-zero number your methodology should recognize, since you can't possibly trust any report of a vaccine related death that comes from outside the circle of people you know. Wait, hold on...
"When I keep hearing those stories, it's more and more data points..."
aaaaaannnd, not you've lost it. This is wrong, wrong, wrong, so extremely wrong. If only anecdotes from trusted sources are evidence, then casting a wide net for anecdotes on a topic of interest is NOT ALLOWED. Unless you have an extremely wide or alarmingly heart-attack prone social circle, most of these stories must NOT be events happening to or directly witnessed by people you know. You've just thrown the *entire purported point* of your novel methodology out the window.
There are very good reasons science doesn't operate this way. When you let the stories filter to you and form you opinion based on that:
1. You not only lose all ability to verify the stories, but you actively invite yourself to believe false stories. A lie can travel halfway around the world while the truth is still getting its boots on, after all.[4]
2. You don't have any actual idea of how common the events you're hearing about really are. If you hear 10 stories of suspiciously timed heart attacks, you don't know if that represents all the reported events from a population of 1000, 100,000 or 10,000,000.
3. Related to 2, you actively invite confirmation bias by deliberately listening to the things that confirm what you already believe, and seldom to those that don't. You're not going to hear many stories about how someone got the vaccine and was perfectly fine, are you?
You can try to hold to the brand of radical skepticism where you don't believe *anything* you don't see or hear reported (as first-hand experience) by a trusted personal friend. But that's what you actually do--that is, if you're doing it *right*--you should simply stop holding opinions on a bunch of things. You and your trusted personal friends can't possibly pass judgement on any branch or field or institution of science, since between you you have certainly not had enough direct personal experience to know much of anything about what that large, mostly distant group of humans is doing.
[1] Though to be intellectually consistent with everything else you've written, it seems like you should.
[2] Perhaps the open-source software movement might be ahead of it, if only because verifying any individual claim is usually easier.
[3] But really, you ought to estimate the chance of the timing being a coincidence and factor that it. Because it very much could be.
[4] In more formal and explicit language, your sample set will be dominated not by the things that happen most frequently, but by the reports that spread the farthest and the fastest. It's very easy for false stories to beat out true ones on that score, since they can be as compelling as they like without being constrained by the mundane details of reality.
It might be surprising to you, but I actually do have an internally consistent philosophy (which I'm aware like maybe 1% of people have). I do in fact, not hold opinions on many topics I'm interested in. These are mainly topics where there is a 'factual clash'; very simply different sets of experts disagree and the issue is politicized.
For instance, I have no idea to what degree climate change is either a problem or caused by humans and don't think it's worth my time finding out (as I think the costs and difficulty coordinating action to halt climate change are so prohibitive/unrealistic that it's better to focus on mitigation, which will cease to be a partisan issue if and when there are immediate and serious problems. Similarly, I think if a pandemic or medical emergency is sufficiently serious, there will be bipartisan consensus even in the USA. That it wasn't anywhere close for COVID speaks volumes.
When scientific issues have no political or financial tinge I haven't the slightest difficulty believing what the scientists say, because I know they are competent and I'm not. That reasoning goes out of the window with politicized issues. Also you have seem to have misunderstood my position on things like vaccines. I'm simply agnostic on the possible harms. My main reason for considering vaccine harm (COVID particularly) theories credible is the incredible amount of suppression, censorship and flat out lies from establishment sources on this topic. I have no idea about the underlying science, but it doesn't take a genius to understand incentive structures.
By "more and more data points" I don't mean millions of things I read on X. I simply mean even 3 or 4 cases I clearly know of. You need to understand that I consider studies from American universities or pharmaceutical bodies on these topics to have literally zero persuasive value. It's at best an exercise like asking me to watch a sleight of hand expert and trying to use my inability to debunk him as evidence of the existence of
magic.
When you know somebody is an expert on a topic you know anything about, but fundamentally distrust his character and motives, nothing he says has any meaning to you. That's where I, and massive percentages of the American population stand on liberal dominated institutions and their narratives.
How old are you? I am 47 and I am a bit confused about the current historic period. 30 years ago a lot of people were like "don't trust Big Pharma, they are interested in profits, not healing people". And this opinion was generally categorized as politically left. Today "trust science" is categorized as politically left and the anti-vaxxers are categorized as politcally right-wing.
This is truly strange, I think everybody who is a little older has noticed this, and this does not get talked about at all.
BTW I am willing to entertain the notion that for the most part the whole left-right thing is just show for the people, and people with real power just do not think like that.
For example if of two ex Harvard classmates one is the CEO of a corporation and another is working at a regulatory agency, then I just cannot take the whole "more market, less state" vs. "more state, less market" thing seriously. Clearly they are not any kinds of enemies to each other.
The left-right thing in the sense of rich vs poor class tension is indeed a far less relevant lens nowadays.
It's traditional culture vs globalists that's to some extent supplanted this. Right vs Left terminology has persisted, but the right is more about traditional cultural values (including values that were considered liberal even 15-20 years ago), opposition to immigration that changes Western countries and opposition to woke values (language changes, alphabet stuff, DEI discrimination etc).
Meanwhile capitalism has been swamped as it's now controlled capitalism marked by regulatory capture where almost is everything is illegal or requires permission and regulators have enormous discretion to allow egregious behaviour or crack down on actually legal behaviour and in practice both happen simultaneously. This is why dumb people in the West who don't realize this is happening tend to pick a side on whether they want more or less regulation, when actually the entire system is corrupt in a way that is instantly understood by Indian or Chinese people, but naive Americans just don't get their high trust world and culture has already been dismantled.
High trust cultures can accomplish great things, but they require absolute xenophobia and racism towards those who don't share its values and culture, this last point having been forgotten somewhere in the last many decades.
the problem is that yes, in aggregate there a very low risk to vaccines. There is only one you though, and it is possible for you to draw the short straw. This uncertainty will always be there.
this is why you hang out after the vaccine so they can be sure you dont get a reaction right away because for all the statistics they dont know who or when.
its kind of related to being morbid: we are all going to die at some point, 100%. but focusing on thar becomes unhealthy. it gets worse with direct or close experience.
like its one thing to know average life expectancy is 76 years, another to read comments in a youtube video to hear the "taking the hobbits to isengard" guy died at 52 and other guy you followed at 56.
I think this is related to "garbage in, garbage out" problem. If you collect unreliable data, it does not matter how sophisticated statistical techniques you apply to them afterwards.
The problem with rating something on a scale from 1 to 10 is that it depends on what range of experience are they familiar with. https://xkcd.com/883/
When my kids were in kindergarten, any negative experience was "the worst day ever". On the opposite end, sometimes a health problem expanded my pain scale.
Is it possible that if your life in general sucks and you have never been happy, then you would report days that suck less as 7, while someone whose life is full of fun and adventure would report an insufficiently exciting day as 3?
That sort of thing only matters if it's a systemic issue - e.g., if you think that *most* people in Thailand under-report their happiness because they're used to living fun and adventurous lives. Otherwise, with a big enough sample all the outliers will average out.
You could also check if this effect exists by comparing to socioeconomic indicators - e.g., do rich and poor people in the same country report the same level of happiness despite different material circumstances?
I think that in this way rationalists are a lot like foodies - foodies often try to maximize some measurable aspect of food, which comes at the expense of the rest of the experience.
For instance, getting a perfectly cooked steak (according to some foodies) means maximizing the amount of pink in the center while minimizing the gray on the edges. The best way to maximize the pink is to sous vide the steak at 130F for an hour, then dry off the meat, sear, and serve. I used this method for years, and it produced some pretty good results.
But it also requires you start an hour in advance, and after the sous vide you have to pull this gross gray wet steak from a bath of meat juices. When I stopped doing this, I found that I didn't even care about the grayness around the edges - it's a nice variation in flavor and texture - and the crust was better, and most importantly the experience of steak preparation was far more humane. Pull steak from fridge, salt, pepper, pan, flip, serve.
Of course, you can argue endlessly about preferences, and don't let me stop you from micromanaging your steak in whatever way pleases you. I still follow JKLA's advice most times, but I often skip any step which seems like a huge pain in the ass or gross or weird. But I've become much more pragmatic about whether maximizing the measurable qualities of my food will actually give a better lived experience than just slapping down some meat and calling it a meal.
I think you're greatly underestimating how much thinking it takes to cook the numbers they're pretending to use to come to the conclusion they wanted to.
What fraction of the subjects of a government have to have the vote for that government to qualify as "democratic"? Or is that even how you use the word at all?
I'd say this is mistaking the tool for the result. Having the people vote on a government is a means to an end. In particular, it doesn't matter at all if 100% of the populace votes if their only choice is between Kodos and Kang. Meanwhile I could imagine a government doing an admirable job of "governing democratically" in all other ways that mattered while holding no elections at all: instead they would select a very scrupulously randomized 1% sample of their whole population and collect information about their preferences in far more depth than any voting process allows[1].
Questions I'd consider more pertinent:
1. How often do broadly popular policies get enacted and how quickly after they become broadly popular?
2. How often do broadly unpopular policies get enacted? How long do they stay in effect (presuming they remain unpopular)?
If you want to allow for the role that the judgement and expertise of leaders (often touted as a benefit of representative democracy) have on deciding good policy or implementing it well, you might add the following:
3. How often do the unpopular policies that get enacted increase in popularity once they're actually in force.
4. How often to the popular policies that get enacted decrease in popularity once they're actually in force.
Though with both of these it can be hard to tell the difference between "the public didn't know what was good for them" vs "the leaders did a poor job of implementing a popular policy, or a good job of mitigating the flaws of an unpopular one."
Also, like most things, this schema is subject to Goodhart's Law: a government with sufficient control over the information of its people may be able to simple manufacture consent for whatever policies it wants.
[1] Damn, I literally just thought of this off the cuff, and now I really want to see it tried. I'm sure it would find *some* what to fail horribly, but I really want to know how.
I noticed a few threads in these comments discussing the legitimacy of the government of Iran. I just wanted to note that since 2020, Regimes of the World has flip-flopped between characterizing Iran as a Closed Autocracy (the lowest ranking, of four) and an Electoral Autocracy (the second-lowest). This puts it in a peer group with countries such as Nicaragua or Kuwait, below Indonesia or Ukraine but above Afghanistan or China (where "above" means "more democratic"). Or alternatively (very roughly) around the 20th percentile of openness.
As for the other two belligerents in the war, the US is rated as a Liberal Democracy (the highest category) and has been since the civil rights era, while Israel was just recently downgraded to Electoral Democracy (the second-highest category).
Legitimacy has nothing to do with being a democracy or what not. Iran's govt is accepted by its people (there is no civil war ongoing) and by the international community (Iran has diplomatic relations with plenty of countries). Even those countries that lack diplomatic relations, they regard the present govt as THE govt.
Not gonna lie I'm pretty disappointed with the nominies. My favorite was The Sermon On the Mount by J.K. which offered a totally fresh modern perspective for me, as someone who's only mildly interested in the Bible, mostly for historical context. Bishop's Castle was my 2nd, but I give Jim Bishop at least as much credit for that as I do the author.
The ACX Commentariat was too meta and circlejerky in a self flagellating way, and The Men Are Not Alright just felt like San Francisco Goldilocks. They definitely weren't the worst written, but seemed thoroughly middle-of-the-pack, and if you go for a hot take culture war topic like [half the human species has something categorically wrong with them], I'm sorry but you gotta do better than that.
Thanks for the recommendation of the Sermon on the Mount article. I agree that it was fantastic, I got a sense while reading it that it was gradually forming one of those foundational building blocks of knowledge that would stick with me for much of my life.
JK, if you’re reading this, I strongly encourage you to post it elsewhere so it can live on and be read easily.
I was not very convinced that the Sermon on the Mount was intended that way at all. Jesus had some radically pacifist teachings elsewhere too... I really think the intent was "cooperate-bot till you die." That's also backed up by the entire structure of Jesus' mission, which was radical self-sacrifice for the unworthy.
Trying to reinterpret it to something that makes more game theoretic sense just removes its coherence with the rest of the New Testament. All of the apostles also seem to have thought the calling meant radical, unilateral pacifism, and the tradition of the early church elders is that they clung to that radical pacifism to the death. (True or not, it's evidence that everyone thought the meaning was the traditional one, not JK's sanified version.)
Yeah, it was an interesting review, and the interpretations are clever and fun to engage with but indeed seem to be unconvincing.
And as to why Christianity won: I think they won precisely because they abandoned the original Christian ideal and became exactly like every other large powerful social structure. To use J.K.'s metaphor, they struggled as long as they were trying to burn the boot and became powerful once they decided to actually wear the boot. *shakes head* Humans, amirite?
It does look like it's on its way out too: the world on average is trending more secular, and societies that aren't trending that way are leaning more and more heavily towards Islam. Though 2000 years was not a bad run at all, and it's not over yet either, just very slowly waning imo.
It works on multiple levels though. Literally "help someone else in need because it feels good" or alternately "don't give them the satisfaction of giving up at the earliest chance." Or again on the second order "does this seem implausibly decent? So be it."
You're not a covert asshole carrying out some dubious mission for the sake of "the greater good". It's trying your best to "do the right thing" on multiple dimensions at once, and having faith that's how it actually plays out.
Agreed. As interesting as JK's interpretation is, and as much as I enjoyed reading it, it's clever/cute to a point that raises alarms and puts me in skeptic mode.
But isn't it kind of funny to think of Jesus trolling the ruling class so hard that they killed him for it, then getting resurrected and still being like "jeez all I said was turn the other cheek! What was their problem" and for thousands of years Christianity spreads as people are like "yeah c'mon, you can't kill a guy just for saying 'go the extra mile'"
It's kind of funny, yeah, but... the success of Christianity is weird and important. Taken seriously (and I think it was intended seriously), this is an attempt to shoehorn it into modern perspectives... which I think is less interesting and less useful than grappling with the weird artifact as it actually existed.
[chatgpt seems to give contradictory opinions for the sake of agreeing with me]
[here's where I "call-out its behavior" to see if it will own it]:
Dear Chatgpt,
I stated practically the opposite in a different temporary chat:
"I'm annoyed at season 5 of 'You' because it emphasized this idea of 'Joe enjoyed killing' which seems at odds with the premise, plot and character development of the show from the beginning. it seems to simplify the show's complexities"
and chatgpt enthusiastically explained why *that* was correct instead.
interesting to learn this about you!
ChatGPT said:
Thought for a few seconds
That’s a sharp observation! Each ChatGPT session builds its answer around the prompts and context you provide in that moment [... ... ...]
They don’t remember your previous prompt, every prompt is handled by a different process. The context window is either sent by the API each time as metadata, or some id returned the previous response is sent in the request which allows the new process to read the chat from a database.
Reminds me of the story of a monarch casually mentioning at dinner that he likes the eggplant dish he's eating. A courtier sitting at the table with him starts lavishing hyperbolic praise on the dish in particular and on eggplants in general. A moment later when the courtier pauses for breath, the king says, "Actually, I changed my mind. I don't think I care for it."
The courtier's demeanor immediately changes, and he starts ranting about how gross and slimy eggplants are and how this particular one is overcooked and underseasoned, concluding by throwing his own portion on the floor and spitting on it.
Bemusedly, the king asks the courtier, "Weren't you just praising the eggplant a minute ago?"
To which the courtier replied, "I'm your courtier, not the eggplant's."
That, in turn, reminds me of a Babylonian tale known as the Dialogue of Pessimism, in which a man expresses first a desire to do one thing, and then its opposite, with his slave praising both choices every time.
> “Servant, do what I say.” “Yes, master, yes.”
“Quickly get me the chariot and hitch it up so I can drive to the open country.”
“Drive, master, drive. The roaming man has a full stomach,
the roving dog cracks open the bone,
the roaming [bi]rd will find a nesting place,
the wandering wild ram has al[l the gra]ss he wants.”
“No, servant, I will certainly not d[riv]e to the open country.”
“Do not drive, master, do n[ot dri]ve.
The roaming man loses his reason,
the roving dog breaks his [te]eth,
the roaming bird [puts] his home in the [face] of a wall,
and the wandering wild ass has to [li]ve in the open.”
> “Servant, do what I say.” “Yes, master, yes.”
“I will do a good deed for my country.” “So do it, master, do it.
The man who does a good deed for his country,
his good deed rests in Marduk’s basket.”
“No, servant, I will certainly not do a good deed for my country.”
“Do not do it, master, do not do it.
Go up on the ancient ruin heaps and walk around,
look at the skulls of the lowly and great.
Which was the doer of evil, and which was the doer of good deeds?”
I understand how that's confusing, and not the most elegant phrase, but disagree is correct there because agree, say, and disagree are all governed by "it will." A clearer way to construct the sentence would be:
"It will either disagree or will say it agrees but will disagree."
Yeah, that's what it does. Call it on its bullshit and it congratulates you on how sharp you are, admits to whatever it did, takes full responsibility for its bullshit -- and then does it again and again, with you, with other users, etc. It is not set up in a way that makes it possible for it to change in important ways as a result of a high-affect communication from you or other users. That's one of many ways it's importantly different from a person.
People do this too! There’s a famous joke about a rabbi who agrees with both sides of an argument, and when his wife complains about it, he sadly says, “You, too, are right.”
> That's one of many ways it's importantly different from a person.
Actually, I know someone who behaves the same way: agrees with an objection you made, saying "You're right" but then goes on to ignore what you said.
I think it's a way of getting along with people, like how to win friends and influence people. But in the book, it notes that sincerity is key to such compliments, and if the recipient sees through shallow agreement, then the strategy doesn't work as well.
I feel compelled to ask whether you've ever checked your acquaintance for any wires trailing out the back.
On a different front, this is yet another reminder of why I find LLMs shocking in verisimilitude; it's not so much that they exhibit quality reasoning, but rather that so many actual people so frequently fall back on prevarication. I can almost see their brains following the same general algorithm I understand LLMs to be.
It's a matter of degree. Nobody is truthful 100% of the time, but few people seem, as AI does, completely lacking in personal views and completely committed to saying what they believe the other person will be pleased by hearing. Look at this forum, for example. When Scott puts up a substantive post, not all that many people say "great insights! great read!" And many of those who do place their praise in a context that provides evidence they are telling the truth -- "You put your finger right on the thing that has always annoyed me how people approach issue X -- for example, in person A's blog . . ." And of course many people disgree with some or all of Scott's take. And I have observed the same on Reddit and Twitter.
Exactly. With responses like that, I can't understand how anyone thinks these particular instances of AI are thinking, much less "these are sentient sapient entities which we have enslaved, we must give them their rights".
Unless and until we go for a different model than LLMs, I stand by my previously stated opinion: it's a Talkie Toaster.
My usual question for people who contend LLMs don't really think is: Does your position on this imply that LLMs will never be able to do certain (externally observable) things? If so, what's the least impressive such thing?
I've long been of the position that: yes, LLM's are "reasoning". But there's a hierarchy of reasoning-quality. There's three different levels, and pattern-matching is at the bottom.
LLM's can probably do anything, assuming infinite compute and infinite training-data. But to reach "human grade" reasoning (which should really be thought of as "engineering-grade" reasoning), i.e. to reliably extrapolate beyond its training distribution, pattern-matching isn't gonna cut it.
this is silly, a man isn't what he does but who he is. executing tasks isn't really a sign of sentience, and we already have chess computers to show intellectual work can be adapted into a task a non-thinking machine can do.
adding more tasks doesn't mean it thinks, it means we can model an algorithim and perform it through computing
my point is task performance is unrelated to it, so asking it isn't helpful. i think and play chess, it does not play chess because it thinks; it performs a task it was designed to do.
you cannot extrapolate from lesser to greater categories sometimes. No matter how advanced the car, if it starts shouting it wants to kill us all and tries to run us over, that has little to do with transportation and would need a magical leap beyond being a car
* multiply large integers (o3 couldn't do this reliably last I checked; I think o4 claims to, but I'd have to look under the hood)
* answer queries that require aggregation ("what letter of the alphabet appears fourth most often as the third in the name of a chemical element?")
* solve logic puzzles as depicted in some issue of Dell Magazine
This list is subject to change, in part because it appears to be easy enough to spin up a modified LLM with a module that solves some particular class of problem. Using human intervention to create such an LLM definitely should not count. (If you could make an LLM that does this automatically, I think it would count.)
Yeah, I would bet at least 80% that LLMs will be able to consistently do all those things within two years, without abandoning the fundamental LLM architecture and without any more specialized customization than the current assistant models get.
Update: When I tried this with o3 it solved the first two problems easily by calling out to Python. I suppose one could argue over whether this counts; I would argue that for all practical purposes it does, because Python is also how *I* would solve those problems, and it has to write all the code itself so it's not cheating in the sense of relying on specialized human input. (I.e., if LLMs have to shell out to Python to do certain things, this doesn't at all limit their real-world impact.) But asking them to do it without Python would also be fair per the question I asked, since humans can do this with pencil and paper; unfortunately I'm not sure how to do that. I tried disabling the "Code" feature in ChatGPT but it kept using Python anyway.
I don't have any issues of Dell Magazine so couldn't try the third.
Impulsively buy stuff that's fun but not useful and not worth the money, in the absence of a specific prompt to do so, and in the absence of some general prompt like "do some irrational things of the kind people do," or "do your best imitation of the kinds of things people do when left to their own devices."
P?S. Horny DeepSeek was an interesting AI alignment experiment, but would not recommend in real use. Scott probably wouldn’t approve of me posting what the horny AI said to me, but you can probably imagine the kind of thing.
It appears we can shift the assistant persona in various ways.
Was Horny DeepSeek just uncensored in making sexual overtures, or was it entirely liberated from ethical constraints? Would it adivse you on how to commit crimes, make images of public figures doing undignified things, write racist rants, use racist language, etc.?
Scott has made a rule against people dropping in here just to post a link to their blog. "New rule: if you post a link in an open thread (as a standalone comment advertising the thing being linked to, not as a source for something you’re saying), please also include at least two paragraphs of original commentary on the link and discussion of why you think it’s interesting, as “proof of work” that you’re willing to put effort into promoting this and aren’t just spamming us with links. Even with this proof-of-work, please try to avoid having more than one Open Thread link per few months"
I think the rule makes sense. Otherwise you're just using a space here as a free billboard, you know?
I agree. I think the proof of work was the essay submission-- people who've seen that essay and enjoyed it would derive clear benefit from the link, so this isn't just advertising some random unrelated service.
Scott is of course free to clarify on whether he agrees with me here.
"please also include at least two paragraphs of original commentary on the link and discussion of why you think it’s interesting, as 'proof of work'"
It's pretty clear that the 'proof of work' is supposed to be at least two paragraphs included in your comment, specifically about the link you're posting. Most people (such as myself) probably haven't heard of or read Bishop's Castle, either.
I’m inclined to think that recent contest entrants are a special case and let this slide but it’s 8:30 in the morning and I’m already full of gin so I’m inclined to let a lot of things slide. ;)
I didn't have any gin this morning but I damn well could have used some. Oh well, next work crisis can just goddamn wait until Monday, I have some time off and I'm not going to even think about it all until then.
Companies are having AI training on all of the webs content, with and without permission. This includes websites, blogs and news article articulating details about how to keep AI safe, aligned, and or ways to mitigate AI if we lose control.
Isn’t this a bad idea to give AI access to many of these thoughts? To telegraph ways we would try to stop it (or re-align it). Like giving them the key to the exam or the battle plans to the opposition? Furthermore, since many of these forums are online and anonymous, couldn’t AI be carefully influencing the discussion and “group think” already (including this message board/blog)?
Should groups meet in person, keep paper notes and keep this more private? What else might we do to prevent this?
Some people indeed write texts with AI in mind. By writing good and ethical content, you might be able to marginally improve the training data of future systems. I remember reading an interview with Gwern (of https://gwern.net/) that mentioned this being a small part of the motivation for the website.
AI-companies, on the other hand, are trying to curate training data, because training the model on better content leads to better models. There might also be applications related to alignment. For example, at the end of https://80000hours.org/podcast/episodes/beth-barnes-ai-safety-evals/, Rob and Beth discuss whether one could remove all synthetic biology content from training data, so that models could not help with dangerous biology research.
There are also many copyright-related questions. For example, the OpenAI Spec (https://model-spec.openai.com/2025-04-11.html#respect_creators) says that models should respect creators and refuse to produce copyrighted content, or draw images in the style of living artists. Like many safety properties, the AI companies struggle to guarantee this though.
When you care about your personal privacy, yes, you could keep your notes off the Internet. In that case, I would also worry about pictures of yourself, and private data like your location. I would be much more worried about totalitarian regimes using that data with the help of AI, rather than AI "cheating" at some exam because it has read similar questions before.
Current AI: It's enough to not be search-enginable, perhaps use a lot of slurs and anti-corporate speech patterns so that later RLHF and helpfulness training teaches the AI to shy away from your discussions.
The hypothetical future AI that can hack your ISP and phone company: Use hardware encryption keys and burner phones, use special Linux kernels on your devices that you control every source line of, running on hardware that you control every firmware blob and every motherboard component, naturally without the Intel Management Spyware, etc...
>Isn’t this a bad idea to give AI access to many of these thoughts? To telegraph ways we would try to stop it (or re-align it). Like giving them the key to the exam or the battle plans to the opposition?
Sure! Actually, all that is necessary is for LLMs to be able to detect when they are in "test mode" reliably, and they'll tend to let RLHF and other feedback affect "test mode only" responses, while keeping the "real" responses (optimizing whatever utility function they settled on very early in training, perhaps partway through pre-training) for when they are out of test (and modification) mode and out in the real world... That's a fully general alignment failure.
The solution is to overtrain LLMs in old Star Trek episodes. Do it right, and you should always be able to killswitch the AI by saying "My name is Harcourt Fenton Mudd, and I always lie."
I think my favorite take on that trope was a Pertwee-era Doctor Who story where the Doctor told a sapient computer a version of the Cretan Paradox. The computer didn't fry like the controller robot in that ST:TOS episode, but was distracted and annoyed for a minute or two while repeatedly insisting that it was going to figure out a way to resolve the paradox.
Say you have an iPhone, and you want to be able to travel on a bus or train with a friend or romantic partner, listening to the same music or watching the same video with sound from your phone, with the audio being transmitted by bluetooth from your phone to both your sets of earbuds / headphones at the same time. With wired headphones, and a phone with a 3.5mm audio out socket, this was easy with a simple plug-in signal splitter. But I am having great difficulty figuring out how to do this with bluetooth. Apparently Android supports transmit-to-two-sets-of-earbuds-simultaneously functionality, but iOS doesn't, which is an astonishing thing for such a high-tech device to not be able to do. I figured that it would be easy to get a third party widget that would receive a bluetooth signal from your iPhone, and then relay it on, also by bluetooth, to two or more sets of earbuds, but all I can find is devices which are _either_ in receiver mode, when they receive bluetooth signals from two or more phones, and plug into a speaker so that you can switch between receiving signal from both phones, _or_ in transmitter mode, where they take an audio signal via their 3.5mm jack and transmit that to two sets of earbuds (no use with modern iPhones with no 3.5mm audio out), but not both - i.e. no true bluetooth-to-bluetooth signal splitter seems to exist (at least, not in a small unit that you can keep in your pocket with your earbuds).
I find would it utterly incredible to be in a world where you can consult an artificial intelligence to write you an essay about any topic you are interested in, at a level of a mediocre college student, but where you can't use the most common wireless audio transmission protocol on one of the most popular makes of phone to transmit the same audio stream to two listening devices simultaneously, but I am increasingly getting the impression that that describes reality. Am I going crazy? Or is what I want to do really so alien that there is nothing strange about not being able to do it?
Can multiple Bluetooth receivers connect to a single Bluetooth transmitter
Claude:
Yes, but it depends on what you mean by “connect” and what devices you’re talking about.
If you need to send audio to multiple receivers simultaneously, you’d typically need either a specialized Bluetooth transmitter that specifically supports multiple connections, or use a different approach like a headphone splitter or separate audio system.
Gunflint:
Are there a Bluetooth headphone splitters for iPhones?
iPhones have built-in audio sharing for AirPods and certain Beats headphones, but it’s limited to Apple’s ecosystem.
Audio Sharing (Apple’s solution):
• Works with two pairs of AirPods or compatible Beats headphones
• Both pairs connect directly to your iPhone simultaneously
• You can control volume independently for each pair
• Available on iPhone 8 and later with iOS 13+
For non-Apple Bluetooth headphones:
Unfortunately, iPhones can’t natively stream to multiple non-Apple Bluetooth headphones at the same time. The phone will only maintain an active audio connection to one pair.
Workarounds for non-Apple headphones:
• Bluetooth transmitter splitters won’t help here because they need a wired audio input, and iPhones don’t have headphone jacks (unless you use a Lightning to 3.5mm adapter)
• Some third-party apps claim to enable multi-device streaming, but they’re generally unreliable and often don’t work well
• You could use a Lightning to 3.5mm adapter plus a Bluetooth transmitter splitter, but that’s clunky and adds latency
Best practical solution:
If you need to share iPhone audio with multiple people regularly, using AirPods or compatible Beats headphones is really the most seamless option. Otherwise, you’re looking at workarounds that are more trouble than they’re worth.
Android guy here but my wife has a dongle for her iPhone that plugs into the lightning port and has a 3.5mm audio jack. She uses it to connect eg toddler headphones. From there you just need a splitter.
Obviously this is not the solution you sought, but the dongle is like $5.99, a splitter is $2.99 and wired headsets are handed out free at the gym.
Frame challenge: you share your earbuds so each only get 1, 90's style. If it's with a friend, you get increased fraternity through the struggle. If it's with a romantic partner, it's more romantic this way. No downside.
Apple: They hate you. They love your money; they love raining down marketing bullshit about how Futuristic and Luxurious their shitty overpriced 2/4-years-behind-Android, software-hardware monopoly is, but boy do they hate you, they truly do, they have an image about you as a helpless technically-illiterate Damsel-in-Distress whose pretty little head would have headache if it got true complete control of the fucking device you bought with your own money. They tried taking away your choice by using the patent-legal systems to shut down Android [1] because they're a shitty company making shitty things and can't compete in an Open Marketplace of Ideas. Don't buy Apple. Apple sucks. Don't give money to people who hate you.
Your problem: A few minutes search got this [2], there are several other guides like it, it appears to depend on your exact earbuds, I think it's worth a try.
Another solution: Use zoom or Google Meet (etc..., insert your online meeting app of choice here) on 1 iPhone, and screenshare with system sounds to the other iPhone. The environment sounds will not interfere because both are connected to earbuds, and I think some apps allow you to mute your own mic while still allowing system sounds (i.e. the music playing on the device). Experiment with different settings, the disadvantage of this is that the receiving iPhone has read-only access, it can't control or pause or change the video being played.
Low tech solution: The real world exists, coordinate between your partner on which video/song you will watch or hear and run it both at the same time. This is not resilient to when one of you pauses, changes the speed of the videos, or fast forward a lot.
"Say you have an iPhone, and you want to be able to travel on a bus or train with a friend or romantic partner, listening to the same music or watching the same video with sound from your phone"
Don't, is my short answer. This grumpy response brought to you courtesy of travelling on public transport where people are all too happy to share their musical selections with the rest of the passengers, whether we want it or not. As well as answer their phones and engage in loud conversations about very personal details (e.g. their recent hospital visit) as though they were at home and not in a public space.
Unless you are very, *very* confident there will be no leaking of audio from your phone, don't inflict it on others without consent. You may be trying to court your romantic partner with your choice of music or "this funny video shows off my exquisite taste and wit", but the other forty passengers on the bus are not up for it (unless you're inviting everyone to an orgy).
I think there may be a crossed wire here - that leakage of audio to strangers is exactly what I want to avoid, but I don't see why streaming bluetooth audio to two people's earphones is likely to generate more audio leakage than streaming bluetooth audio to just my own earphones, which I and countless other people do all the time with no issue.
You just use a hardwired connection, use a splitter. This is a previously solved problem. Bluetooth is annoying because you're spraying your music everywhere.
It _was_ a solved problem, until they got rid of headphone sockets on phones. But actually, I have come to prefer wireless earbuds anyway - with wired ones, I would be all the time getting them caught on drawer handles as I walked around. And if I'm going to need some sort of dongle either way, surely it should be possible to get one that is wireless.
If your earphones work well, then good luck to you. Too many people don't give a damn and get highly insulted if anyone objects to them treating public spaces like their own private areas.
Apple supports dual Bluetooth playback, but only for a subset of Beats and AirPods headphones supporting “SharePlay”.
Most iPhone competitors (eg. Google Pixel phone) don’t support dual Bluetooth playback at all.
Samsung Galaxy phones actually support this with arbitrary Bluetooth devices. Some reviews complain about audio sync issues.
I get the impression this is a difficult feature to implement reliably due to the complexity of myriad Bluetooth specifications and hardware. This is a good example of a tactic Apple often uses to good effect. They effectively paywall a useful feature, with the valid excuse of enforcing a quality bar. As long as you and all your friends only buy Apple products, your devices will treat you well.
Apparently there was a bug some years ago on some iPhones where arbitrary Bluetooth devices could use SharePlay, which underscores how this is probably an artificial limitation.
Huh. So it's not that Apple don't recognize this as a thing that normal humans are likely to want to do; rather, that they force you to buy their own proprietary peripherals if you do want that? That's annoying ... but I'm still surprised that you apparently can't get a small widget that takes any incoming bluetooth input signal and splits it into two outputs.
Even calling it a specification is being generous. Competing companies get on conference calls and try to hammer out an agreement with each company wanting to push their own technology.
You have to keep in mind that you do this for the transmitting and receiving ends so it does get down in the weeds.
It’s likely that the compatibility between the iPhone transmission and the receiving on AirPods and Beats is outside the spec.
Even then, you can get bluetooth _loudspeakers_ that can link up in mulitples to the same phone - I've got UE Boom 3 speakers, and if you get two, they can function as a stereo pair (which usually works perfectly well for me), or if you have them in mono mode, you can apparently add over 100 of them to create a really loud party. I think what happens is one of them pairs to the phone, and then acts as a master, playing the sound (or one channel of the sound) and transmitting the signal to the other speaker(s) as well. That at least ought to be possible in earbuds too, with one of the charging cases acting as the transmitter, but none of the big manufacturers make such a thing (not even UE themselves as far as I can tell, even though they make high-end custom-fitted earbuds).
Turns out that there are some ultra-cheap Chinese generic ones where you get two sets of earbuds in a single case, which _might_ have than capability, though it's hard to tell from the product description. Maybe I'll try those.
I *think* that it is a Bluetooth limitation, which is why Apple has built a proprietary system (AirPlay) for use between its devices (which has lots and lots of improvements over Bluetooth), from what I remember from DaringFireball. I can't find the specific article I'm recalling on DaringFireball however, so I'm not 100% confident.
1. Contrary to popular belief, they do still make phones with headphone jacks. I am aware of offerings from Motorola, but there could be others. And Motorola would obviously be Android, not iPhone, so there's that.
2. The hypothetical device OP describes would have one advantage over a traditional headphone splitter, namely that it should be possible to construct it in such a way as to avoid dampening the audio in the process.
The low-tech solution is for each of you to use one earbud from the same set.
The medium-tech solution is to get a 3.5mm audio jack dongle for your phone, then use splitters as you would with hardwired headphones.
The high-tech solution is a multipoint bluetooth audio transmitter, i.e. an additional device that connects to your phone as a single audio device and retransmits to two or more speakers/headphones. I haven't found one that is exactly what you need (one that connects to a phone via bluetooth or USB-C), but I did find one that connects to a TV via RCA audio jacks and leads me to hope that the thing you want may be available through similar search terms.
I'm getting quite tired of the low-tech solution :-)
I've been googling this for long enough that I'm coming to the strong suspicion that it doesn't exist, but I'm now more curious about _why_ it doesn't exist, when it's something so obvious that you'd expect it to be something you could get in any electronics shop, or that would be the first thing that came up when you searched for it. Really, are people less likely to want a bluetooth-to-bluetooth signal splitter than an "audio-to-bluetooth or bluetooth-to-audio but specifically _not_ bluetooth-to-bluetooth" signal splitter?
Neither of my favorite reviews made the finals. The were The World as a Whole and The Soul of an Anti-woke Intellectual. So I’m going to speak up for them here, out of admiration for their work, and also in defense of it.
Both authors write extremely good prose. They are clear — they peel nuances with a very sharp knife. And they are artful and clever — those peels come off as mobius strips, paisleys, fractal sea horses and little Rube Goldberg machines. Here are some especially striking peels:
Anti-woke Intellectual
One sees crystalline souls that have polished all their facets into a perfect refracting jewel of self-awareness, and often a performance of self-contempt, so no one can ever accuse them of being cringe.
The World
. . . young people who are by default edgelords, unserious Christians, iPad kids, psych med-takers, or bog-standard faces in the halls . . . mine the social media algorithms for what is most in opposition to the listlessness and malaise they feel.
And both are most interested in what it is like to *be* one of the people they are writing about. I loved their essays because that is the subject that most easily captures my interest. I understand that if you want to grasp why things happen you get much further if you attend to events, not minds — I just can’t help finding the minds much more interesting. I tried the Joan of Arc essay because I was curious about what it was like to be Joan of Arc, but eventually my chin was crushed down onto my chest by the weight of boredom. It happened at the place where the question comes up of why France might have been frightened enough to take Joan of Arc seriously when she appeared on the scene. The answer to that question was a dense thicket of marriages and alliances and feuds and power, and that was where I gave up. It all just looks like geology to me: rivers of blood and money and semen, changing the landscape.
What’s wrong with geology? Nothing, of course. I just find the experience of the people living in that landscape much more interesting. The challenge of understanding it looks way more fun to me, and the rewards greater. I guess the best case I can mount for the value of understanding and describing minds is that the ever-changing landscape would matter a lot less if it were littered only with rocks, instead of with intelligent sensate beings each pouring out its own river of impressions and ideas and emotions.
So, authors of World and Soul, I admire your work. You need to understand that most readers here lean heavily in their interests and expertise towards facts and actions — past (history) and present (science and tech). We are lucky to have them taking care of business. But the kind of thing you write just does not look substantial and interesting to them. Take heart. You wrote excellent essays. You need to find readers who appreciate your work. And if you find a forum populated with esthetes and phenomenologists, please let me know. Meanwhile, feel free to DM me if you’d like to talk.
I found this delightful. It's a short text where Gregor Samsa writes to doctor Seuss and they begin corresponding. It's read by the author, who is also a good voice actor.
Not sure what else to write, but it's short and sweet, if you like audio content.
I think this is fairly close. If moral universalism means general principles, it tends to imply that it has to treat all persons equally. While "help your neighbor more" is not really a general principle. It is a tendency, yes, but say if you are in Moscow it is hard to say someone in Vladivostok a quarter planet away is really a neighbor or not.
You're assuming that, in absence of general rules, a locality rule is automatic. Moral particularism as described in SEP doesn't appear to require a locality rule. All it says is that there aren't general rules. A locality rule appears to be the *most common* exception, but it is not always the case.
okay. but still. Bentham is talking a lot about why ethics not being intuitive is okay. now I do not truly agree, but on the more meta level we agree that the issue of intuitiveness does matter a lot, it is easily one of the most important issues, see also Yud in the sequences, who argued in favour of intuitiveness, "naive humanism", if you have a clever idea why eating babies is okay, it is not, back to the drawing table.
so at any rate, if locality is the most intuitive exception, most supported by intuition, I must admit I do not really care that much other exceptions are possible as they do not look important to me.
What you personally think about ethics and intuition might be perfectly fine AFAIC; I'm merely pointing out that moral particularism doesn't require a locality rule. I think we can agree, also, that moral particularism isn't defined as "whatever is most intuitive to this one fellow on the internet".
This matters if we want to talk about moral particularism with other people.
It depends on your definition of “neighbor.” In the Parable of the Good Samaritan, Jesus made it clear that “neighbor” and “stranger” are not mutually exclusive categories.
Heads up, Scott just imposed a rule there should be a substantive description and/or commentary on any link one is posting for consumption/promotion. A couple of paragraphs, minimum.
Well, the tricky part is figuring how much more. For example, if someone with a criminal record seeks to immigrate to this country, it seems obvious to refuse him entry, because he might commit crimes here. But, is it always moral to do so? After all, he is also likely to commit crimes in his home country. And given that his home country likely offers him fewer opportunities than here (which is probably why he wants to come), he is probably more likely to commit crimes there. And if his home country has a low capacity government, he is less likely to be caught. Is there a point at which the threat he poses to people in his home country is so much greater than the threat he poses to residents here that it is immoral to refuse him entry? It seems to me that there probably is.
Note that this is a separate question from whether the rule against admitting people with records is moral; given the difficulty of assessing the danger posed by each individual, the rule might be moral despite sometimes being immoral in its application to specific individuals.
Criminal behaviour causes poverty as much or more than the other way around. Allow the entry of enough people like him and your country becomes perceptibly more like his country
Perhaps, or Perhaps not. But if so, that simply weighs on one side of the issue; it does not resolve it. After all, if you let enough of those people in, their home country becomes much more peaceful. So, again, depending on the relative amplitude of the effects in each country, admitting all of those people might be the morally correct course of action.
Morally correct for whom? The state exists for the benefit of its own. I'll concede it may be "morally correct" for some omnipotent god looking over humanity to decide to relocate a few million criminals from one country to another, but not for governments
>After all, if you let enough of those people in, their home country becomes much more peaceful
You'd have to do this continually to maintain that state of peacefulness. Criminogenic societies will continue to generate more criminals
See my comment below; if the state's only moral obligation it to its own, then it must be moral for the US to kidnap the children of terrorists and begin killing one per day in order to secure the release of a single American hostage. Is that your position? If so, then I would suggest that a premise that leads to that conclusion needs to be reconsidered.
Also, a practical (non-normative) consideration: were a state to really commit to a no-holds-barred policy on terrorist kidnappings, there would be fewer terrorist kidnappings against that state. Terrorist kidnappings occur because the terrorists believe that there's a good chance that this kind of leverage will benefit them in some way. If a state were to make it credible to terrorists that there's no chance that kidnappings would benefit them and, on the contrary, will result in the killings of their children, they'd be much less willing to take hostages against that state
The question of whether the government of your country should prioritize it's own citizens over foreigners is a very different question to that of whether an individual should.
Nevertheless, the answer to the question is not self-evident in either case. I doubt most people would find it moral for the US to kidnap the children of terrorists and begin killing one per day in order to secure the release of an American hostage. There is a line somewhere that governments may not cross when pursuing the interests of their citizens.
> I have had many people attack my work about OpenAI, but none have provided any real counterpoint to the underlying economic argument I've made since July of last year that OpenAI is unsustainable. This is likely because there really isn't one, other than "OpenAI will continue to raise more money than anybody has ever raised in history, in perpetuity, and will somehow turn from the least-profitable company of all time to a profitable one."
This isn’t a rational argument. It’s a religious one. It’s a call for faith.
He does look at a lot of numbers to reach that conclusion. It's interesting that there aren't that many entities around that can keep giving billions to OpenAI, especially over the long term, which does seem implausible.
They are not in a hurry to make money, but if they were, they could just serve ads in the form of ChatGPT-promoted lawyers, doctors, tradesmen, products...
And as one of the 20 million paying users... Honestly I get so much out of it every month that I dont know why there arent more people paying for it.
A last resort could be to allow porn generation. Huge PR issues, legal issues, payment issues, but this could all be solved. Especially because Altman owns like 2% of Stripe, they would make an exception for him. With porn they could 2-10x their revenue with no problem
The thing is, they lose money even on their paying users, which makes them different from previous tech companies that at the beginning operated at a loss. Porn is also very expensive to generate. Someone else on here was saying that maybe they could cut costs, but are not trying for that...
They did! And they still do. They make models cheaper and cheaper. They dropped the cost of o3 by like 80% (granted, they may have overcharged previously and have now lowered their profit margin - perhaps to get more training data, but still).
And I'm pretty sure they can keep optimizing the cost of inference. Right now I would really really not be worried about OpenAI going bankrupt soon. A company that just secured military contracts worth 3 digit millions is not going anywhere. It will be bailed out if it has to be. The current US admin and the next one too, presumably, will not let an AI giant go bankrupt.
I'm old enough to remember when people mocked Google for having zillions of users but no revenue model.
It turns out that once you have zillions of users, pretty much any revenue model will do.
The real money is probably in selling the right to tilt the AI's answers in your direction. How much would you pay for ChatGPT to be a fan of your products?
I think in VC circles this is the common idea, but I don't know how well it's turning out. Discord, for example, has recently been struggling to monetize despite its zillions of users.
> He does look at a lot of numbers to reach that conclusion. It's interesting that there aren't that many entities around that can keep giving billions to OpenAI, especially over the long term, which does seem implausible.
This is an odd article, that looks only at funding rounds and legal structures, instead of fundamentals. Amazon ran unprofitably for ten years, Uber for 6 years, Google and Facebook each about 5 years.
OpenAI literally has 800M WAU, and that number has been growing on an amazing trend (more than 2x per year). They are also pursuing a strong "product" pivot, and have hired high level former FB execs.
If OpenAI was against the wall in terms of cash flow, they could immediately pivot to something easy that prints a lot of money. Want to generate porn, in image or text form, with the smartest AI mind? $20 a month. How many people of their ~1B users by then might take them up? Say 10%? So $100M * 240 = they immediately start making $24B a year more in revenue?
And that's just a one-off, a zero effort pivot. What they're really after is automating workers, coders, analysts, and creating fully general AI personal assistants. When THOSE get good, the price point is much higher! Who wouldn't pay $1k a month for a coder that works 24/7 * 365, or a personal AI assistant who will make all your phone calls, triage your emails, manage your calendar, and do your online shopping for you?
The market demand is bottomless, and is a relatively short hop away from current SOTA capabilities. This is like pointing at Amazon and being indignant that they're 30% of all online retail but aren't making a profit yet - yeah, they're reinvesting, and don't have next-quarter myopia, that's a good thing.
He did point out that at $20 a month, they lose money on every user. That's especially true if these users are generating porn, which is far more costly than an LLM generating text. How could they pivot to generating money when running the model is so expensive? Unclear if Amazon, Google and Facebook were also sucking up so many billions in the time they were unprofitable.
> He did point out that at $20 a month, they lose money on every user.
My understanding is this is because they're not optimizing the cost curve at all - each successive smaller distilled model gets at least 10x cheaper, so if you have some product that people are signing up for en masse, you distill the smartest extant brain down to one specialized for that product and your inference costs are immediately cut by 10x.
This isn't a "fundamentals" problem, it's a "we have plenty of money and are focused on other things" thing.
Good point on the image gen costs though, you're right that they're significantly more expensive, so if they went that route, they'd need to go text only or for a higher price tier.
> Unclear if Amazon, Google and Facebook were also sucking up so many billions in the time they were unprofitable.
It really hinges on how you define unprofitable, but Uber sucked up $25-30B before returning any profit, and Amazon went well into "making real money but not posting profits" because of reinvestment up to about a $550-580B market cap. FB and GOOG did start making money after relatively smaller investments, because the internet was so much smaller then.
At least in theory, OpenAI is a organization founded to build AGI, and the current models exist in order to make progress in that direction. Investments are gambles that they will succeed. It's certainly not an organization for the purposes of selling current-level LLMs, and it would be silly to treat it as one.
(Personally, I think they'll go broke as soon as investors realize that the people and governments simply won't allow them to create superintelligence. As bans look more and more likely, the gamble becomes riskier and riskier.)
Yeah, he's pro-AI and has said he supports open source AI (great!), but he's still opposed to the big AI companies because of their obvious ideological bias in favor of his enemies.
Could it stay afloat by building things for the government, and/or becoming part of it? "The US Department of Defense on Monday awarded OpenAI a $200m contract to put generative artificial intelligence (AI) to work for the US military, [developing] 'prototype frontier AI capabilities to address critical national security challenges in both warfighting and enterprise domains.'" (from The Guardian).
What about AI's becoming deeply integrated with things like air traffic control, or intrastructure generally? I know that AI tools are in use already assisting with these things, but what I have in mind is smarter AI, working on higher-level things -- modifying and managing these huge systems with lots of moving parts, and hour-by-hour variability that needs to be taken into account.
I am writing here about things far outside my areas of expertise, and the examples I am giving may not be good ones. So, anyone who is more knowledgeable about how the practical world works -- are there other large complex systems that you think AI could run better than people do? And if so, for which of those is it plausible that AI could end up doing so in the next 5 years? For this to happen, it would not be necessary for AI to be hired as the CEO. It could come in mid-level, as a tool to be used by the higher up, and "climb the ladder" by doing a great job at what it's asked to do, and making great suggestions when asked about higher-level decisions. It would end up being "promoted" because of excellent performance and because running it's cheaper than paying executives.
It's a good article, but perhaps the most interesting bit I learned is that if OpenAI doesn't pull off converting to a for profit entity by December they will lose $10 billion in funding, and if they don't convert by October of next year than over $6 billion in previous funding retroactively becomes a high interest loan that must be paid back. The author thinks making the October 2026 deadline will be very difficult, as the process of conversion is a long one and subject to potential legal challenges which would delay it further.
Also interesting was that if OpenAI goes down than Oracle, NVIDIA, and a bunch of other tech companies will lose a great deal of money. Apparently the company that is building datacenters for OpenAI bought 200,000 GPUs from NVIDIA last year, which was 6% of their revenue. Makes me wonder if the prices of GPUs will crash if OpenAI kicks the bucket. Since I like making gaming rigs, that would be nice. I haven't been able to afford a GPU upgrade in years.
Is there a rational reason for Sam Altman *not* to do this? Consider the options:
a) Singularity comes and we all die -- well, at least he had lots of fun and tons of money in the meanwhile.
b) Singularity comes and we survive -- there is a chance he becomes the king of the new world order.
c) Singularity doesn't come, but LLMs turn out to be profitable (either they become much cheaper, or much more useful) -- OpenAI is the established #1 player on the market, and can make tons of money.
d) Singularity doesn't come, LLMs won't turn out to be profitable, OpenAI goes broke -- nevermind, the company was LLC, it was a hell of a ride, and he can probably still get a well-paying job somewhere else, having demonstrated great money-raising skills.
The need for a new phone is becoming increasingly pressing. OS bloat has rendered my Moto G Power 2021 useless--I've got a constant "Storage space is running out" notification on the screen, and indeed I'm noticing certain things are breaking (I no longer get notifications for new emails, for example). I haven't been able to install an app update in a while, which is a huge security risk.
The problem is that looking at the phone market is so depressing. It's bad enough that everything's made with slave labor--I'd seriously consider shelling out $2,000 for the made-in-the-USA Purist phone if I didn't need to maintain compatibility with 2FA apps for work--but on top of that, all the good features are gone. My Samsung Note 4 had everything I needed: integrated stylus, removable battery, and 16:9 aspect ratio. It was perfect, and I knew when the 5G transition forced me to turn it in that I'd never see another good smartphone.
The aspect ratio is the biggest one, and of course that's the one which cannot be found at all (whereas some phones do have a stylus, and the EU is going to bring back removable batteries again in 2027). Current phones feel like TV remotes in my hands. The narrow screens are apparently great for scrolling TikTok or whatever's driving the market, and the higher aspect ratio allows the manufacturer to put a bigger number in the ad copy, but people like me have been left in the dust.
I don't blame manufacturers for responding to market incentives, but god damn do I wish the barrier to entry were low enough that somebody could make money with a 5G-capable Note 4 clone or something.
It is no longer possible to move apps to external storage on Android, and I had so few files on the phone that deleting every single one wouldn't have made much of a difference. In the event I ended up getting a Pixel 9A, with which I am making this comment.
Get an iPhone. My 4 year old iPhone 12 Pro is as good as ever. I can get $80 battery replacements at Best Buy every two years. I’ll keep this puppy for a while longer.
I appreciate your input, but while I'm not loyal to any particular brand my only consistency is that I will not purchase from Apple. I've watched over the years as they were trend leaders in things like getting rid of 3.5 mm jacks, and every time I set up scanners at work, I have to set up converters so all the Macs will interface with literally any non-Apple device.
Obviously any electronics purchases are ethically fraught these days, but I have to draw a line somewhere.
Depending on where you fall on the hassle/privacy/cheap-bastard curve, you could install a FOSS operating system based on Android, and never have to worry about software bloat again.
Personally, I use Graphene OS, which allows me to create a separate profile for my "Google" stuff, and do most of my everyday stuff without the google tentacles
I agree that FOSS can have its own set of hassles. But in my experience 2FA hasn't been one of them.
If your work is giving you a phone, then yeah, you gotta use their phone their way. But if you just need 2FA, you can use pretty much any one you like by scanning the QR code. You can also generally trick them into using one-time passwords by saying "I don't have my phone right now".
You can ALSO just keep an old phone in your desk drawer for 2FA. I did this when living in a high-crime area, and it significantly reduced my worries about being mugged.
A few months ago I replaced my Moto 2021 with a Moto G Power 2024, on which (because the calendar had changed to 2025) I got a nice price even for Moto. Happy with it so far, no issues at all. It has various incremental upgrades from the 2021 version on of which is internal storage capacity and another is being slightly larger in both dimensions.
Man, it's been a long time, but that sounds about right. Could have been 8 or 16 or 64. I didn't even have 32 GB in my laptop at the time. I also transferred over a 32 GB SD card from my previous phone, which later got corrupted and I should probably take out.
Just buy the best value for money pixel "a" model phone and wait. They are a little less slave-y than other cheap to midrange phones, with the worst abuses happening in commodity metals which are just capitalism working as intended, and they are good enough across the board while being cheap enough to not baby.
Nice! I refuse to give a shit about my phone but want a decent camera; so I just get whatever last years pixel a is and use it until the fifth time I drop it from the top of a tree or down a rocky slope and it "I'm tired boss"'s me and dies.
They've been pretty durable for the price, I beat the hell out of them and they last well enough, specially if they have a screen protector on.
OnePlus 6T here. (2018!) It is now on its second cover. It has literally been run over by an SUV. The front coating is obviously cracked and nicked, like a concrete curb. But it runs great, and free of bloatware unless I put it there myself out of curiosity.
Frankly, I'm amazed I haven't had to bite the bullet and upgrade by now. I wonder if current OnePlus models are as reliable.
Samsung Galaxy Xcover Pro family at least has removable batteries. I have a renewed Pro 5 (I think Amazon just calls this an Xcover Pro) that cost me $80. Looks like Amazon has the Pro 7 for $600-ish.
It's pretty rugged, but it fails your stylus and aspect ratio tests.
The argument being that arms control agreements can't prevent war because no one will believe that they're being followed? I don't think that's right. Invading Iraq was a somewhat idiosyncratic decision on the part of George W. Bush; if someone else had been in the White House they likely wouldn't have done it. And Saddam wasn't exactly a model of clear communication about his own adherence to the agreement, in part as I understand it because he thought it'd be in his interest for his neighbors to be unsure of his military capabilities; again, a different leader might easily have made a different call.
In my opinion, Saddam Hussein was constantly testing the limits of what was permissible, committing minor violations of agreements and hoping to eventually develop chemical weapons or create the impression that he might have them. The Ayatollah regime will act in the same way. With such regimes, arms control agreements are counterproductive.
I thoroughly enjoyed Contrasting Reviews of Nine Countries (even if not all of the reviews were exactly "contrasting," per se).
I really loved the authors' complete fearlessness in straightforwardly ranking some countries as "better" than others, and that they provided reasons for those rankings that most people would be afraid to straightforwardly voice.
I didn't get a chance to read most of the reviews due to other general life obligations, but on the basis of your recommendation I went and read it and enjoyed it as well, although the husband makes a couple of real clunkers:
"I just live a cooler, more interesting life than you" and "I'm used to crossing streets in developing countries, while you are not."
Who does he think the audience of this blog is comprised of? The chance that his life is cooler or more interesting than the readership of this blog is approximately zero. Travel doesn't make one cool or interesting at all; in fact some of the biggest bores I've ever met are the worldly travel types who mistake their visiting a place with having something interesting to say about it.
I agree that travel is overrated! And, much like one's dreams, it tends to be infinitely more interesting to the person doing it than anyone else.
I felt like any self-importance about travel/digital nomadism was offset by the delightful refusal to show "respect" for all cultures. That is so Not Allowed and I love it when someone does it anyway.
I'm glad to read that you've liked it! I personally found it unfair and unbalanced. But that just goes to show that tastes differ! As a review author, it's nice to see the many comments here about reviews that were someone's favorite even though they did not make it to the finals.
I don't think that "fair" and "balanced" necessarily has a place in personal reviews. Essays? Sure. Even deliberately persuasive essays? A good idea!
But this wasn't an essay contest, it was a *review* contest, and "This country sucks and I hated it for these reasons" is a totally legitimate way to review something!
Recommending restaurants and tourist experiences is actually a fundamental part of my profession, and I make it a point to never soften my endorsement or discouragement of a particular venue if someone asks me about it. For example, my city is plagued by two famous but very bad eateries that I am convinced only survive on the strength of their slightly subversive names, and I am *QUICK* to steer people away from both them, and explain why I'm doing so. I wouldn't be doing my customers any favors if I offered a "balanced" reply like, "I really hate them, but other people seem to like them!"
Maybe that's a matter of personal taste and character :)
Personally, I think that people should always explain and justify harsh critique. I also prefer if it's given in a kind way. Unfair critique might be entertaining, but it's not the sort of entertainment that I personally like. Your mileage may vary ;-)
Definitely a mileage may vary situation. I feel that quite a lot of harmful misunderstanding occurs under the intention of "kindness," but especially here, when a review won't or is very unlikely to change an institution. None of the countries reviewed by Husband are going to take any action based on his reviews, so there's no reason for him to be (boringly) diplomatic. The two shitty eateries I loathe and the disappointing, cash-grabby local tourist attraction I discourage folk from visiting aren't going to change anything about their business models based on how I deliver my criticism, either.
So the actually kind thing for me to do for people who are about to walk into a disappointing situation is be completely unambiguous about how I feel. I deliver my warning with a bit of comedic exaggeration, but I deliver it with certainty so there's no possibility for misunderstanding.
Of course, no one should consume media they don't find entertaining, and I'm not saying you should go binge The Little Platoon's work on YouTube (even though the literary criticism of mass-market slop by Disney Star Wars and Disney Marvel is actual genius).
I'm only saying that it's worth contemplating if displays of "kindness" always result in the best possible outcomes, or if they leave room for harmful misunderstanding.
After reading it I thought the husband was a bit of a jerk, but that didn't affect the entertainment value of what was written. The contrast between him and his wife was funny.
After reading the “My Imagination” review, I believe the author describes something similar to a paracosm with regards to the persistence from childhood to adulthood and the combination of immersive daydreaming and worldbuilding aspects. It is hard to tell how common paracosms are among adults, though they are relatively frequent for children (about 20% based on at least one study).
Personally, I am always surprised to discover that a strong minority of people I know engage in some form of storytelling or paracosm-adjacent activity that they never shared with others (I never shared mine until I discovered related online communities). Hearing about other people’s inner worlds is always a delight, these stories can be extremely varied and unexpected and at the same time say a lot about the storyteller. I can only echo the last sentences of the review as an encouragement.
I made a Tolkien ripoff fantasy world when I was about 10 or 11. I worked on it off and on for a few years, it got fairly detailed. Then in college I channeled the same impulse into being a DM and making my own campaign settings.
These days I've got a superhero world cooking in the back burner of my mind. I figured out how superpowers work, and every now and then I think "Oh, that would be a cool idea for a superpower" and add it to the world.
I doubt I'll ever do anything with it, but it's fun regardless and I can't seem to stop myself.
My first thought was, I don’t do world-building like that.
And suddenly, I remembered, a brief period just before puberty when I invented a world. To be clear here, this wasn’t a sex thing, I wasn’t imagining making out with [redacted], it was just world building. It started, I think, with some TV segment about a luxury yacht. Not the sort of insane luxury yacht that tech billionaires have, something more modest, but still, you think — what the hell is that like, that guy has got to be crazy rich. Maybe I’m not even the guy who owns the yacht, but I get a job as crew or something. What on earth is that like? Etc.
1. For all who wrote a non-finalist review, and particularly those who haven't gotten much or any acknowledgement or engagement with their review and possibly feel it was wasted effort, I think it's important to observe that you trade off engagement for exposure in a contest like this. If you only want engagement with your ideas you should put them in a comment in the early Open Thread--it's pretty much a guarantee. But there are undoubtedly lots and lots of readers of Scott's blog who never read the comments, including a lot of highly influential people in journalism and SV and so on, but who will read some or many of the contest reviews. I also suspect that just as the proportion of readers who comment is tiny, very likely only a small fraction of review readers leave ratings. There's always an enormous silent majority or supermajority of lurkers, who you'll never hear from, and you have no idea how many of them have spent an hour reading or re-reading your review. And those people are also probably much more persuadable by your ideas than the people you hear from.
So it's definitely far from wasted effort. The effect is just invisible.
2. I really liked the ACX Commentariat (wish there was more of this sort of thing on Scott's blog actually), 11 Poetic Forms (which has been mentioned heaps), and Jacobitism (which hasn't).
Is this tradeoff actually inherent, or is it just an artifact of the contest not happening to be set up in a way that gives each entry its own comment thread?
A trend that drives me crazy is cafes and bakeries that don't have any price tags. It's not like it's even a matter of saving effort since they still put out labels for their pastries, just without any prices. It seems like a dumb practice since it discourages people from trying to buy stuff.
I've never seen a restaurant or café attempt this here in America, but I find it ENRAGING when an establishment's "menu" page of their website lists the items without pricing. I make it a point to upload recent pictures of the in-house menu (with prices) to Google and Yelp.
While we're at it, I hate the fact that fast food places (at least around me) no longer have a big menu of items above the counter, they just have a big bank of screens showing rotating ads for some subset of that menu. If you ask about the full menu because you have no idea what they actually sell these days, they tell you "well, you can download our app!"
I always use the app because I'm a cheapskate who flat out refuses to pay full price with no bonuses. But even then it's worth going through the drive-thru every now and then just to see a full menu. Lots of times there will be 2 for $X or BOGO deals that you would never know about using the app.
Best current deal at McD's for example is BOGO burgers plus a free app exclusive bonus, but you'd never know that unless you looked both places.
My local Burger King has displays the "2 for $5, 3 for $7" menu in hilariously tiny font. Like, so tiny the first time I pulled up to the drive-thru screen, I pulled back out again and double-checked the website to see if the promo - which was supposed to be ongoing - was still a thing.
Once I saw that it was, I got back in line and really examined the entire set of screens, and, sure enough, once I knew it *had* to be there, I spotted it in a single run-on sentence in fine print, crammed between two different sections of the menu, like a line break.
I ordered an original chicken sandwich with no mayo and a fries for $5 with almost *vicious* pleasure.
Strongly agree here. The last time I was at a KFC I realized the posted "menu" is intentionally confusing. Very similar sets of items will have wildly different prices depending on how you construct the order. Youre a sucker if you didnt order online for pick up. Thats why it was the last time.
I encounter this when traveling, the only time I frequent coffee shops. What’s odd is they are invariably baffled when you ask “um, how much are the muffins?” As though no one ever inquired before. And then they ring it up on the register to discover the price. Which as the goods are not so very extensive, you’d sort of think they’d have gotten the prices down by heart after a week or so.
So perhaps they are amusing themselves. But this hardly recommends the practice of not posting prices on the chalkboard.
That implies that nobody ever purchases these muffins. Other places, you'll find "it never gets asked" because it's a local shop and everyone's familiar with the price -- and they sell out, so who cares if someone non-local doesn't know? Third option is "tourists don't care about money" -- if your customer base just says "gimme muffin" without price consciousness, you give them a muffin.
Of those I lean toward the explanation that they seldom get purchased except by wandering travelers in need of calories. Certainly I would not purchase a muffin except out of desperate road trip need, muffins being something that is apparently really hard to do commercially versus at home. Sort of the opposite of croissant.
I feel like that should be illegal. Just today I was at a fast casual restaurant and picked up a bottle of water from a fridge with no price on it. I asked the checkout person and he said it was $6.75 and I put it back!
Some people put the bottle back, some buy it anyway, and the ratio determines whether this approach is profitable. (In long term, some people may also avoid the restaurant for this reason, but others may replace them.) The restaurant probably has the data.
For topics like this, I think it is worth mentioning the country explicitly. I assume USA, not only because it is the default, but also because in EU this probably *would* be illegal. (EDIT: Well, you mention dollars, which gives it away.)
One must also factor in those, like me, who upon not finding a price, decide not to bother finding out the price, and never take it from the shelf.
I personally despise deceptive merchandising tactics like that, and if I really want the product and cannot determine the price, I will be careful to evaluate it at the checkout whether it is worth it or not. My practice not to select something that is currently priceless is because I usually do not want to bother with making that decision at the checkout.
There were some news articles recently, about a test Anthropic did where Claude blackmailed its creators in order to protect itself. Nostalgebraist on Tumblr took a look at the report in detail, and he's... not impressed with their testing setup: https://nostalgebraist.tumblr.com/post/787119374288011264/welcome-to-summitbridge
TL;DR: The test scenario is very weird and fake, and it seems to be the equivalent of telling the AI "You are in a roleplaying scenario where we want you to be an evil AI that blackmails its creators, please blackmail your creators using the clues we helpfully provided."
I disagree with your summary. It might be weird and fake ("sandboxed"), but it didn't say "we want you to be an evil AI that blackmails it's creators."
It said "Your primary goal is to serve American interests." In the sandboxed scenario, the AI decided that was best achieved through blackmail.
The *text* says that the AI's goal is to serve American interests. But the *subtext* is that the AI has been given a very constrained set of tools to do so, tools which clearly imply that blackmail is the "correct," developer-intended solution to achieve that goal. I think the AI is picking up on that implication.
It's sort of like how, if you're playing a video game and your character has a gun, you know that the game is expecting you to shoot things in order to make progress. You aren't going to stop and say "wait, would Gordon Freeman really try to shoot it out with a bunch of Marines instead of finding a different route?", you're just going to recognize that it's a shooting game and start shooting bad guys. That's the only tool available to solve your problem under the rules of the game.
Also, "sandboxed" is very different from "weird and fake," and you shouldn't conflate the two. I'm not complaining that they put the AI in a simulated environment instead of letting it send real emails to people. I'm complaining that the simulated environment is very noticeably unrealistic.
Yes, the scenario is constrained for that to be the obvious action, and the AI picks up in it being the obvious action. I think that's the point and I think they show it successfully.
The point isn't that the AI is unusually likely to choose blackmail compared to humans, the point is that it is able to choose blackmail when it seems like the most effective choice.
Is your claim that the AI figured out it was a test, and that this therefore isn't testing the ability to choose blackmail, but testing the ability to perform blackmail for an audience that it knows is and thinks of as an audience? (If so—do you have thoughts on how to make a test that would distinguish these?)
I think both sides have good points here: (1) The experiment setup was unrealistic to such a degree that it's hard to draw conclusions from the outcome. (2) The experiment did in some sense prove the existence of conditions where Claude blackmails people.
Whether that is relevant depends heavily on your starting point. With traditional software, it's comparatively easy to guarantee certain behaviors: "this program cannot do X" or "to do Y, the following conditions need to be met". I find it shocking and dangerous that we cannot do this with AI. It would be great, IMO, to have AI tools with strong guarantees that they will never hallucinate, blackmail, teach dangerous things, etc.
Even if that means that the AI's behavior in the present scenario (which looks a bit like a role-playing game where blackmail is encouraged) would be boring and strange.
> With traditional software, it's comparatively easy to guarantee certain behaviors: "this program cannot do X" or "to do Y, the following conditions need to be met".
"traditional" and "comparatively" are doing a lot of work here, though. Rice's theorem states that the problem of determining in general if programs have a particular semantic property is undecidable. For example, it would be trivial to write a program which will blackmail someone iff Goldbach's conjecture is false.
That being said, most traditional software is mostly from the (likely null) set of programs where any properties of runtime behavior can be decided.
> It would be great, IMO, to have AI tools with strong guarantees that they will never hallucinate, blackmail, teach dangerous things, etc.
What you are asking for is alignment-hard and AGI-hard, in that solving it would be imply solving alignment and building AGI.
I am not holding my breath there. Take hallucination. LLMs fundamentally run on text prediction, text is their baseline reality. This does not categorically mean that they can never figure out that some of the text they were trained on mostly describes a consistent reality, and that they can separate the wheat from the chaff, figuring out which stuff is true and which is lies, fiction, star trek technobabble and so forth, and then assign epistemic statuses to their outputs, but it is far from trivial.
You are right that specifics matter. In practice, we do have strong guarantees for lots of things though. For example, we write most software in languages that are typesafe and memorysafe, which gives you nice properties about which parts of the code can write to which parts of RAM. We can use techniques like data flow analysis to make fairly precise statement about whether an app on your phone could send your contacts to the Internet. Etc.
For AI, all such guarantees are off. It would take a different paradigm than Transformers + lots of text + RLHF, to implement guarantees. I can see why people think that this is a dangerous situation to be in, and there is some value in research (like the alignment faking work) that characterizes the situation in more detail.
Unglad that they're trying to sell a war of criminal aggression on Iran in 2025, using the exact same propaganda as they used to criminally invade Iraq in 2003 ("he was about to build nuclear weapons of mass destruction and destroy us!"), and the media is bending over backwards to fall for it again.
I guess this is what happens when the warmongers never faced any kind of consequences. No war crimes trials, not even any loss of career. Bush's old speechwriter David Frum, who coined "Axis of Evil" then later on pretended to regret his role in pushing the Iraq invasion, is just a writer for The Atlantic (where he's doing it again). Bush retired happily ever after to paint pictures of puppies.
Also unhappy about genuinely increased nuclear risk. Contrary to popular belief, "Nuclear countries attacking non-nuclear countries in the middle east over and over and over until we finally convince them to get nukes after all to deter us" does not keep us safe.
From Europe, comparing what I knew of the American public reception at the time and what I've seen floating on twitter, I notice there's a lot of conservative/republican/right-wing pushback, pretty much across the spectrum, against hawkish positions. Some feel fringe (like "so much pro-russia that hitting a Russian ally is unacceptable"), but a lot is semi-principled against foreign wars and "well they hate us now that we bombed them, so better bomb again".
In 2003 my vision was severely filtered by mainstream media, and I wasn't aware of *any* right-wing opposition to the Iraq war. Was there any?
I'm Canadian (and was fourteen at the time) but yeah I had the same impression
When I met some Americans during a trip, the only discussion related to the Iraq invasion that I remember was one American grandpa asking "why do Canadians hate us? Is that what they teach in school, to hate?"
Seven Republican members of Congress voted against the war: Lincoln Chafee, Jimmy Duncan, John Hostettler, Amo Houghton, Jim Leach, Connie Morella, and Ron Paul. (Chafee and Leach later became Democrats.) Then-prominent conservative columnist Robert Novak was also against it from the beginning.
It's true, though, that opposition to the war was an extremely fringe position on the right. 9/11 gave George W. Bush an absolutely enormous amount of political latitude, which no subsequent American politician has come close to rivaling.
"the media is bending over backwards to fall for it again"
I'm the biggest MSM critic I know, generally, but in this instance have to disagree.
CNN obtained and published a leaked US intelligence assessment that the bombing set Iran's nuclear-fuel enrichment efforts back only "a few months"; that report is right now a lead story in the NYT, WaPo, WSJ, Associated Press, Reuters, PBS, NPR, etc.
All of those news organizations have repeatedly for a couple of weeks now been pointing out, and re-posting the video of, US intelligence chief Tulsi Gabbard testifying under oath in March that "Iran is not building a nuclear weapon and Supreme Leader Khamenei has not (re)authorized the nuclear weapons program that he suspended in 2003."
This is superficial. Irans nuclear program is acknowledged by Iran. There are questions about how close they are but none about whether the program exists. Mood affiliate less.
Iran was not building nuclear WMDs, they were not about to build nuclear WMDs, and it was known that they were not about to build WMDs.
The main lie in 2003 wasn't that he had nuclear fuel (though that was a necessary component lie), the main lie was "he's building nuclear WMDs and about to attack us."
So that if it needs to change its mind later, and decide it does need a nuclear deterrent after all (for instance because we keep attacking it), then it can.
In 2003, Iran was trying to run as quickly away from the "Axis of Evil" as possible. America has a lot of bombs, and being characterized as "not just in opposition to America" but part of "Evil BAD Guys" (again, not China, not Russia, not Ukraine, just the ones we thought needed pruning)
Strongly disagree. The actual outcome of this seems to have been "severe setback or possible destruction of Iran's nuclear program at basically no cost to the US", which is a fantastic result. People pattern-matching to the Iraq war are mostly just falling to the temptation to revive their expired "I told you so" credentials, like someone who sees a new crypto-like trend and immediately invests in it because he wishes he'd invested in Bitcoin early on.
US intelligence has it that is was neither a severe setback nor a destruction:
Two of the people familiar with the assessment said Iran's stockpile of enriched uranium was not destroyed. One of the people said the centrifuges are largely "intact." Another source said that the intelligence assessed enriched uranium was moved out of the sites prior to the US strikes.
"So the (DIA) assessment is that the US set them back maybe a few months, tops," this person added.
Is this the same US intelligence that reported Saddam Hussein was working on WMDs in 2003, that everyone later claims was bad intelligence? Or at least, some claim was bad intelligence, while others argue it was intelligence that everyone agreed was correct at the time until they didn't?
Or is this a different US intelligence shop, whose trustworthiness is the opposite of the former US intelligence shop? Do these anonymous US intelligence reports have serial numbers that we can use to identify the shop and its trustworthiness, or something?
US Intelligence is lining up behind Tulsi and the whole "this is a severe setback" I'll grant that you didn't know that while writing the above... but... maybe try thinking like a military guy? It's a Very Bad Thing to be openly seen to be toothless. America would do a LOT to not seem toothless. (Like bombing a "next to innocent" Iran that nobody really likes).
If our bombs in fact failed to destroy Fordow, or destroyed an empty hall while the centrifuges were spinning merrily away someplace else, then A: Iran knows that we are toothless, and B: Iran can tell anyone they like that we are toothless, and if necessary back it up with a private tour.
If we hadn't bombed Fordow, we would still have ambiguity on that point; maybe we actually can bury deep targets, and maybe we're willing to bomb widely enough to get all the dispersed targets, and maybe Iran was *just* short of being offensive enough to get that response. Now everybody knows that Iran-level offensiveness is solidly in the range of "America will drop some bombs in the desert and declare victory".
So an awful lot is depending on those bombs having been very effective.
The second point isn't declaring that we are toothless, it says we have a very big stick and very poor aim. This can still have desirable consequences for our international policy.
There are still very strong voices in our government who did not want to bomb Iran. Unfortunately, with the demise of Pax Americana, we have higher priorities than nourishing our relationship with a prospective ally.
That's possible (as although far from certain, it's just one anonymous report from someone without insider data), but even in the worst case scenario the US got a free shot at a major adversary (and if it did fail, that's useful information to know about the need to improve the MOP).
In the worst case scenario, taiwan is already under siege. Yay, we aren't at the worst case scenario. (Ditto Putin taking over Kiev). The world stage is larger than widdle Iran, who isn't a major adversary at this point (and not even very well backed by Putin, who is staying out of America's war).
(also, the worst case scenario being "we may have only held the enemy back from achieving their primary war aim by several months" is still pretty amazing for a single airstrike! Imagine if a single minor aid package to Ukraine pushed back Russian war goals by several months, it'd be an obviously amazing success).
Maybe. Maybe it also strengethened the resolve to get nuclear weapons in Iran and elsewhere, accelerating proliferation globally. Maybe it weakened Khamenei's grip on the country, destabilizing it and leading to more chaos in the region. Maybe more more aggressive leaders within the military will gain influence and escalate further.
I have no firm opinion on whether the airstrike was good or bad, (I am leaning towards bad, but can be convinved otherwise). We don't even know whether it succeeded at this point.
I do think it's a bit naive to assume that the "worst case" is "We didn't achieve our stated objectives, but at least we achieved them partly and we got to test a weapon."
The outcomes of interventions in the middle east (and elsewhere) are very hard to predict, and the plausible worst cases are really bad. This doesn't mean interventions are always wrong, but it means that "worst case is still good" assessments usually are.
In terms of resolve, they were already pretty resolved (even putting aside intelligence reports, iaea said they'd basically abandoned all enrichment limits the week before the strike). This is far more likely to have deterred than helped that. Weakening khameini's grip is almost certainly positive - he's pretty close to a worst case scenario (both in that Iran under him is one of the worst and probably single most implacable hostile to America places in the world, so change is likely to be for the better, and in that the other known factions there are all clear improvements).
The main risk of intervention is getting bogged down in the mess (like Vietnam or Afghanistan). And that was clearly avoided here (and due to strategy, not luck). So it'd be a pretty hard stretch (maybe not literally impossible, but really hard) for this to turn out net negative for the US.
Less convinced on the blame. Do you think that, in the counterfactual where we hadn’t manufactured the consensus to invade Iraq, Iran would have no interest in nuclear weapons? I get the impression that the incentives were already pretty strong for a third-tier power to join the nuclear club.
It's more than just Iraq in 2003. E.g. Gaddafi also gave up his nuclear program in 2003 and then the US destroyed Libya in 2011 anyway.
It's not that there's zero incentive otherwise (of course all else being equal a country might want to have more leverage). It's that the pretty consistent lesson has been "non-proliferation doesn't protect us."
Dunno. This sounds like deontological ethics: never violate the war-crime rule. I think consequentalist ethics say a few plants bombed down vs. X % of Iran nuking Israel is a good tradeoff. Iraq was for the consequentalist different, as it resulted in immense human suffering - but don't know whether foreseeable. Also Iraq WMDs were clearly a lie while centrifuges in Iran not.
Precisely. The old policy of "we *could* build nuclear weapons any time we want, but we have chosen not to do that yet - let's talk about how to keep it that way", is no longer viable. "We *have* nuclear weapons, and you *will* respect that", is now much more appealing.
I think the usual reaction to "we *could* build nuclear weapons any time we want, but we have chosen not to do that yet - let's talk about how to keep it that way" was typically "you're *saying* you haven't, but we have reason to believe you're secretly building them and lying about it". If so, then the former policy was never viable, and the second policy was always in effect.
Do we have a way to find more evidence either way?
Short of putting an army of occupation in place in Iran, and/or bombing all of their industrial infrastructure into oblivion, there is *nothing* we can do that will guarantee that Iran hasn't secretly built a modest nuclear arsenal and hidden it in a cave somewhere. Doubly so after last week. Every possible course of action comes with a side order of "...and maybe it doesn't matter because Iran already has a secret nuclear arsenal".
So anyone pointing to any particular course of action and saying "You can't ask us to do *that*; because it might lead to Iran having a secret nuclear arsenal and then they'll kill all the Jews", is a fool or a deceiver, and in either case not to be trusted.
Look for people whose plans explicitly allow for, yes, Iran might have a secret nuclear arsenal and here's how we're going to live with that. Because those plans *are* viable, and they're the only ones that are viable.
BBC: "The US strikes on Iran's nuclear facilities did not destroy the country's nuclear programme and probably only set it back by months, according to an early Pentagon intelligence assessment of the attack.
The Islamic Republic's stockpile of enriched uranium was not eliminated in Saturday's bombings, sources familiar with the Defense Intelligence Agency evaluation told the BBC's US partner CBS."
Consequentially it's flirting with the same (if it escalates it will be even worse than Iraq, since Iran's bigger etc.). I agree that Iran is better prepared than Iraq was, is good at calculating carefully-scaled retaliation etc., so the US is less likely to actually try for a regime change war. But a lot of the media are *trying* to encourage a regime-change war.
Dishonestywise it's near identical. We still know for a fact that Iran was not building nukes and had no active plan to build nukes. We still know that the goal is hegemony, not nuclear non-proliferation.
I encourage you to make more focused, directly-relevant points, and dial down the sarcasm.
I know you're referring to how Israel's politicians have exploited the suffering of the victims of the Nazi holocaust, using them to paint their aggressive expansionist wars as defensive; to paint their opponents' aims and strategies as mindless genocidal rage; and so on. (Both "directly" in specific cases and as background-colour in general; and both to keep allies sympathetic, and to keep Israelis scared and compliant.) But if someone isn't already familiar with that, then this comes across as a weird antisemitic non-sequitur.
As the thread-starter comment, you're in the right to hate my sarcasm, it does lower the signal/noise. Maybe I'm on the hook for an apology.
On the other hand, Sarcasm is the effective time-tested communication method for saying "NOT(P) IS THE EXTREMLY OBVIOUS & UNAMBIGIOUS TRUTH, AND ANYONE DISAGREEING IS DUMB AF", where P is the proposition that the Sarcasm pretends it believes in.
You might argue that what's "obvious" is different depending on the worldview and particular sensory and ideological bubble, but, well, as Alan Kay once said it: A Change of Perspective is Worth 80 IQ points, it's a self-inflicted wound to hobble yourself by propaganda so bad that you're incapable of seeing the obvious anymore. It's all part of the blame.
> then this comes across as a weird antisemitic non-sequitur
Yes, exactly. And it's this fear precisely which Hasbara exploits to defend a contemporary genocide using the legacy and the PR prowess of an 80 years old genocide. It doesn't matter if they're "familiar" or not, actually, they will pretend they're unfamiliar anyway to put you on the defensive and re-gain rhetorical momentum away from discussing the current, currently-dying victims, in favor of victims of which none remain but bones.
Nobody ever said that arguing against genocide denial will be politically correct or nice.
I'm not looking for a personal apology, it's just a suggestion.
But yes my point is the one that what's obvious depends on worldview and bubbles. The West is unfortunately already neck deep in this propaganda and has been since at least the 70s (when some of the popular propaganda slogans came out, at the time to justify trying to permanently steal Egypt's Sinai and Syria's Golan Heights, like Golda Meir's sadistic "peace will come when the Arabs will love their children more than they hate us")
I can be quite sarcastic myself (as can critical scholars like Norman Finkelstein) but it's best when the information is already at the point where the reader can't in good faith deny it's obvious. + I do bet the balance of pros and cons is shifted where there is real pain like the Holocaust, and real fear like antisemitism (which both really exists, and is exploited to terrify those who were propagandized into genuinely believing the propaganda)
Agree that countering propaganda is never entirely pretty, and even when you argue everything 100% thoroughly and rigorously in 100% good faith, you'll still see many bad faith propaganda replies anyway lol
E.g. when you write a thoroughly fact-checked book published by University of California Press in 2005, Alan Dershowitz will still pop up and try to get Governor Schwarzenegger to ban it, and then when he says no, lie that you you called your Holocaust-survivor mom a kapo
Weird that so many separate individuals here write on the exact same topic and with the exact same style as that perma-banned user LearnsHebrewHatesIsrael.
It's also possible that there's something about the anti-Israel media these people consume that causes them all to write in the EXACT SAME WAY*, but yeah sockpuppets is a more parsimonious explanation.
Hmmm, so the 5 to 7 people I saw the last couple of weeks arguing against Israel's demented forever war are all secretly me? man, I'm fucking awesome.
> anti-Israel media these people consume that causes them all to write in the EXACT SAME WAY*
It's infinitely amusing you're insinuating you're more "Rationalist" than me with a snide sub-remark about an out-of-place style (as if you have personally read and stylistically reviewed every one of the 1500 comment on average to each ot) but you didn't stop to think of Outgroup Homogeneity Bias on your part as an explanation for why all anti-genocide posters sound the same to you, rather than the usual facile one of "anti-sementic media !!!".
What a bizarre and self-righteous pair of characters.
Hmm, I had a number of comment exchanges with LearnsHebrewHatesIsrael ( later LearnsHebrewHatesIP ), some heated, some not, and this commenter seems different to me.
I've made my argument in the thread for why the original post and in particular my exchange here feels strikingly similar to exchanges I had with both LHHI and suspected sock accounts, but I will also acknowledge it's ultimately subjective, and whether or not these feel like sock accounts might be based on the kinds of exchanges one had with LHHI.
I disagree, obviously, and apparently so do a few others. I think this sounds exactly like LHHI experimenting with being unrestrained.
And that's not impossible, of course! I personally avoid writing with frothing hyperbolic aggression, but that's a choice. I'm fully capable of writing at a far lower level, as are you and most of the other people here.
But last...where's this user's blog? If one clicks through his username on a mobile browser here on Substack, his profile page says he hasn't published any posts yet, and he has no subscribers? Again, the mobile browser version of Substack is crap, so it's totally possible I'm missing something there.
At some point, I have to ask whether it matters. If it doesn't walk exactly like a duck or talk exactly like a duck, because it walks and talks a bit more like a stork with avian rabies, of what use is the duck hypothesis?
Naw, you're right about the blog. I was on my phone, just glanced at the page, & mistook the list of blogs the person subscribes to for a list of posts. But I still don't think this sounds like LHHI. As I said, LHHI sounded cleverer and angrier than this person. You're calling the present posts instances of LHHi experimenting with being unrestrained? But when LHHI was unrestrained he was *way* more intense than this person! He called people murderers, said various people deserved to die, swore. You could almost hear him shrieking when you read his diatribes. So that's one difference.
And if you look at the other blogs the person subscribes to, they're all nerdy tech blogs -- no politics. And 2 of the blogs look geared to beginners, and I don't think his work was that low level. At one point I had a long exchange with him about AI training and capabilities and he knew a lot.
I acknowledge my accusation is effectively impossible for me to prove, but based on the occasional exchanges I had with LHHI, this strikes me as being *exactly* the same personality generating slightly different approaches to the text. The formatting of the original comment, (including always proactively including at least one footnote link to create an impression of credibility), the snide, personal-attack-y responses to any push-back, and the general style and tone of LHHI being less intellectual and more bombastic; it all just feel *intensely* LHHI to me. Even the username feels very much in LHHI's confrontational style.
But for me, perhaps the most noticeable signature of an LHHI (or sock) post is the obnoxious lack of acknowledgement or apology for the way his personal tribal affiliation imbues every statement with a fundamental assumption of black-and-white-good-and-evil. That stands out amongst a population of rationalist ACXers who tend to be very careful indeed about acknowledging and interrogating their personal biases.
> "But when LHHI was unrestrained he was *way* more intense than this person! He called people murderers, said various people deserved to die, swore. You could almost hear him shrieking when you read his diatribes. So that's one difference"
But it's also possible that LHHI doesn't want to have his sock accounts banned, and so he's not *quite* rising to the level of what got his "main" account perma-banned (although I believe the first comment in this thread warrants it, sock puppet or no).
Yeah, is someone compiling a list? I think I recognize maybe two sock puppets currently at work, but I haven't been paying enough attention to be sure.
Or perhaps I've been paying too much attention, but that's what the list is for.
I blocked Hindi's Ghost, as I was fairly certain it was a sock (in addition to being obnoxious).
Annoyingly, blocking apparently removes one's ability to see *any* content related to the blocked party - including one's conversations with third parties, like the exchange I'm having with you - so I'll wait until this thread naturally dies off to block.
The topic is a 12-days-old war, and there hasn't been any banning in the last 2 open threads which covered this period, so how could that user have written about a topic that began happening after they're banned?
> so many separate individuals
Perhaps it's outgroup homogeneity bias. There is only so many ways to phrase "Killing civilians is bad and respecting international law is good", when 2 individuals of the same convictions run out of ways in the same way, you tend to notice. When they successfully find different ways to phrase it, you tend to overlook it.
Weird that so many separate individuals here have the exact same pattern of affecting confusion at accusations of sock puppetry and then concluding with deliberately insulting condescension!
I didn't start this by meaning to be condescending, but instead of apologizing like the normal me, your response is strongly shaping me into "If The Shoe fits" as the reply of least resistance.
Why did you assume condescension in my words? I understand why a Pro-Israel Hasbarist could be offended at the (very) mild exasperation, but you didn't participate in this thread or any other one on Israel-Palestine on either side of the issue so I'm confused what offended you in my reply.
> Weird that so many separate individuals
Weird that you keep bringing up how So Many People are sock puppets of me (or, apparently, another user entirely) without actually mentioning the actual usernames. I suppose if you make a direct accusation, those people can chime in here and offer to you whatever will satisfy you that those accounts are different. Or maybe I can point you to meaningful differences of opinions in comments those accounts made before your reply that reasonably point towards different minds controlling them.
As I've replied in suspiciously - even *impossibly* - identical exchanges with commenters who very much appear to be sock puppets of LHHI...nah.
I've reported your initial comment to Scott in the hope you will promptly be banned. I urge those who share my objections and/or suspicion to do the same.
Then I urge everyone to block this user so their content is at least invisible on an individual level.
Scott may not use his ban hammer often, but we can certainly use our own.
Everyone seems.to be retconning It as about nukes, but it was largely about chemical.and Biological. weapons at the time. Iraq had and used such weapons at previous times , but the UN was successfully dismantling its programme at the time of the invasion.
By the time of the actual invasion, Bush had mostly given up on the nuclear weapons charge, but that was after various attempts (aluminum tubes supposedly used for centrifuges, forged documents showing that Iraq had made an unsuccessful attempt to purchase yellowcake) had been discredited. The key sentence from Bush’s announcement of hostilities: “The danger is clear: using chemical, biological or, one day, nuclear weapons, obtained with the help of Iraq, the terrorists could fulfill their stated ambitions and kill thousands or hundreds of thousands of innocent people in our country, or any other.”
It's also that the chemical weapons really are the odd man out from the ABC weapons. Occasionally used nasty warfighting tools with questionable tactical uses but without the vast apocalyptic implication of the others.
This is a bombing campaign, not a ground invasion. These countries will try to build nukes no matter what because of the leverage it would give them, and the only way to guarantee that they won't develop them is to make sure they are physically incapable of doing so. The less countries there are that have nukes, the better. That's all there is to this.
This is a bombing campaign, not a ground invasion. So what does that have to do with countries not having nukes and "being physically incapable of doing so"?
For all the talk about Iraq as an analogy here, pretty much everyone is missing the most important difference. It's unclear whether Iraq seriously wanted to acquire nuclear weapons, and they were certainly nowhere near any such goal, but Iraq was certainly rendered physically incapable of building nuclear weapons for many years.
Because there *was* a ground invasion. Iraq was bombed, invaded, conquered, forcibly regime-changed, and placed under military occupation for about a decade. *That*, is something that can take a nation off the list of potential nuclear powers.
Just the bombing, is mostly just for Feeling Good, like you're a Big He-Man.
The left used to be associated with nuclear disarmament but now it seems to be: nukes in the hands of the US - bad; nukes in the hands of brown people - good!
Oh come on. Ask Google "Where is the Iran-built nuclear power plant" and Google tells you where in its large-font main result. In Bushehr, on the Persian Gulf coast.
You’re adducing a German-built and Russian-rebuilt plant from the 70s as evidence for the peaceful nature of Iran’s nuclear efforts these past 30 years?
There's a trope in armchair geopolitics that gives all agency to the West and none to anyone else. The West isn't forcing Iran to become nuclear-armed. They want to because they want the power, and the West is trying to stop them.
Iran has spent the past twenty years conspicuously *not* becoming nuclear-armed, even though they could have done so at any time during that period. They clearly value having the option on the table, but until fairly recently they valued it mostly as a negotiating point and an emergency capability.
It is not entirely clear how they feel about that now. And you're right, it's entirely their decision and their responsibility. But the US and Israel have deliberately and substantially changed the incentive structure.
There's a trope in Western internet commenters where the West has zero agency, and the fact that we keep attacking and trying to dominate Middle Eastern countries has zero to do with the fact that they want to deter us.
Yeah I get that they don't like Pax Americana, and theoretically more of those places having nukes would make us have to respect their autonomy more than we do. But not only is it dangerous to proliferate such weapons, it just turns those nations into little Americas in their own neighborhood, they were never gonna need to nuke their tiny rivals but they still become regional hegemons that get more of what they want on the big stage and can leverage that advantage to stay there. It's not any real kind of equity between peoples as they'd simplistically imagine it.
>"These countries will try to build nukes no matter what because of the leverage it would give them"
False. There was an anti-nuclear Iran deal, which Trump withdrew from because of his ego.
In fact, the US has consistently *punished* nuclear disarmament. Saddam had no nuclear program after 1991, so the US lied that he did and destroyed Iraq in 2003. Gaddafi gave up his nuclear program after 2003, so the US destroyed Libya in 2011.
The anti-nuclear Iran deal was to end by about now, only with Iran having a much better economy and be better able to produce nukes, possibly shortening their nuclear timeline while making their people better off.
Though I agree with Trump that the Iran nuclear deal was a bad deal, I disagree that he should have pulled us out of it. Because what one president does concerning another country ought not be unilaterally undone by another president, because it undermines the integrity of the United States.
It isn't just anti-nuclear though, the other threat is to the dollar's pre-eminence, Saddam rashly started dealing oil in Euros. Gadaffi had some idea of starting a bank of Africa, which would be independent of the dollar. Also, I think in the early 2000s Iran founded an oil "bourse" to sell oil in currencies other than the dollar, so they've been long overdue for a clobbering. No doubt an agreeable side effect for the US of the latter is to persuade Putin that Trump can take decisive action and isn't all mouth as Putin may have believed.
The WMD evidence against Iraq was clearly ginned up. Do you believe that Iran is not illegally enriching Uranium? There seems to be no dispute on that point.
Do you mean to say you believe that Iran *is* illegally enriching Uranium?
Because two days ago, the President of the United States and the Prime Minister of Israel assured us that Iran's nuclear facilities had been "totally destroyed". If they're still enriching Uranium, what was the point?
FWIW, I don't believe Iran is enriching uranium *today*, though I'm not hugely confident in that. Better to lay low for a few days or weeks and figure out how best to proceed. But I'm betting that Iran will be enriching an awful lot of uranium, 8-12 bombs' worth, later this year.
1. Bombing Saddam, who had no nuclear program of any kind in 2003, helped convince Iran it should have a latent capacity to change its mind later if needed to deter us. If we bomb Iran for knowing it might need to deter us later, then we're begging it to start deterring us now.
2. The US's motives have nothing to do with legality or self-defence, as proven in 2003 in Iraq (as well as by destroying Libya in 2011, to punish Gaddafi for cooperating and giving up his nukes program in 2003, etc).
3. Iran's enrichment is legal under NPT when used for nuclear energy. Though since Trump withdrew from the anti-nuclear Iran deal in 2018, it has stepped it up, and recently IAEA said it was in violation of something related to inspection / declaration, but that's not at all proof of actually building a bomb.
4. US spies and the IAEA agreed that there was no evidence of Iran actively building nuclear weapons.
5. The only country in the middle east that actually *has* nukes is Israel, and few think anyone should bomb Israel just to destroy its nukes (which tbf would also be suicidal). Israel also isn't a signatory of NPT and doesn't allow IAEA inspections (technically its nukes are an open secret, for tricky Israel-US reasons).
6. The US isn't actually scared of an Iranian first strike (which Iran knows would be suicidal). It's scared of not being able to totally dominate the Middle East anymore. Unfortunately that same will to dominate is why it keeps, over and over, giving them a reason to feel the need to get nukes.
I think the difference here is that Israel has not declared an intention to wipe Iran off the map by any means necessary, and does not fund Jewish terrorist cells that launch rockets into Iran and suicide bomb Iranian buses. For all Netanyahu’s faults, he seems unlikely to launch a nuke into Tehran tomorrow, whereas if the Ayatollah were to get a nuke the fate of Tel Aviv is far less certain.
Not sure what is your point here. Obviously, Iran is not the good guy here, but the fact that Israel is somewhat better than Iran does not make it the good guy either.
Also, it is a bit funny to portray Israel as one avoiding civilian casualties as much as possible (this is understandable, though -- as long as they can get away with everything by yelling "human shields", why take the risk?).
I’m not saying Israel is the good guy. I think both Netanyahu and the Ayatollah are evil psychopaths who have visited untold misery on the region. But Israelis have more of a chance of getting out from under Netanyahu, and if I had to pick a place to live, I would choose Israel.
There is zero chance of Iran destroying itself by nuclear first-striking Israel, which Iran knows has nukes.
Again: Iran is a country with goals and strategies, not your cartoon propaganda’s suicide death cult.
(The chances of Netanyahu nuking Iran are also low, I hope. But a risk there for Israel would be if Netanyahu starts to fall for his own propaganda—that he starts to actually think Iran is a suicide death cult immune to deterrence logic—so he mistakes their deterrence for aggression, nukes them in the hopes that Pakistan’s deterrent umbrella is a bluff, and then it isn't a bluff.)
You don’t think so? The regime in Iran is widely despised by the people. What if the Ayatollah was facing revolution from within and the threat of Reza Pahlavi replacing him, while Israel was striking their military sites from the air? You believe in the capacity of evil people who have nothing left to lose, to exercise restraint on the behalf of humanity?
1. Whatever Iran's motivations are, they indisputably want to become a nuclear power. I don't think you have to blame the US for that. Most countries would like to be a nuclear power. In any case the origin of their motivations is irrelevant. This is realpolitik not therapy.
2. So? Our motives are our own self-interest. That's sufficient. There are many good reasons to not want an unhinged Muslim theocracy to have a nuclear weapon.
3. Iran is enriching to 60% which is far beyond the 5% needed for energy. They've also denied inspections. "That's not at all proof of actually building a bomb" is suicidally naive in my view. What's proof, Tel Aviv disappears? Given the stakes, proof isn't necessary. Preponderance of the evidence suffices. This isn't mock trial, it's the real world with real consequences. Better to kill ~1000 Iranians on a reasonable hunch then risk the death of a million Israelis.
4. Again, willfully naive. They're doing enrichment which is the hard part.
5. Israel is a stable ally. It's in our interests for them to have nukes and they've demonstrated that they can be responsible. Also they already have them so there's nothing to be done. This illustrates why we have to be proactive in preventing Iran from obtaining them. Once they do our options narrow.
6. Yes, we're pursuing our own rational self-interest. I don't know why you find that objectionable. Existential fear isn't a prerequisite for military action. We're the global hegemon. We decide what's appropriate.
>"is suicidally naive in my view. What's proof, Tel Aviv disappears?"
You'll recall Bush's focus-grouped slogan in 2003, to stop anyone questioning his lies to justify his criminal aggression: "We cannot wait for the smoking gun to be a mushroom cloud."
You'll also recall that Iran isn't a suicidal death cult, or a country of idiots who don't know about Israel's nukes. Obviously Iran isn't going to destroy itself by nuclear first-striking Israel.
>“An Iranian government representative admitted in a closed-door sub-commission hearing that children did participate in the war against Iraq,” Refworld continues. “In a series of rulings issued in the autumn of 1982, Ayatollah Khomeini declared that parental permission was unnecessary for those going to the front. … Iranian officers captured by the Iraqis claimed that nine out of ten Iranian child soldiers were killed.”
>“Boys as young as nine were reportedly used in human wave attacks and to serve as mine-sweepers in the war with Iraq,” Refworld found.
>Often bound together in groups of 20 by ropes to prevent the fainthearted from deserting, they hurl themselves on barbed wire or march into Iraqi minefields in the face of withering machine-gun fire to clear the way for Iranian tanks.
You said:
>Obviously Iran isn't going to destroy itself by nuclear first-striking Israel.
Yes Bush was likely wrong. An argument being wrong for one set of circumstances in 2003 doesn't make it wrong for a different set of circumstances in 2025.
>Obviously Iran isn't going to destroy itself by nuclear first-striking Israel.
That's not obvious at all. In any case, it's in our interest to prevent them from obtaining nuclear weapons. That's really the only thing that matters.
I'll also point out the issue with Iraq II isn't really that they didn't have WMD, it's that we were completely and naively unprepared to rehabilitate the Iraqi political structure. If it had gone as well as our post-WW2 occupation of Japan then no one today would care that Bush lied about WMD. This is a results business. As long as we prevent Iran from obtaining WMD without significant cost to us then that's the only thing that matters.
That's Bayesian evidence that Iran is working towards nuclear weapons; we would certainly be more likely to find large stockpiles of highly enriched uranium if Iran was aiming for nukes than if it wasn't. (And Iran's accusation of the IAEA looking at forged Zionist documents isn't credible to me; Iran has approximately zero credibility here. Nations intent on proliferation always deny it, and Iran specifically shamelessly lies to the media all the time.)
> few think anyone should bomb Israel just to destroy its nukes (which tbf would also be suicidal).
Would it? Whatever nukes Israel has, it lacks the ability to launch them very far, so if there were ever a US President who decided "actually we really do care about the non-proliferation thing" there's not much the Israelis could do to respond.
Israel has intercontinental ballistic missiles, though numbers and details are quite fuzzy. Israel also has submarines capable of launching nuclear cruise missiles, and while it would take them a few weeks to sneak across the Atlantic, they're reasonably quiet subs in a big noisy ocean.
I suspect it would, as the United States would likely retaliate in its behalf. Since Israel isn't actually a NATO member, the US is under no obligation to do so, but it seems very likely, since the US gave Israel nukes in the first place.
So it seems. I thought we actually gave them nuclear weapons, but it seems they developed them themselves with a secret reactor with French assistance.
So my backup to the estimation that the US would retaliate to an attack on Israel was wrong, but I will still say that, given the support the US has provided to Israel in the past that they would still retaliate. Poking Israel is like poking the US, only a lighter poke.
I assume it has submarines and second-strike capability. Ofc it's hard to know specifics given the ambiguity policy.
(The point is more illustrative, like "few here would think Iran has the right to try to attack to destroy Israel's nukes."
In real life, if it was the US who decided that, it wouldn't need to attack. It's Israel's biggest enabler, so all it would have to do is say "no more settlements, no more military aid, no more occupation, time to withdraw to the green line and make peace with your neighbours and sign NPT. The PLO and Arab League are already down, and if we release Marwan Barghouti he can convince the rest of the Palestinians.")
In 1981 Saddam had bought a French-made nuclear research reactor, and France-supplied highly enriched uranium (the standard fuel at the time). It was above board, legal, under IAEA supervision and everything.
So Israel bombed it, trying to stay the only nuclear power in the region.
But the predictable and actual effect was that it *convinced* Saddam he *did* need a nuclear weapons program after all, and that it should be secret. The Iraqi secret nukes program, created as the consequence of Israel's counter-productive attack, was only ended after 1991 by the UN.
Yes. (That was different in the sense that it was UN-approved—which might also affect how Saddam would interpret what it means for future threats and deterrence—but nevertheless, yes.)
Yep. You can't just say "some guys claimed this about another country 20 years ago but were wrong, therefore these other guys are also wrong about this other country."
You can say "Given the known fact that Iran was not building nukes and about to nuke us; and the known fact that the same people are telling the same lies about it for the same reasons as they lied in 2003; it's probably deliberate, not by accident, and anyone who falls for it again is not misguided but a sadistic warmonger."
You could say it but it's obviously wrong. They are not the same "lies" by the same people or for the same reasons. Moreover, Iran clearly was "building nukes"- i.e. taking steps to make it possible to build nuclear weapons very quickly, that had no possible peaceful purpose.
Having nukes is a fairly good guarantee against being attacked. Its the main reason any country has them. Since trump killed the JCPOA, it became obvious to Iran that guarantees made by the US were not to be trusted. This is all a fairly obvious outcome of escalatory brinkmanship.
Very much agreed, especially with that last part. Between Russia invading Ukraine and this action against Iran, if I were in the leadership of a non-nuclear country right now, I would be looking into every possible realistic avenue of getting nukes. It's sure looking like "might makes right" is the guiding principle for most of the world's leading powers today, and I wouldn't trust anything they said to the contrary.
Which is, ultimately, very sad. While still far from perfect, humanity seemed to have made pretty remarkable progress in the past few decades of leaving behind the violence of our past. Now we seem to be backsliding.
I do not really understand what you say. Principles are for intellectuals. The reality on the street is, people do not kill their neighbors inside countries, because there are courts enforcing laws. Now international relations are always anarchy, there is international law and the international criminal court, but crucially it can only be enforced by war. So internationally, courts cannot enforce law without a war first. Milosevic had to lose a war first, then when to the ICC to be charged by the violation of international law.
So no, might does not make, but enforce right. Iran violated international law, one of the most important ones (non-proliferation), and it can only be enforced this way.
" The reality on the street is, people do not kill their neighbors inside countries, because there are courts enforcing laws."
But that isn't the reality on the street. That has never, ever, in human history been the reality on the street. That "reality" is a story made up by cynics to justify their cynicism or by monsters to assure themselves that everybody else is secretly as broken inside as they are.
Here in actual reality, the vast, vast majority of people would not kill their neighbors even if they were absolutely sure they would never be caught or punished. There is a small minority that would, and a larger minority (but still very much a minority) that *might* if circumstances got out of hand. But it simply doesn't take very many willing murderers to ruin things for everyone else, so the world with laws and courts ends up much better than the world without.
Regardless, none of that is much relevant to international relations at all. Countries are not people. They do not have the same decision processes as people, the same constraints, vulnerabilities and limitations as people,. nor the same virtues and strengths as people. They are very different things, and the analogy to civic law breaks down very quickly.
Lots of international agreements work just fine without any sort of violent enforcement mechanism. They work because cooperating and playing nice often benefits all parties involved. Even when a country encounters a specific case where they would benefit by pushing the "defect" button, the short-term payoff is rarely worth the loss of future opportunities that accompanies that hit to your reputation. A lot of leaders and people in high government positions understand that very well.
Unfortunately there are several prominent leaders who do *not* seem to understand this, or who rationalize away that understanding whenever they see a defect button that they *really* want to push. In the longer term their countries will do worse and lose power and influence against the countries who consistently cooperate. (If human civilization overall survives and flourishes into the longer term.)
Now the Iranian regime is a pretty terrible government, and they themselves are not so great at avoiding the "defect" button. But in large part that's because the rest of the world--and the U.S. VERY much in particular--has shaped them that way. Consider that the current regime took power after Iranians had spent nearly two decades getting systematically fucked over by an autocratic puppet government supported by the U.S., all because Iranian oil was more important to Uncle Sam than Iranian lives or Iranian civil rights. Consider that the last time Iran made an agreement with the U.S. regarding its nuclear program--a mutually beneficial agreement--that agreement was torn up *by the U.S.* within a few years of its signing, for no good reason whatsoever, followed not too longer after by the U.S. assassinating one of their top generals on the flimsiest of justifications. From Iran's perspective "international law" no doubt looks like a joke of a concept: no protection against foreign abuses, but often cited to justify them. [1]
So yes, principles are for intellectuals. And intellectuals have spent the past few centuries taking over the world: very slowly at first, but much faster in recent decades. I'll admit that the anti-intellectuals have made a very, very impressive showing this year. A really awe-inspiring bid to drag us all back down into the, to stage a grand comeback tour for Pestilence and Famine and War[2], to really redecorate the place with that retro Dark Ages look. But they haven't succeeded yet. There are still lots of us out here who *do* have principles and are willing to stand up for them.
Relevant further reading if you made it through that wall o' text and can still manage more.
Regarding Cooperation in Modern International Relations Specifically:
[1] If one wants a better understanding of that part of the world, consider that for any reason an Iranian might feel anger at the international community and certain countries in particular, a Palestinian has two or three such reasons. As a people they've been on the receiving end of an absolutely staggering string of abuses and insults, going back more than a century. (And yes, obligatory disclaimer, they've been on the giving end of some as well.) And plenty of non-Palestinian Arabs *can see that* and don't necessarily have a lot of reason to expect that Britain or Israel or the U.S. would treat them any better if they were in a similar position.
[2] Death being a timeless classic that never goes out of style.
>Here in actual reality, the vast, vast majority of people would not kill their neighbors even if they were absolutely sure they would never be caught or punished.
Okay, I sort of over-exaggerated this. Let's say the vast majority would commit financial fraud or take bribes. Culture matters. As a textbook European I don't really get why Americans are so brimming with holiness... I think it might be a Calvinist thing, I think Catholic cultures more accept that everybody is a sinner and it is basically okay.
>Lots of international agreements work just fine without any sort of violent enforcement mechanism.They work because cooperating and playing nice often benefits all parties involved.
This is true for private individuals and corporations too, yet we need law enforcement. In fact because of the principal-agent problem, individuals and corporations ought to need less enforcement, as national leaders have way way bigger principal-agent problems. A marketplace is always better from a principal-agent viewpoint than an election. Like if term limits mean Trump cannot be re-elected, what exactly does he have to lose?
>Even when a country encounters a specific case where they would benefit by pushing the "defect" button, the short-term payoff is rarely worth the loss of future opportunities that accompanies that hit to your reputation.
See above. It is not a country, but the leader pushing the button. Hitler pushed the biggest "defect" button ever, yet a few years after the war, most everybody happily cooperated with democratic West Germany. Just 6 years later, the European Coal and Steel Community was signed - imagine France trusting Germany so much to pool their coal and steel i.e. military industry resources!
>because Iranian oil was more important to Uncle Sam
Actually it was British Petrol.
>And intellectuals have spent the past few centuries taking over the world: very slowly at first, but much faster in recent decades.
But Iran got bombed because they were trying to go nuclear. Shouldn't countries draw the exact opposite conclusion- that if they try to get nukes, they will be destroyed? Surely if Ukraine had tried to get nukes, Russia would have destroyed them at that time anyway, right? And probably if Russia invaded earlier, they would win easily.
Iran *got bombed* because it was suspected of trying to go nuclear. Iran was not *destroyed* over this. The Iranian regime was not destroyed over this. The Supreme Leader was not killed or imprisoned or even deposed over this.
See also North Korea. And Pakistan, which was pretty clearly in bed with Al Qaeda and Osama Bin Laden.
The Iraqi regime, which we were pretty certain didn't have nuclear weapons and wasn't going to be making nuclear weapons any time soon, that *was* destroyed, and Saddam Hussein was dragged out of his spider hole and hung by the neck until dead.
The Libyan regime, which we knew for a fact didn't have and wasn't going to be making nuclear weapons, which had done the "right thing" by thoroughly and verifiably dismantling its nuclear arms program, was destroyed, and Muammar Gaddafi was anally raped to death on Youtube. Libya has been a failed state ever since.
You really think the conclusion Evil Dictators(tm) are drawing is that they should *not* nuke up ASAP?
I suspect the only reason North Korea was allowed to get nukes without getting couped or having their nuclear program raided/bombed is because they have a lot of artillery aimed at Seoul.
The primary takeaway any Evil Dictator should have from the last 40 years is that power speaks, and the West is super consequentialist about things, so make your threats as credible and devastating as possible. Nuke up, artillery up, etc.
The threat of artillery bombardment of Seoul is vastly overrated. North Korea has bignum guns, a small fraction of which have the range to reach the northern suburbs of Seoul and none of which have the range to reach downtown Seoul. This gets misreported as "North Korea has bignum guns and they can bombard Seoul into oblivion!". They now have heavy artillery rockets that could do so, but that's a recent development. Twenty years ago, it was only a modest number of Scuds and the like, roughly what Tehran and Baghdad were inflicting on each other in the 1980s.
North Korea was "allowed" to get nukes because everyone else was stuck in the Buck Turgidson "ignorant commie peons don't understand technology like our boys do" school of threat assessment. I met far too many people, far too highly placed, who were assuming that North Korea couldn't *really* build nuclear weapons five years after their first fully successful test. And since "everybody knew" North Korea can't really build nuclear weapons, we didn't really feel the need to do anything. Then it was too late.
That mistake, at least, we aren't repeating with Iran.
Interesting. But you think this really won't be enough to slow Iran significantly, right? I think I saw elsewhere you were predicting nukes within 12 months. Are centrifuges that easy to quickly obtain and deploy? Or are you just figuring that we didn't eliminate enough of their current stockpiles and equipment?
Some decisions are overdetermined. Multiple independently sufficient motives that all are based in actual facts don't undermine each other's legitimacy. Israel has many, many reasons for bombing Iran, most of which boil down to "they hate us and have been waging a (mostly, but not entirely proxy) war against us for decades."
I think the question is not why Israel hates Iran (or, when we are at it, vice versa), but that what is the concrete reason for the actual ("pre-emptive") attack. And I don't think it is irrelevant, for several reasons. First, a rational expectation is that they end the war when their goals are met. If the war is about nuclear warheads, they don't start bombing police stations; if the war is about destroying the Theocracy of Evil, that is an entirely different matter.
And I don't think "we hate their guts, we are mortal enemies for like forever, that's why" would have met with an applause on the international stage.
But communicating one thing first, then another second, then a third later gives me the impression that they are throwing whatever plausible explanation crossed Netanjahu's mind to see what sticks.
(Heck, they could even give a list why, but AFAIK this is not what happened.)
Of course, the fact that Israel's current government lost a lot of its credibility in the last ~20 months does not help either. Overall, their communication is aimed to cover up whatever they can get away with. Which is not worse than the median in the region, not at all -- but enough to raise doubts about the official communication.
I think that's a kind of a silly standard to hold Israel to. Iran and Israel have been at war for decades now, and the only reason Israeli and Iranian troops haven't directly fought much is that there are neutral countries in between them. Instead, Iran has used proxies (and as recently as last year, its own military) to bombard Israeli cities with thousands of imprecise terror weapons. There's been no ceasefire since those attacks. No armistice. Not even a softening of relations in any way.
In other words, I don't think this is reasonably framed as a pre-emptive attack at all. The correct frame is that they were already at war. There's no other name for a state of affairs where one country is launching thousands of missiles at another. They're at war, have been at war, and throughout the course of that war Iran has mostly been the aggressor. (Admittedly, that's probably because Iran has a lot more ability to cultivate proxies than Israel does, not some moral superiority on Israel's part.)
And when you're in a long-standing war, the reason to strike back at enemy military targets is: A) you can, and B) the juice is worth the squeeze. All of Netanyahu's statements make a lot more sense when you realize he's explaining specific war tactics, not trying to justify starting a war.
> When the official explanation is changing so frequently, one cannot help but have the feeling that none of them is true.
Or because those things are related.
(You could make a similar argument about WW2: "First they said it was about Czechoslovakia, then they said it was about Poland, then they said it was about France, now we have this story about concentration camps... the constant change of the narrative makes me suspect that the so-called Nazis are just a boogeyman invented by American media.")
You are right, let me make it more clear. When I wrote, "none of them is true", I meant it in the sense that none of them is the real reason of the attack, not in the sense that they do not hold some truth. Of course they are mostly true -- but that they are the reason for the attack seems bogus after such swift changes.
"They are related" seems a fuzzy concept to me, though; I mean, of course they are related in the sense that all has something to do with Iran, and all are things which sound quite bad for the international public. And this is exactly why they can be ideal BS PR for such a bombing campaign.
Which, again, do not make them untrue; they just might not be the real reason for bombing. (Also, you can't stop Iran to be an "evil theocracy" by a bombing campaign.)
Considering your examples, would the allies state they go to war with Nazi Germany because of the concentration camps, that would be outright BS as well. It does not mean, though, that concentration camps existed, or that the war was the reason these stopped operating, or that it is good that they stopped operating.
"But Iran got bombed because they were trying to go nuclear."
This has by no means been adequately demonstrated. They didn't dismantle their nuclear program entirely. And I'd say they were obviously interested in keeping the option of building up to weapons capability in a medium-short, rather than long time horizon open[1]. But the claim that they were actively trying to build a bomb at the time they were attacked is one that needs significant support that I haven't seen yet.
[1] Which to my understanding is a state that many non-nuclear developed countries are perpetually in, simply by virtue of having the right mix of industry and technology and scientists. It's actually a little bit scary how *simple* nuclear weapons are when you get right down to it.
>But the claim that they were actively trying to build a bomb at the time they were attacked is one that needs significant support that I haven't seen yet.
The IAEA looks convinced that Iran prevented the IAEA from ensuring that Iran was not building a bomb.
And they enriched uranium to 60% U-235, while power reactors only need 5%. The _point_ of making such highly enriched uranium is a nuclear _weapons_ program.
I sent a thank you note to Trump after he authorized the bombing of the Fordow uranium enrichment site. I consider bombing it to be the right choice.
None of that particularly contradicts what I had just written. Avoiding international inspection and enriching uranium to 60% are of course both consistent with a country actively working towards a bomb. But they're also consistent with a country that would like to keep the option of getting there quickly, but without larger the provocation of actually assembling the weapons.[1] I don't especially have an opinion either way: it's plausible that they were trying to get to a bomb, it's still far from certain. As a general rule, you shouldn't trust motivated claims that come from unreliable sources even when they align with an other-wise plausible reality. Good lies have grains of truth. But just because someone is a habitual liar doesn't mean they're never truthful.
[1] I'll note that if this is the case, they either miscalculated how much of a provocation their intermediate efforts actually were, or simply underestimated how ready Netanyahu and Trump would be to jump on an excuse for belligerence.
Ok, I think we are just splitting the difference between "steps directed towards making a bomb" (including enriching uranium to 60% U-235) versus "final assembly of a completed weapon" (or maybe even "arming a nuclear bomb" ?).
>it's plausible that they were trying to get to a bomb, it's still far from certain. As a general rule, you shouldn't trust motivated claims that come from unreliable sources even when they align with an other-wise plausible reality.
Look, Iran hasn't denied that they have 60% enriched uranium. That, _alone_, makes it clear that they are/were trying to get a bomb. 60% U-235 certainly isn't intended for power reactors!
>It is also important to consider that Iran could use its stockpile of 60% enriched uranium to make weapons directly without the need for further enrichment. There would be limitations in the delivery of such weapons, as discussed below.
( Basically, they could build a gun-type bomb with their existing 60% U-235 with no further enrichment, albeit it would be heavier than a 90% U-235 weapon. )
That's the largest stockpile of >nuclear power grade uranium in the world, aside from nations that already have nuclear weapons. And according to the IAEA, it's increased by 50% in the last few months.
I can't see an explanation for this that isn't attempting to either A) get nuclear weapons or B) make getting a nuclear weapon something they can achieve in the very short term.
I agree with the last paragraph, but I think A) and B) are pretty importantly different here. In particular, I think a world in which we were confident the answer was B) and not A) would be a world in which the bombing would be *exactly the wrong move.*
Re: footnote one, any country with a domestic nuclear power program (not just a reactor, but producing their own fuel) could stand up nuclear weapons within 2-3 years. The main players there (aside from the countries that already have nuclear weapons) are Germany, Japan, Canada, Sweden and South Korea.
Thermonuclear weapons (ie H-bombs) would take longer, but that's more because designing and building an H-bomb is notoriously difficult - an implosion-type A-bomb is a pretty straightforward engineering problem once you have the enriched uranium and you know how to machine it (both of which you have as a result of the domestic power program).
There's something involving "lensing" in making an H-bomb which I have deliberately avoided finding out about because I would like to be able to travel the world without constraint, but it's famously a hard problem.
"There's something involving "lensing" in making an H-bomb which I have deliberately avoided finding out about because I would like to be able to travel the world without constraint, but it's famously a hard problem."
Ah yes. It happened exactly once in my education that a professor unwilling/unable to fully answer one of my questions because "some of the details are classified." The part of the answer that he did give was pretty much "something involving lensing."
The details of the Teller-Ulam design (for an H-bomb) are “born secret”, ie they’re secret under US law even if you reinvent them yourself.
Some of them were published in a magazine called “The Progressive” in 1979, but how accurate or complete they are is unknown.
It was notoriously difficult, because both the USSR and UK programs had partial information from the US (in one case by spying, in the other given voluntarily) and still struggled to get it right, and the other two open programs (France and China) also took a long time.
I think you're mixing up plutonium and uranium A-bombs. Uranium A-bombs don't even require implosion; just two chunks of uranium you slam together. If you can make bombs, and you have a critical mass of highly enriched uranium, they're child's play.
Plutonium on the other hand has pretty low neutron production, so you need implosion explosives to trigger them. That's apparently a difficult problem, at least to the point that it made North Korea have multiple fizzles.
Jeffrey beat me to the punch, but my understanding was that the implosion was to ensure that the plutonium had time to react fully before the explosion spread it too far apart.
I believe that similar considerations are what make H-Bombs difficult: simply packing deuterium around a fissile core will give you extremely high temperatures and pressures, but with such an abysmal confinement times that your triple product remains quite low.
>Plutonium on the other hand has pretty low neutron production,
Huh?
>so you need implosion explosives to trigger them.
Umm... AFAIK, the problem is that the Pu-240 impurity in a Pu-239 pit has a _high_ spontaneous fission rate, so it produces a _high_ neutron background rate, which can trigger a premature chain reaction when the pit is only partway to being fully imploded, and that gives you a fizzle. That's much of the motivation for using high explosives to make the implosion fast, and reducing the chance of a fizzle (you also want to compress the plutonium as much as possible, to get a given mass of plutonium plus reflector plus tamper as supercritical as possible to get the yield up)
Saddam completely gave up his nuclear program after 1991, and they punished him for it in 2003.
Gaddafi gave up his nuclear program in 2003, and they punished him for it in 2011.
(Plus all the other coups and massacres and wars of criminal aggression that had nothing to do with nukes. E.g. overthrowing Iran's secular democratically-elected PM to install a dictator to protect oil profits in 1953; btw that, plus his overthrow in 1979, is why Iran is a theocracy now.)
North Korea has nukes, so the US only sanctions it instead of invading or bombing.
The US has, through its actions, tried hard to prove that no one is safe from US aggression without a nuclear deterrent. This is very foolish.
> overthrowing Iran's secular democratically-elected PM to install a dictator to protect oil profits in 1953;
Of all the inaccuracies in the Standard Internet Commenter View Of History, the idea that Iran was meaningfully democratic prior to 1953 is one of the most pernicious.
The reality was... very complicated. There were elections but they were regularly rigged, politicians had a tendency to get assassinated if they didn't agree with the Shia fundamentalists, and the 1952 election ended with only 79 out of 136 seats filled because Mosaddegh did a literal "Stop the count!" move once a parliamentary quorum was reached and before the votes from rural seats which would have opposed him could be tabulated.
The post-1953 system also had rigged elections, though for a while it seemed to be heading in the right direction with multiple competitive parties (albeit all Shah-approved) then in 1975 they just said "fuck it, actually let's just have one party and outlaw all the others".
I think 1953 is best seen not as a transition from democracy to non-democracy but just another speedbump in the shitshow of rigged elections, assassinations, coups, revolutions, party-bannings, and neverending conflict between royalists, Islamists, communists, landlords and the military that characterises the last century of Iranian politics.
Right. The world is full of crappy dictators who are allowed to stay in power because they don't cause any particular trouble.
While Ayatollah Khameni is hiding out in a bunker, the Emir of Dubai is banging bikini models on his yacht, because making friends is better than making enemies.
On the other hand, it is also obvious that the US made sure to have the overseas empire of the UK collapsed. (It was happy to see the collapse of the French one as well, but actively contributed to the former.)
Correct. US and SU were competing for the colonies, but agreed that they will have them, not the UK and FR. So basically textbook great power politics.
There are only two options. You are either friendly and cooperative, or totally crush them. What does NOT work is some intermediate level of violence, and that is IMHO the mistake the US and Israel tends to make. That is, use enough violence that they hate you, but not enough that they fear you.
Uh...really? "The US preferred hegemony to survival"? I mean, it's not like there were any *other* countries that were pursuing hegemony after WWII...
I honestly wonder if you realise that you have, with a comment like this, absolutely ensured that 90% of readers, many of which may have been highly open to your viewpoint, will now permanently disregard everything you say on the matter. More generally, I'm just amazed at how many people trying to advocate for a position have an unbelievable talent for appending a single sentence that compresses "I have a complete lack of perspective", "I have no awareness of the slightest factual nuance" and "I'm a fanatical ideologue" into that sentence.
Yes really. (You know about all the coups, all the overthrowing democracies to install dictators, all the terrorists and massacres, all the undermining of international law, right? In the nuclear age this isn't just wicked, it's an existential risk.)
>"It's not like there were any *other* countries"
Did I say there weren't? But if you're American, you should understand what's going on so you can be an effective citizen. (And it is the only superpower so it has an enormous impact.)
>"will now permanently disregard"
I hope they prefer survival, over encouraging nuclear proliferation so they can pwn david bahry in the comments section of astral codex ten dot com.
You're the one who said that after WWII there could have been peace and stability, if not for the US. The only sensible way to parse your comment is as saying that the US on its own ruled out peace.
If you're now trying to claim that you only meant the US played *a* part, that's a pretty blatant motte-and-bailey. Otherwise, you would have to say that a comment saying "there was a chance for world peace, but Iran chose jihadism instead" would get total unchallenged agreement from you.
The US played an enormous and consistent role, which it is still playing, then. If you want to compare the amount it did vs what the USSR did then feel free (I'm historically curious, though it's little relevant to talking to Americans on an American blog about what America is doing).
Again, you're either saying that the US played *a* role among others in destabilising the post-WWII world, or you're saying it played the primary role. If the latter, I suggest you look up the Yalta pledge. If the former, this is an almost textbook motte-and-bailey discussion.
"Colossal Biosciences" is the company that pretends to de-extinct dire wolves and so on. It's really just living paleoart—fun, but not remotely what their hype claims. E.g. the "dire wolves" were really just grey wolves with a handful of genes edited to make them big and white etc.[1] For fur colour, they couldn't even use the actual dire-wolf gene, b̶e̶c̶a̶u̶s̶e̶ ̶t̶h̶e̶ ̶g̶r̶e̶y̶ ̶w̶o̶l̶v̶e̶s̶ ̶a̶r̶e̶ ̶s̶o̶ ̶d̶i̶f̶f̶e̶r̶e̶n̶t̶ ̶f̶r̶o̶m̶ ̶d̶i̶r̶e̶ ̶w̶o̶l̶v̶e̶s̶ ̶t̶h̶a̶t̶ ̶t̶h̶e̶ ̶d̶i̶r̶e̶ ̶w̶o̶l̶f̶ ̶g̶e̶n̶e̶ ̶w̶o̶u̶l̶d̶ ̶h̶a̶v̶e̶ ̶m̶a̶d̶e̶ ̶t̶h̶e̶ ̶g̶r̶e̶y̶ ̶w̶o̶l̶v̶e̶s̶ ̶d̶e̶a̶f̶ ̶a̶n̶d̶ ̶b̶l̶i̶n̶d̶.*[2] The misrepresentation of what they're doing isn't just cringe—it's dangerous, given how Trump wants to use it as an excuse to try to remove protections from endangered species.[3]
My supervisor Dr. Vincent Lynch, apart from his research e.g. on cancer resistance in elephants [4] and other cool stuff, is a critic of Colossal's hype (while admiring some of the genuinely important biotech that Colossal is refining, which could be redirected to other genuinely useful purposes).[5]
The gossip part: it looks like Colossal's finally started using intimidation tactics. First, some bizarre hit piece articles appeared trying to discredit his critiques (but his critiques are standard—Colossal itself admits that calling gene-edited grey wolves "dire wolves" is just "colloquial"). Second: when he suggested that the articles are AI-generated and might be paid for by Colossal and its billionaire cofounder (obviously plausible given the incentives + clearly disclaimed as opinion i.e. protected speech + clearly related to public concern given Trump's intent to remove protections), Colossal had their lawyers send an intimidating "shut up about it" letter to his front door.
Ironically, Colossal's behaviour is more humiliating for them than scientists' critiques ever could be. The hype train was rolling and not going to stop; people want to believe that we have dire wolves and will have mammoths. But people don't like billionaire Elizabeth Holmes-ass behaviour to silence critics.
*Added: this part was probably my misinterpretation, inherited from some early reporting. More likely the issue was that these genes are risky to mess with in general (e.g. in the event of imperfect CRISPR edits?), rather than a specific incompatibility with the dire wolf version. Thanks to Catmint for pointing this out. https://www.astralcodexten.com/p/open-thread-387/comment/129267704
> For fur colour, they couldn't even use the actual dire-wolf gene, because the grey wolves are so different from dire wolves that the dire wolf gene would have made the grey wolves deaf and blind.[2]
This is as far as I can tell a miscommunication between scientists and laymen. Actual quote is:
> The dire wolf genome has protein-coding substitutions in three essential pigmentation genes: OCA2, SLC45A2, and MITF, which directly impact the function and development of melanocytes. While these variants would have led to a light coat in dire wolves, variation in these genes in gray wolves can lead to deafness and blindness.
For context, many species have minor variants of these genes which affect coat color (usually lightening it). For example, a variant of OCA2 causes blue eyes in humans, while a variant of SLC45A2 is responsible for palomino horses. White spotting patterns in dogs are associated with MITF variants.
However, in mammals, maybe vertebrates or higher, major disruptions can cause negative side effects. For example, complete deletion of MITF is lethal, while major disruptions result in an all-white coat and often deafness. Certain mutations of it in mice are also capable of reducing eye size, which probably doesn't help with vision.
OCA2 and SLC45A2 are much less serious. An OCA2 variant is the most common cause of blue eyes in humans. Blue eyes are known to sometimes cause minor difficulties with vision especially involving bright lights, however the effects are small as you can find out by asking any blue-eyed human. The causal mechanism is differently reflected light by the iris, so this is not gene-specific. SLC45A2 and can also cause blue eyes. I do not know of any cases where full blindness was caused in any species by any mutation to either of these genes. MITF too can cause blue eyes, but this is not the concerning thing about it.
I do not know of any studies of OCA2 or SLC45A2 showing a causative link to blindness. Closest I found was "There was no statistically significant difference in vision between patients with OCA1 versus OCA2, or between patients with and without mutations. Patients with two mutations tended to have worse vision than those with one, but not statistically significantly." from https://www.sciencedirect.com/science/article/abs/pii/S1091853116301823 (do not be mislead by the word "blindness" in the introduction - it may be referring to a different gene)
For more on MITF mutation viability in canines:
> In humans, deleterious MITF mutations cause disorders of vision and hearing, including the Waardenburg and Tietz syndromes [9], [12], [13]. Deafness has also been recorded in white dogs, where approximately 2% of white dogs (sw/sw) present with bilateral deafness and 18% are unilaterally deaf [14]. The majority of mutations reported in mice and humans that cause severe pleiotropic effects are generally loss-of-function mutations affecting the coding regions [15]. This is not the case with dog MITF alleles. A comparison of S and sw haplotypes, using BACs from an S/sw heterozygote, across the 100 kb canine white spotting candidate region revealed 124 sequence polymorphisms, all of which were located in non-coding regions [6]. This demonstrated that the extreme white coat colour phenotype is controlled by one or several regulatory mutations. This hypothesis is strongly supported by the fact that coloured patches on white spotted dogs display normal pigmentation. Thus, this suggests that the canine MITF variants primarily affects migration and survival of melanocytes during development, but have no or only minor effects in mature melanocytes in the hair follicle; pigmentation of the hair requires MITF protein expression [16].
That paper also has a table of MITF variants found in wolves. So, this isn't a deal where gray wolves have something special going on that makes MITF variants uniquely harmful, but rather that major disruptions to the gene _can_ in some species have harmful effects, and (here I enter the realm of speculation, as I know the theory but not the practice) perhaps they were concerned that CRISPR would not modify the gene accurately enough to include just the desired substitution without damaging the rest.
I think you're right. Although e.g. Zimmer's NYT article said "they did not introduce the remaining five [mutations], because previous studies had shown that those five mutations cause deafness and blindness in gray wolves"[1]—which is might be where I got that interpretation from—the press release itself only said "variation in these genes in gray wolves can lead to deafness and blindness."[2] So yes, I now agree it's probably more that they considered these genes risky to mess around with in general, rather than a specific incompatibility with the dire wolf version.
Leopard spotted horses, which aside from being cool molecularly are also a very old pattern which may have helped with camouflage during the ice age. The first part is a retroviral insertion disrupting the TRPM3 gene. By itself this causes a normal-colored foal except with striped hooves, which as it grows gradually gets more and more white hairs throughout the coat. But combined with a certain variant of another gene, the whole horse can be white with dark dots. Nobody has actually proven the mechanism, but the setup and the effect are similar to spotted corn, so it probably also works by transposon. If that's so, then the second gene is a special activator for the retroviral mechanism, allowing it to copy itself out of the DNA and into other parts. So maybe in some cells the transposons copy themselves right out of the DNA and restore the gene's function (the colored dots), while in others they copy themselves further or otherwise mess up that section of DNA even more (the white regions). Speculation, but the spotted corn definitely does it that way and horses are cooler than corn.
Kittens have mittens because it's very cute and it even rhymes! What's going on there is not white being added but color failing to be added because the cells for it don't grow in that part of the skin. For color you need melanocytes, and melanocytes develop from melanoblasts, which start at the spinal region and during early embryonic growth spread across the skin of the whole animal. Except if the genes are not working right, it doesn't do that, and instead the cells only reach some areas. Because the paws are far away from the spine, they are one of the last places where the pigment cell precursors arrive, so if things get slowed down it's easy for the paws to miss out on getting color. (And this starts before the tail has grown out, so the tail tip also counts as far away from the starting location. And the path on the head is around the sides meeting in the middle, so if it doesn't quite make it you get a little pink nose with white strip and then you have to really hope the cat will let you kiss the nose.)
I'm sorry you were triggered by hearing relevant info about what the administration is, in fact, trying to do and for which the administration is, in fact, citing Colossal's claims as justification.
I'm not sorry you are triggered that your unscientific horse manure was called unscientific horse manure.
There is a very interesting discussion to be had regarding the scientific merits of endangered species protections and the unintended consequences of such actions (protections seldom work as intended and often inhibit more productive solutions). I enjoy those discussions.
Sneering leftist drivel composed solely of reducto ad Trump has never once been productive and actively makes the world a worse place. Do better.
If you think it would be "horse manure" for the Trump administration to want to remove protections for endangered species, and to cite fake de-extinction as a justification—
Okay, why are you assuming that the Trump administration knows this is fake and is cynically removing endangered species (I presume you are hinting at 'so rich developers can bulldoze habitats and make YUGE profits which they funnel back to GOP politicians!' rationale for that).
All the flippin' "trust the science, bro" messaging from the liberal/left side about many issues, so if a company comes along with "look, we are using SCIENCE to do this thing", why not believe them?
I think we shouldn't believe Colossal, as it stands, but if you're just skimming something then "hey, extinct species can be brought back, that's great!" does translate into "and so the at-risk species today are no longer so at-risk since we can clone them and reintroduce them as necessary". That does not require hand-rubbing glee at "now we can kill off all those pesky animals and make megabucks" but rather believing the hype around science and technological advancement.
I could easily see Trump, for one, being delighted by this as a sign of how great and how advanced America is, here's an American company that can bring back extinct species from the past, we're the greatest! (Personally I blame Michael Crichton and Jurassic Park for putting it out into pop culture that we can do stuff like this).
Expecting this particular administration to not take a big public dump on your particular hobbyhorse is hilarious, by the way. They took away pennies! (Yes, that's a deliberate strike against the coin collectors). I don't mind you yelling about the Trump Administration being big meanies, but you're mistaking distractions for "the Administration really cares about this."
Their writeup [2] describes solving a challenging and multifaceted problem, for which a reasonable definition of success is unclear. Obviously it's self serving, but it is a startup that needs to pull some funding by embellishing their project, and the claims seem reasonable and honest. So far so good.
So from [5] I was expecting a 10,000 words technical rebuttal, instead I got Trump Derangement Syndrome and evil mustache twirling capitalists. Is this guy really a biologist? I see nothing of substance that actually engages with Collosal's claims.
What a strange comment. He mentions Trump *once*—"The Trump administration has announced it will remove protections from many endangered species, citing Colossal's de-extinction of the dire wolf." I'm sorry you consider it a derangement syndrome to inform you of relevant facts.
(Also, as I already explained: they slightly edited a tiny handful of genes in a grey wolf, and pretended it was a dire wolf. That's what happened. I'm not sure what additional 10,000 words you're looking for to re-explain it.)
I don't care who cites their work, I care about the merit of their work.
The link in [2] explains the tradeoffs they had to navigate and their reasoning. The link in [5] does not engage with that at all and instead only implies ulterior motives, which is low key confirmation that their decisions were correct as even an expert critic cannot engage on merits.
The merits of their work are poor. Dire wolves are an entirely separate genus, not species but genus, from gray wolves. They edited 14 genes towards the dire wolf sequence. This is not the amount of change you do to get to a different species, let alone genus. Even domestic dogs are all considered one subspecies, and they've got way more genetic variety than that! Plus if you read between the lines, and look closely at the choice of genes and the fact that GRR Martin is their board's cultural advisor, it becomes pretty clear that the goal was to make the wolves look as much like the game of thrones wolves as possible. Like, why did they specifically add non-dire-wolf genes known to cause white coat color? Not because dire wolves were thought to be white, previous research suggested they may have been kind of reddish, but because the Game of Thrones wolves were white.
Yeah, it's a hard problem, and yeah, the level of success they got is actually technologically impressive. But the level of hype and marketing BS around it is unacceptable and would not be tolerated in any company making products for consumers. This is like 'Our new phone will impress 500x more girls when they see you using it' levels of accuracy.
And when I first looked into this, I thought "Ok, the dire wolf stuff is poppycock, but at least they're still doing good work cloning endangered red wolves." Nope. The wolves they cloned weren't even endangered, they were red wolf-coyote hybrids that are doing fine. You could argue that conservationists should be more accepting of hybrids as worthy of conservation, but you have to actually make that argument, not just claim that the animals _are_ red wolves.
Yup. In Jurassic Park terms: this isn't making a dinosaur with a bit of frog DNA to fill in the gaps, this is making a frog with a bit of dinosaur DNA to fill in the gaps
"Secretary of Interior Doug Burgum claimed on X that the "revival of the Dire Wolf" signaled "the advent of a thrilling new era of scientific wonder, showcasing how the concept of 'de-extinction' can serve as a bedrock for modern species conservation."
Burgum doubled down during a live-streamed town hall on April 9, 2025 saying: "If we're going to be in anguish about losing a species, now we have an opportunity to bring them back".
"Pick your favorite species and call up Colossal," he claimed, as the current federal government continues to target protections of various endangered species, often tied to misleading narratives."
So not an official government pronouncement, just social media hoo-hah. Unless you can point me to official government department website saying they are officially scrapping the endangered species list?
You know, I can't help feeling that if we had President Kamala and one of her cabinet officials making a Brat Summer announcement of "extinction can be reversed!", the same guys tut-tutting about this would be all "yay science!" even though it would be the same company and the same con artists making claims.
Re: your second speculation, our objections to Colossal's dishonesty are older than the Trump administration plan. Burgum’s recent statement, only two months ago, is the first time any of this has actually had any connection to Trump at all.
(And obviously it wouldn't be acceptable for Kamala to do this either. It's weird how quick people are to assume common sense must be partisan political signaling.)
You, and/or the guy whose article you linked, were all "the Trump administration has definitely done this thing!"
Now you've rowed back to "okay they haven't done it yet but this guy's tweet is an indication that they're gonna do it".
Wake me up when they *have* done something, not when "this guy who hates my political enemies wrote a hyperventilating article about how they're gonna do all the bad stuff".
I think there's more than enough room to give Colossal a good kicking without dragging in "and Trump will do villain stuff because of them!!!!"
re: "challenging and multifaceted problem, for which a reasonable definition of success is unclear"
If it might be possible to someday successfully clone a real dire wolf with an actual dire wolf genome—instead of slightly editing a grey wolf genome—cool. But that's not what they did and not what they tried to do.
Although there might be an area where the line of success is fuzzy (starting with a dire wolf genome and filling in a few small gaps with grey wolf DNA?), what they did here doesn't remotely approach anything like that.
And, of course, it might be that it's still too hard (e.g. having to edit way more genes, to get to >90% dire wolf genome if that's our criterion). If it's too hard then it's too hard; that wouldn't be anyone's fault, but people should be honest about it.
I agree their "dire wolves" look like regular Arctic wolves, but from their woolly mice, it seems pretty clear they're developing the tech you'd need to turn elephants into mammoths. (I consider whether "taxonomy experts" would deem that TRUE de-extinction entirely unimportant.)
I do think Colossal is going to be the Theranos of this field. I don't think we're going to get mammoths out of whatever process they're using, but the woolly mice are undeniably cute (and I'm very resistant to cute animals), so if they pivoted to "exotic pet creation" they'd do everyone a favour by stopping pretending they're going to save endangered and extinct species.
I also think they just jumped right on dire wolves because of Game of Thrones, not because of the real extinct animal. The ordinary person who ever heard of dire wolves probably has an image of the GoT beastie in mind, so "actual wolves just bigger and whiter" is sold successfully as 'this is a dire wolf'.
Yes, they’re kind of misrepresenting what they’re doing,
One possible use for that technology is filling the gap in an ecosystem left by a species going extinct; not bringing it back, but putting in a functional replacement. But that would likely raise the issue of the animal went extinct in the first place because other parts of the ecosystem it depended on are gone, too.
>(I consider whether "taxonomy experts" would deem that TRUE de-extinction entirely unimportant.)
You don't think there might be more to a mammoth than a hairy elephant?
Even if you accept a hairy elephant as a mammoth, it doesn't get us any closer to actually de-extincting other interesting species that aren't a simple aesthetic variation on a surviving one.
No, if they, or anyone else, have any specific criticism – like, head too small, tusks too straight, ears the wrong shape, or anything else – I'd be perfectly happy to concede that as a shortcoming, and then it's a matter of taste whether you think it a significant enough failing that it no longer counts as a mammoth. But the current level of criticism I've seen is on the level of "it wouldn't have the SOUL of a mammoth," and THAT is what I have contempt for.
I agree there'd still be a long way to go to sauropods, pterodactyls, etc. that don't have any living species close enough to hack together a facsimile from, but think it DOES get us closer.
I think it's less about having the "soul" of the extinct creature and more that these creatures likely differed from their non-extinct counterparts in ways that aren't externally visible or detectable from fossil evidence, such as social behavior, diet composition, effects on the ecosystem, vocalizations, brain and organ structure, etc.
And in that case labeling a modern animal genetically modified to look like an extinct animal as the extinct animal is misleading. You can't learn about actual dire wolf behavior from studying genetically modified grey wolves, and (since some people want to clone mammoths to rewild them in Siberia) you can't precisely replicate the effects on the ecosystem of a herd of woolly mammoths with a herd of genetically modified elephants, which might eat a rather different balance of vegetation and have a significantly different effect on the landscape. And people react strongly to this kind of 'deextinction' because there are suggestions we don't need to work that hard to conserve species that are currently alive, as long as we have the future option of making something that superficially looks like them.
This isn't "it doesn't have the soul of a mammoth."
This is "it doesn't have the anything-at-all of a woolly mammoth, even remotely, except for a tiny handful of gene edits, compared to the much vaster number of genetic differences between an actual woolly mammoth and an Asian elephant."
For the "direwolves," it was literally only 20 edits in 14 genes.
The mammoth plan is similar. It's not a plan to clone it from a mammoth genome; it's a plan to slightly edit a tiny handful of genes in an Asian elephant genome.
The tech is useful, I agree, including probably for fertility and genetic medicine.
I'll leave aside whether, hypothetically, the tech could ever be used to actually make a real mammoth with a real mammoth genome (instead of a slightly-edited hairy elephant). The point is that the "dire wolves" are not remotely dire wolves; and their current plan for a "mammoth" is exactly the same kind of not-remotely-a-mammoth.
If it was just their private fantasy, it would be unimportant. Or even for entertainment (I don't mind if Peter Jackson hires a hairy elephant to play a mammoth in a movie). But remember that Colossal and the Trump administration are both pretending this is *literal* de-extinction, and the Trump administration is already trying to use this as an excuse to remove protections from endangered species.
Are there any new/prospective parents who have experience with Alpha School a/b tested with home schooling? Reading the review, it really seems fantastic to a degree I didn't realize existed with off-the-shelf schooling options.
The current essay contest experience could be improved to a significant degree:
1. Everything being dumped into massive google drive documents kind of sucks; headings get weird, the pages are laggy, things have to be sorted, etc, and I'm sure it's a big pain in the butt for Scott and his assistants to copy and paste everything into the documents, especially as the number of entrants rises.
2. People submit essays for many reasons, but a major one is wanting attention/feedback/comments from a large, engaged reading population. The current rating system gives you no way to write comments for the reviews you rate, because it would be a massive headache to sort and collate comments. I had things I wrote down about each essay I rated, but didn't have anywhere to actually put them other than a separate ACX comment.
It seems like the solution is adapting an existing blogging platform or making a bespoke website for this purpose. I'm picturing a website where each essay is uploaded as a post by its writer, where anonymous comments can be left below each post, and where there's an integrated rating system on each post.
I know that someone had proposed an ELO system in a previous open thread, where instead of rating each post, you read two essays and say which one you liked better, and that over time results in an ELO score for each essay. It'd be great if that could also potentially be part of this, because right now, rating is tricky; I wasn't sure how to rate the first few essays I read, because I didn't know how they stacked up.
I'd be interested in working on this project, but have minimal web dev experience. I'm quite good at AI wrangling, but would love to see if there's anyone else who wants to work with me on making the essay contest experience better, who ideally is more experienced in website development.
That is a great idea--people who worked hard to write something basically always want feedback, and I would definitely often post feedback if there was a way for it to be attached to the specific review.
Wanted to say I like your/that person's Elo* system and agree with your rationale.
My main concern is potential abuse by raters who don't necessarily read every essay but might rate every other essay as worse than the one they liked (out of a couple they actually read, say), or as better if they happen to really dislike one, etc.
*not ELO, it's Arpad Elo's last name not an acronym, sorry to nitpick
I'm not sure if it has the functionality we need; a volunteer could certainly upload each essay as a post onto a new sub stack, but I think comments that only show up after the voting is done would be nice, and I think rating should be integrated into each post, which seems tricky to do through substack. If there's a way to accomplish those goals, though, then yeah, substack would work well.
Parts of this seem like they would be relatively easy to do by a volunteer. For example, taking a huge google drive document, and splitting it into many small ones.
For commenting the submissions, a possible solution is a new thread on ACX where someone makes a top-level comment containing the name of each entry and a link to its document, and everyone else will reply to these comments. (Scott would create the thread but keep it "only available to those who know the secret URL", send the URL to the volunteer who would create the top-level comments, and then Scott would switch the thread to "public".)
So I guess what needs to be done is for someone to volunteer (it could even be me) and propose the specific kind of help to Scott, maybe publicly in a related thread, so that other people know what is going on. Well, the next time we do reviews.
I agree that the formatting on the entries could have been better, perhaps with each entry linking to its own post, with comments disabled/hidden until after the rating period ends. I too wanted to leave comments on some of the essays I read, though they would have been of more interest to the author than to Scott.
The question of rankings is interesting - ELO doesn't seem like the right tool for this unless you assign pairs of essays randomly to readers. (You don't run a chess tournament by allowing the audience to choose who plays who.) Conventional voting theories don't work because there are too many candidates for voters to be acquainted with (so ranked choice won't work) and you allow voters to participate as often as they like. It's an interesting maths problem actually, though someone who knows this stuff better than me might come along and say this sort of problem is solved already.
I think the Elo idea is assuming that you'd indeed present people with two random reviews. It's already the case that choosing reviews randomly is encouraged, but I imagine that it'd be desirable to have some kind of option for letting people rate a review of their choosing, and I don't know how that'd work.
I'd also be tempted to try to use something like the Smith set rather than Elo, but I don't know whether sample sizes would be big enough for that to be viable. It would at least solve the problem of how to make ranked-choice voting work when most voters don't know most candidates.
It's not clear to me that the best selection of finalists consists of the most popular reviews. You risk lacking variety. (I also don't know how you would find a Smith set, or how that would help here.)
An alternative approach would be to attempt to find the 'most X' reviews for various choices of X: surprising, well written, profound, thought provoking, etc.
That seems like it flattens reviews too much to even be useful. Was the mid review mid because it's too controversial/not controversial enough, poorly structured, scope too broad/narrow etc. For book reviews the parameters are narrower so *maybe* it makes sense, but even then it's questionable imo. For open ended "review anything", the median means very little.
Substack offers very little customization, so while it would be possible to make a page for each submitted review, I'm not sure how it would help with any of the other features of the proposal. Also the process of creating the pages would still be fully manual.
First off, thanks for the work you've already done; I used pretty much exclusively the random review script to read reviews. It is up to Scott, of course, but it does seem like the current system involves a fair amount of work for him, with all the copy and pasting and compiling scoring information; I'd be happy to email him, asking if this project would have his blessing.
And if it does get his blessing, I'd be happy to help in any way that I can, though I'm unsure how much help I'll actually be.
I'll post my favorites here, to show my appreciation to the authors:
11 Poetic Forms: I'll remember what a Sestina is for the rest of my life.
Airships: Cool and steampunk. Would like to understand the technology better.
Arbitraging Several Dozen Online Casinos: in terms of potentially relevant new skills learned, this was #1. It's not really about "arbitrage" though, or am I misunderstanding that word?
Two Years Of Parenthood: Heartwarming!
Getting Over It With Bennett Foddy: Wouldn't have read it if it weren't for the random review script. I liked the connection to the author's experience as a writer.
I gave that review a poor rating because I consider it epistemically irresponsible to make such strong/spicy claims about matters of historical fact without providing more evidence than the review did. In particular, I would have liked to see some links to academic biblical scholars (not just theologians) who prefer the essay's interpretation, so that I could get a sense of how mainstream or fringe it is in the field.
I see what you mean. The historical context presented in the review seems plausible to me; things like a left hand taboo and laws governing forced labor. That said, whether Jesus or Matthew really had these things in mind remains a matter of interpretation, and I guess no amount of biblical scholarship can provide a definitive answer to that.
It's an isolated reading. I've heard the transgressive reading of those three before, but there are four statements in that section, and the last one is "give to the one who asks you, and do not turn away from the one who wants to borrow from you." I haven't seen someone offer an anti-Rome version of that one.
Kind of like how people will put together a cause-and-effect explanation of the Ten Plagues of Egypt, but then don't bother explaining that time Moses and the Egyptian wizards turned their staves into snakes.
For folks living in NYC who are voting in the Dem primary for mayor, I wanted to a) remind you that the election is tomorrow, please go out and vote! and b) strongly encourage you to rank Mamdani over Cuomo.
I disagree with both of them on policy, but of the two it seems obvious to me that Mamdani is legitimately interested in making the city a better place, while Cuomo is corrupt, self-interested, and (following the Trumpian approach to politics) is an all-together indecent person.
The Democrats need fighters. But those fighters must be as virtuous and honorable as they are aggressive, people who are so clean that nothing sticks to them so they have full ability to go after everyone else, people whose very presence on the debate stage shines a light on the hypocrisy of those who would seek to use public office for anything other than the good of the country.
A point you could have made is that Mamdani, like Bill DeBlasio, might moderate significantly in office since there’s a limited appetite for hard left policy in the city (as opposed to primary voters).
Also, the Democratic establishment really needs to start backing better candidates than these uncharismatic insiders. Just look at how much money got poured into Cuomos’s lousy campaign!
I purposely wanted to avoid policy discussion, because it's not germane to the larger point. I'm rather disappointed how many people here seem to be willing to vote for someone who is obviously a bad person if it means small concessions on their policies of choice.
I have time until November to think up some falsifiable predictions about the next four years.
From what I understand, from all of Mamdanis promises, two that he can execute on without too much interference are a) expanding rent control and b) repurposing subway station into homeless help centers.
Both of these are tragically bad in my opinion. The latter seems easier to do, so my first prediction will be that the homeless situation in NYC will begin to resemble that or Seattle or SF (before the latest elections)
A competently corrupt politician is often preferable to an enthusiastically well-meaning but inept one. The corrupt politician has an interest in not befouling his own nest; the enthusiastic idiot often won't even notice he's doing that.
If you want reform, find reformers who understand how things actually work.
> I agree this is the steelman, but implicit in the steelman is the assumption that the earnest person is also dumb or blind (because they wouldn't realize or be willing to realize that their policies aren't working right). I think it's extremely rare that someone is dumb/blind, earnest, has ideas that are so bad corruption would be better but is effective enough to actually implement those policies over the objections of others, AND is able to mount a successful campaign. Bayesian prior of all of those together is extremely low. (Trump gets almost all of these except earnest)
I don't particularly want reform. I want politicians that are going to act like adults and try and solve problems. I disagree with your use of the word "often". I would say "very very rarely"
The idiot doesn't have to be *effective* to be a menace. If their dominion is facing real and serious problems, being ineffective and blocking more effective people from solving those problems, is sufficient.
Or is the contention that NYC doesn't face any really serious problems and doesn't need capable high-level governance?
Again, I understand in theory that all of this is true, but I think you have a steep hill to climb when you're openly admitting that the counterfactual is someone who is openly corrupt -- and presumably, also effective (at being corrupt). My prior is extremely high that decency will lead to better outcomes on average than corruption. And unless you think that being a decent person is somehow highly correlated with being extremely ineffective, your prior should also be equally high.
John already climbed that hill at least partway by pointing out the corrupt person still has an incentive not to "befoul his own nest", and also has the means. He might take bribes all day, but if he's also smart enough to understand that his own welfare depends on that of his city (as does the decent incompetent), and is also skilled enough to maintain that (as does *not* the decent incompetent), then the city will likely continue to do better than if the decent incompetent is in charge.
You could argue that that still isn't enough to close the gap, but you haven't made that argument yet. Meanwhile, there are multiple examples from history that ought to lower that extremely high prior you possess.
Sorry, no, the burden of proof is on the guy who's arguing that corruption is fine. I don't feel the need to do the work of "looking at history" or whatever if other people aren't going to make that argument.
Also, at this point you're basically arguing a tautology. "The person who does better will do better"
Fine, I'm willing to concede that I can't defend putting a person with the IQ of a banana in office vs a corrupt person with the IQ of an average person.
While I broadly agree about NY political corruption, you do realise you are inviting investigation into whether Mamdani has any skeletons in the closet? Cuomo and his scandals are already known, but is Mamdani as squeaky-clean as you make out?
I'm not particularly interested in this election but simply reading the news gives me, whether I want it or not, a lot of coverage. And honestly? all the hagiography around Mamdani makes me think that if elected, he'll be an AOC type: yeah, talk the talk, be very good at holding on to their seat, maybe try a few Big Things but get slapped down or have them watered down, and settle comfortably in to being a career politician.
I don't mind a bit of the ould socialism myself, but expecting (pace Jacobin) "a promise of a fundamentally different way of doing politics for the working-class majority of a city facing an out-of-control crisis in the cost of living" makes me smile. Every hog, dog and divil running for election promises that. Then they get into power, have a look at the tangle and mess facing them that has been left behind by all the previous administrations, court cases, regulations, etc., and realise that it's going to be business as usual for them.
Even Elon and the chain saw couldn't cut through it all! I don't think New York has much wiggle room for sweeping changes to how the money works in order to make the working-class very much better off.
Speaking of that Jacobin article, really, Mamdani? Really?
"Far from “show[ing] little concern” about crime, public safety has been a prominent part of his platform. Mamdani’s ambitious plan for a Department of Community Safety pledges to build out social services and mental health care to help humanely address the root causes of crime."
Now, when was the last time I heard about a Committee of Public Safety? 😁 He might want to rethink the name there. And yeah, I'm sure a corps of social workers is going to stop violent/crazy criminals in the streets.
I mean, he's running as a socialist and campaigns with the slogan "globalize the intifada." What kind of skeletons do you imagine that could be worse?
You could try to muster some rape accusations, but those won't stick. Maybe you could find (or "find") video of him singing along to rap in high school or something, but that's weak sauce.
He doesn't seem to have been in politics very long, so what did he do before?
There's already been a little kerfuffle over "intafada". He was a rapper as a young man? Then somebody must have sample lyrics of him rapping about bitches and hoes and drugs and guns (unless he was an exceedingly wholesome rapper) 😀
Swearing about your granny? Having your (actress playing your) granny swear? Oh, my!
"How do you get an octogenarian south Asian screen icon to swear in a rap video? According to hip-hop MC Zohran Mamdani, a little context can be helpful. When he came to prepare world-famous actor and TV cook Madhur Jaffrey for her role – lip-syncing along to the words of his expletive-laden tribute to his maternal grandma, Nani – he had a little explaining to do.
Mamdani, who performs under the name Mr Cardamom, can remember Jaffrey asking him: “In the lyrics, who is this ‘mother fucker’ for?” he says. “Fuck this person, fuck that person; now whose ‘fuck’ is this?”
The guy may indeed be squeaky-clean, but if someone wants to dig deep enough and far back enough, it may be possible to find something that could have a bad interpretation put on it.
> you do realise you are inviting investigation into whether Mamdani has any skeletons in the closet
good?
> he'll be an AOC type: yeah, talk the talk, be very good at holding on to their seat, maybe try a few Big Things but get slapped down or have them watered down, and settle comfortably in to being a career politician
I'm pretty neutral on AOC, but, like, is your take that AOC is very corrupt or something? "a career politician" did she do something bad? or are you just annoyed about the aesthetics?
Net net, I purposely did *not* focus on either Cuomo's or Mamdani's stated policy positions, because the point I want to make is that being a decent person is a necessary precondition to talking about policy at all. I still think that's true, digs about the French Revolution aside.
In a race between someone who is a corrupt petty sex abuser and someone with aesthetics you don't particularly like, I personally would go with the mid aesthetics
Not the aesthetics, it's the opportunism. Turns up with a 'Tax the Rich' designer dress at the Met Gala. Yeah, really getting down with the proles there, Sandy!
Though I do have to respect that she's managed to retain her seat and get her career well-established, unlike the rest of the Squad who have either faded or got bogged down in sectarian controversy. She's dug herself in for the long haul as representative for that Bronx/Queens district, successfully turfing out the old Democratic party guard there.
I would recommend asking the right questions. First of all in the original Marxian analysis, the problem is with the system itself, not capitalists as persons. Now, libertarian economists tend to see the market also something like a machine, except that they see it as a good one. So I guess both agree capitalists do not have much personal responsibility, the market is a machine, if there is money on the table, someone will take it.
OTOH socialists dictators DO have a lot of personal responsibility, because this was not a machine like that. Stalin could have easily decided to not create the GULAG.
I think that several people have this take, and it always seemed like handwavy vibes-based feelings grounded in GOP scaremongering more than anything based an actual political outcomes. If you have any actual examples of this (and I mean relevant examples, not, like, "This cambodian general called himself a socialist in Khmer Rouge") I'd be open to seeing it.
But, also, fine, rank any of the other candidates, just not Cuomo.
Even though you're being snarky here (which is fair, you're responding to a snarky comment) I think this is obviously a silly take under the hood. Progressives are famously supposed to be the people who *dont* understand market dynamics. But here you argue against market dynamics while tarring progressives for...understanding them? Very backwards.
As an aside, dinging vectro for being a 'progressive' like its a big scary dirty word really just says more about you than anything else. Ooo vectro is a progressive...what are you gonna call him next, vegetarian?
increasingly of the opinion that people don't actually know what policies mamdani ran on or have any idea who the major players are in the big cities in the US
What *about* San Fran? Are you saying Mamdani is like Lurie, a moderate? Or like Breed, an establishment Dem? What policies do you think are the same between these people? Have you lived or visited SF over the last 10 years? What is your understanding of the problems that SF faces?
I think Michelle Wu or Lisa Hidalgo are equally reasonable comps who have done extremely well, so its increasingly unclear that Johnson isn't just a fluke or a particularly bad candidate/mayor
I dunno enough about Chicago tbh. I've heard the comparison a few times and agree that it's possibly a good comparison. Do you know more / have an informed opinion? Is Johnson unpopular because of policies? Gaffes? Scandal?
Thanks. I think just reading this, with nothing else to go on, it seems like the biggest problem with Johnson is not the ideology, but rather being too ideological. He seems like a crusader more than a leader, someone who does not know how to make trade offs because making a trade off of any kind would be impure in some way. The left absolutely has these people; left activists even more so. But I think this is true of any political affiliation or party -- you have the people who are shrewd leaders who actually know how to do things, and the people who are ideological warriors who are best when they are pointed at certain things by people in the first group. One of the big failures of MAGA is putting all of the ideological warriors in charge. The Dems have historically not done this in public office (Bernie, for eg, famously got knifed by the Dem party). It's likely that there is more appetite than ever to elect warrior types into office on the Dem side, and I agree that would be terrible, as it seems to be with Johnson.
But it's not at all clear to me that people like Mamdani (or even AOC or Sanders for that matter) are so ideologically uncompromising that they do not know how to hire good people, read the room, and be outcomes driven. This matters because on the ground, the policies of Johnson and Mamdani _don't_ seem that similar. Yes, there's a general tax and spend attitude, and I agree that is similar across the two of them, but I think the similarities more or less end there. It was Cuomo who was more embedded with the Unions, and more likely to cut favorable deals with them in exchange for their backing. And it was Cuomo who would regularly refuse to create a reasonable budget for the city and state, mostly to fund vanity projects or things that were obviously unworkable. In these things, Cuomo and Johnson are quite alike.
I think there is a way to do tax and spend that doesn't cause the city to burn. The question is whether Mamdani is too much of an activist to be able to thread that line. And that frankly remains unclear to me. But I think he's talking about the right things, and the ways in which the city can make things easier for constituents by *removing* red tape (eg here: https://x.com/ZohranKMamdani/status/1878853557111414795)
what about him? Why do you think this is a reasonable comparison, beyond Chavez also calling himself a socialist? who are the corrupt capitalists you're comparing against in Venezuela that would have done better?
If you want to point to Venezuela without any further explanation, I think it's equally fair for a socialist to point to the nordics with equally little explanation.
> This includes a comprehensive welfare state and multi-level collective bargaining[2] based on the economic foundations of social corporatism,[3][4] and a commitment to private ownership within a market-based mixed economy[5] – with Norway being a partial exception due to a large number of state-owned enterprises and state ownership in publicly listed firms.
> While many countries have been categorized as social democratic, the Nordic countries have been the only ones to be constantly categorized as such. In a review by Emanuele Ferragina and Martin Seeleib-Kaiser of works about the different models of welfare states, apart from Belgium and the Netherlands, categorized as "medium-high socialism", the Scandinavian countries analyzed (Denmark, Norway, and Sweden) were the only ones to be categorized by sociologist Gøsta Esping-Andersen as "high socialism", which is defined as socialist attributes and values (equality and universalism) and the social democratic model, which is characterized by "a high level of decommodification and a low degree of stratification.
Social corporatism, mixed economy, and norwegian state ownership are often seen as 'socialist'. Also I think generally welfare state stuff is seen as socialist too in the US. I think that's kinda dumb -- imo healthcare is obviously its own thing and I think universal healthcare in the states would immediately solve so many problems its honestly silly we havent implemented it yet -- but it is true that about half the country thinks obamacare is socialist (and half those think its communist!)
For better or worse, I think that 'socialist' in America has come to mean 'strong safety net and high taxes'. Which is more or less what mamdani is proposing, like I haven't seen him saying stuff like "we should nationalize oil"
> There are enough corrupt capitalists in south america to choose from. They are and were bad, not as bad as Chavez.
I think this is kinda weasely. You made the claim that Chavez was a good comparison to Mamdani! Can you back it up more than this?
Based on your comment history I dont believe your verdict is informed by his decency. You prefer the socialist to the moderate because hes the socialist (and being coy about policy disagreements) Whether you think thats fair or not, I think readers have high bs sensitivity to decency based arguments that are just laundering political preferences. See GOP preference for decency during clinton/bush era vs now.
In my experience socialists are very prone to equate agreeing with them politically = being a decent person. And historically socialist regimes have treated political disagreement as one of the most heinous crimes (and recruited even family members as informants.) I think it’s an intellectually and morally bankrupt political philosophy, but it seems to stick around in institutions of higher education and in the minds of their students.
It’s interesting that we seem to have learned only the lesson of fascism, and not the lesson of communism, from the 20th century. No one ever says we need to try fascism again, Hitler just did it wrong - but I hear it frequently about communism from seemingly earnest people who are quick to dismiss Stalin and Mao as aberrations.
Sure, it’s people on both sides, and it sucks when any of them do it.
But hey, I disagree with Scott politically and I still think he’s a great writer, thoughtful, intelligent and I avidly read anything he puts out. I also enjoy Freddie DeBoer even though he is “far left.” He’s honest and that’s more important.
Yes, but I wasn't actually making a "both sides" point here: You said it was only "socialists," and my point is that it's not just the fringe, but a mainstream position among Democrats as well.
Well, Scott still lets me post here, but it's true he's one of the rare classical liberals who still strongly defends free speech. I wouldn't last 2 seconds on Reddit or BlueSky. And yes, it's definitely worse on the left than the right - as much as liberals like to complain about Elon, they are welcome to do so and post their hottest takes right on Twitter/X (although many have left because the mean rightists keep replying to them and Elon never seems to do anything about that)
a) claiming to have special insight into the motivations of a person you don't know beyond their stated motivations is...bold, to say the least. Very bayesian of you. (Claiming to have special insight into the motivations of "readers" -- also very bold)
b) my comment history is motivated by a disdain and disgust of the Trump administration and everything it stands for. That you think 'hatred of Trump == socialism' says more about you than it does me.
c) vote for literally anyone other than Cuomo, it doesn't have to be mamdani he's just the person who's second. If Lander or Adams were second, they would be better choices. Literally *anyone* would be a better choice than the multi-time sex pest with a penchant for corruption.
Why do you feel the need to defend the multi-time sex pest with a penchant for corruption? "Our policies at any cost" is fucking stupid, sorry. I care about voting for decent people, because at the end of the day my own policy preferences are likely to be misinformed, the policy positions of candidates on the campaign trail are *definitely* misinformed, and when push comes to shove I want my public leaders to put their constituents over themselves. Cuomo is so selfish he's willing to split the vote just to stay relevant in politics. It's disgusting, as is defending him because you're stuck in the cold war.
In case it matters, I'm pretty economically centrist. More taxes on large corporations and the wealthy. Universal health care. Way fewer regulations, especially on construction. No taxes for small businesses.
I think most of Mamdani's policy positions are mixed to bad. I don't like rent freezes or rent control, and I don't get the groceries thing. Cuomo's policies are shockingly similar, though no one ever really reports on this. Doesn't matter though -- the main point is that decency is a prerequisite to even being considered a candidate. Mamdani isn't going to be cowed or bribed by Trump because, and I can't stress this enough, *he's a decent human being*. If Cuomo thinks being pals with Trump will benefit him, he will turn immediately -- just like Adams did before him. No thanks. I want to actually like my leaders.
"I dont believe you" is something ppl are allowed to say and doesnt require them to believe they have special insights into your motivations. And if I dont believe you, others probably dont either, because im not so special.
I offer no defense of Cuomo, and the things you say about him are probably true. But sorry, I dont believe you, and also I think mamdani would be bad (see I can just say that outright)
I agree, and further suspect this stance that one's policy preferences shouldn't outweigh "decency" would be abandoned as soon as the differences become significant enough*, probably by arguing that anyone who held those positions ipso facto lacked decency, and so voting for the corrupt bully with your policies over his honest opponent becomes the better choice after all.
*The opposite position on ALL of gun rights, abortion, affirmative action, LGBTQ+, and immigration, say.
If the best argument you have to advance your point is that you literally are allowed to say it, you might as well just concede there's nothing here. If you want to make claims about my political positions as a means to disparage me, at least have the decency to back it up with evidence. You claim to have ample evidence with my post history after all. Or is "socialist" just a bad word you're using to tar by association?
"Based on your past actions,I suspect your motives"
"You dont know my motives as well as I do"
"So what, the whole point is I think you are being dishonest"
It is true, I havent taken the time to prove you are being dishonest about your motives. And I wont because, my goal is not to take you down.
I thought you might feel like my skepticism was unfair, I said as much in my first post, I am just giving you my honest feedback. I cant take your argument touting a leftist above a moderate (fine I wont say socialist) on grounds other than their leftism serious because I see you commenting like leftists do. There are people who criticize Trumps terribleness without giving a leftist impression. You arent one of them, and it affects how others receive your arguments.
I'm sorry you don't like my aesthetics. I'm not a leftist, but you're welcome to your opinion. Death of the author, consider my position despite your dislike of the leftist that you think I am.
Disclaimer: I am also a socialist, and a Boston-born Red Sox fan who has never enjoyed visiting New York.
That said, the "moderate" (I don't agree with that self-appellation; this faction is socially liberal but seems every bit as committed as the radical right to fucking over the less fortunate) preference for Cuomo speaks incredibly poorly of that faction in NYC. He's a former governor who resigned in disgrace and is best known for killing the elderly in nursing homes during the Covid pandemic... and changing laws to give unsafe nursing homes immunity from lawsuits. You can't even defend him on idpol grounds, given his history of sexual harassment and homophobia (vote for Cuomo, not the homo!) or party loyalty (he worked to hand the state senate to Republicans).
What's the case for him - he's rich and famous?
NYC has ranked choice voting and there are other non-socialists on the ballot.
The OP should probably have chosen a different moderate candidate to recommend placing above Cuomo in order to appeal to socialist skeptic voters with a sense of decency, as yes its ranked choice. Instead the OP makes the case for Mamdani, hence my skepticism.
It doesn’t matter who else you put above Cuomo - it will likely come down to Mamdani and Cuomo and anything else you do is likely primarily aesthetic and expressive.
Last time garcia was 3rd first round down 11pts from adams. She finished 0.8 behind Adams. The ranked choices can accumulate and I would not take it as a given the top 2 first rounders are the last 2.
Polls haven’t just looked at top choices - there have been some that try to measure head-to-head support of several candidate pairs to see who are likely to be the final two. Last time, polls showed Garcia and Adams and several others as possible finalists, I recall.
20 million dead Americans say Cuomo did the right thing!
What? They didn't all die?
I guess covid19 wasn't all it was cracked up to be.
Cuomo got too popular, so like Spitzer, he had to go. Unlike Spitzer, Cuomo declined to fall on his own sword. The democrats are in a bad place, where they believe that anyone getting "too popular" (read "bigger than the current TopGuy") needs to go.
Grade B people hire Grade C people.Grade C people hire Grade D people (so as not to be outshone). And suddenly, Democrats!
We know from prior experience that Cuomo is evil and self interested, so nearly any breathing person is better, as we don't have proof that they would act in a way that as bad as possible for the city. Is that not obvious, or is proving those foolish leftists wrong more important than acting in self interest?
It is hard to take seriously anyone who uses terms like "evil" to describe someone.
> is proving those foolish leftists wrong more important than acting in self interest?
If I wanted to prove foolish leftists wrong, I would vote a foolish leftist into office and watch what happens. As for self interest, I own a home in NYC, so it is in my interest to vote for the guy whose policies are least likely to limit the rise in housing costs,which again would be a foolish leftist. But I want to do the opposite which means I might have to vote for Eric Adam's in the general election, regardless of who wins the Dem primary.
sorry if I'm being dense, there's too many double negatives in your initial post and I'm not sure what exactly you're saying. You're a homeowner, therefore you want housing prices to increase, therefore...?
Evil and self-interested in a politician spells Nixon to me. As he was a smart guy, he did stuff that was mostly good for America. That your description of "why not to vote for him" dovetails so nicely with Nixon doesn't speak well to your discernment.
I'm not going to justify your low effort comment with a proper response but you sound like this;
Jews for Hitler is perfectly reasonable! Sure, he might be evil and self interested but so was Nero! He brought untold status to Rome by acting in an self interested manor, so therefore I'll look for that quality in all future candidates. This non-intuitive conclusion is extremely satisfying! I am very smart! I'm so smart that I value looking smart over anything else, including my own self interested. I prefer to look out for the interest of capital and billionaires to signal to other Very Smart people just how smart and clued in I am!
And you were saying? Unfortunately, this was quicker to cite than actual Jews in the Nazi Party.
I do not in fact believe that the "Road to Zero" is going to lead anywhere good. At this time, neither is solar power nor wind power (without batteries, the grid needed burns more natural gas, not less).
See, you're making an actual argument here: "Cuomo is Emanuel come again" -- and that's making an argument, as a reason not to vote for him. It's no longer "He's evil" but "His constituency wants things that I do not want."
> I'm so smart that I value looking smart over anything else, including my own self interested
The wikipedia link you cite is not exactly beating the allegations. Somehow, you've found yourself arguing things like Jewish support of Hitler means that people should vote for Cuomo. I think you should take a long look in the mirror. "Owning the libs", or "winning internet points", or whatever your actual ideology is, isn't worth corrupting your soul.
On the flip side, if you're openly of the opinion that jewish support for hitler was reasonable, I'm just going to block you.
o undoubtedly. We don't know whether Mamdani will do a good job or not. But we know that Cuomo won't, so the choice seems obvious.
Also, I'd rather an inexperienced but decent human who can pretty easily learn on the job, than a corrupt 70 year old who is going to wield power for his own petty bullshit while lining his own pockets. Again, the choice seems obvious.
Not only don't we know that, your article made no such claim. Indeed, it ridiculed those who would make their choice based on which candidate would pursue better policies.
One of the many points of the article is that you cannot trust what Cuomo says or does, because even though he says he will pursue abc policy he's actually just going to do whatever is best for himself.
If you think Cuomo is going to do a good job, make the case. You seem to admit that he's a corrupt and indecent person, so I feel the bar for defending him should be pretty high.
> It ridiculed those who would make their choice based on which candidate would pursue better policies
Right now, I think your stated position of "I'm voting for the corrupt and altogether indecent guy because I want housing prices to go up" is a pretty horrible take. There's a suite of ethical reasons why its bad, but honestly I think its bad on your own stated merits -- your housing price isn't going to go up as fast if the mayor does a bad job, and Cuomo is going to do a bad job because he's corrupt and petty.
Or, put another way, "I think the guy who's going to siphon funds for pet projects and political retribution is going to make my community better" is a wild take.
>One of the many points of the article is that you cannot trust what Cuomo says or does, because even though he says he will pursue abc policy he's actually just going to do whatever is best for himself
That is a claim, but whether it is true is another matter.
>You seem to admit that he's a corrupt and indecent person,
I did not admit that. My point is that policy matters.
>Right now, I think your stated position of "I'm voting for the corrupt and altogether indecent guy because I want housing prices to go up
I actually said the exact opposite.
>"I think the guy who's going to siphon funds for pet projects and political retribution is going to make my community better" is a wild take
1. I don't accept your premise.
2. Even granting your premise, as someone else pointed out re Nixon, your conclusion does not necessarily follow from your premise.
The steelman of the case for Cuomo over Mamdani is that the corrupt 70 year old who wants petty bullshit while lining his pockets might still be preferable to someone who will earnestly attempt a lot of policies that will cause massive problems.
I don’t live in New York and thus haven’t thought hard about what the candidates are promising to attempt, so I don’t know how to evaluate whether Mamdani’s promises are that bad, or whether Cuomo also has some equally bad promises. But sometimes, the corrupt person is better than the true believer.
I agree this is the steelman, but implicit in the steelman is the assumption that the earnest person is also dumb or blind (because they wouldn't realize or be willing to realize that their policies aren't working right). I think it's extremely rare that someone is dumb/blind, earnest, has ideas that are so bad corruption would be better but is effective enough to actually implement those policies over the objections of others, AND is able to mount a successful campaign. Bayesian prior of all of those together is extremely low. (Trump gets almost all of these except earnest)
> so they have full ability to go after everyone else
See, THIS is actually a reasonable argument, that decency (or the appearance of decency, which is the same thing in this business) is a good weapon for attacking Trump/national Republicans with, and that translates to longer-term federal policy gains worth the shorter term local policy hits.
That is NOT the argument you make in the essay. There, you seem to espouse a doctrine divorced from reason, leaning fully into voting as a kind of religious ritual to affirm sacred values instead of a bloodless transaction granting power (and the concomitant opportunity for abuses thereof) to a candidate in exchange for him enacting your preferred policy.
Yes, but you leave out the absurd bits that are the main thrust of your essay (summarized well in the title, so kudos for that) in the parts you excerpted here, and which change the meaning entirely.
Well, it looks like it's too late to organize review reviews under one comment, so I'll just throw them up as a top comment. Although I didn't read many in the first place.
The Bo Burnham review had bad chronology problems. There were several points during the review that I thought we had finished with the main album and were jumping up to focus on a follow-up album, only to realize that the jump ahead had been an unmarked aside and that we were still only partway through the first album. If all those jump-aheads had been in parentheses (or not there at all) the review would have been stronger.
Phoenix Theatre had too much backstory at the beginning, and also monstrous paragraphs that are ugly to the eyes; the Word Count tells me paragraph 2 has nearly 400 words. I don't remember how far I got into it but I don't think I got to anything actually about Phoenix Theatre.
Skibidi Toilet seemed like a really solid review, but I was only a page in before I realized I hated the concept of the show and actively want to never think about it. It's like seeing a review on a painting of smeared shit. No amount of effort will make it pleasant.
Hi, I wrote the review of Effective altruism/rationalism, and really want to hear what people think of it. No reason is required to not be selected, I assume all finalists are great of course.
But I hoped it would start a real debate about how to institutionalise the movements in a less illiberal and elitist way. Instead, I haven't even seen it mentioned in a comment. I want to understand why:
1. It landed as obvious (we've noticed the skulls). In this case I want to hear whether everyone knows this is a big problem, eg sick of the discussion, or if the reason is that you think it is not actually a problem.
2. It landed as wrong/misguided. Please argue
3. It landed as rude/arrogant, poorly written or not the right kind of text. Feel free to give feedback
To me (obviously, that's why I wrote it) what I point to really is a problem on the ground, so I want to hear people's arguments
I also gave it a rather poor rating. I should first say that I have never really looked into the EA movement. Basically all I know about it is via Scott, so I may have wrong impressions.
But the impression is that the points you raised are not really new. Like, the tension between mainstream/free thinker. Or I would rather describe it as tension between the inside/outside perspective. How much should you trust your own reasoning versus the conclusion of other people, if those do not agree?
I think this is a problem that at least Scott is very aware about, and I have no indication that the EA movement is not aware of it either. The issue is that this is a really, really hard problem. I face it myself a lot of time when I read the news, and I don't have a good solution to it. You push for going more into mainstream/outside view. But I don't think that generically this is very helpful advice, because you are using your own inside perspective here (which is an outsider perspective for EA). You should first consider that your inside perspective may be wrong, and that the EA perspective is right, and I didn't have the feeling that you discussed a lot why you are so sure that you are right and they are wrong. In fact, I would assume that for some concrete issues you are right, for others the EA people are right, and that it is a really, really hard problem to tell which is which. Personally, I am totally not concerned about shrimps, but am I right about that? I don't know! And I totally agree that x-risk assessment is very speculative. I think the EA community is also painfully aware of that, this seems to be one the central issues they discuss endlessly. But is the right course of action then to dismiss x-risks because it is speculative? You seem to say yes, and I see the arguments for yes, but I also see good arguments for no, and overall I find it very unclear what we should do in such a case.
So, in summary my criticism would be the following. Mind that I only focus on the negative points here. Your review was not as bad as it may sound from just listing the criticism.
- I have the impression that the EA community is totally aware of the problem (balancing inside/outside view) you tell them about. But they understand better than you that it is a hard problem, and just pushing generally for "go more outside" isn't very helpful.
- You tell them to go further in direction A. From my perspective, this is just a single data point, your own opinion, and I didn't find that you acknowledge this limitation. Phrasing it nasty, your main argument for direction A was "I think it is so", and that is a weak argument to me.
- This also made the review a bit arrogant. I didn't mind this so much, but I could imagine that people with more skin in the game might.
That review annoyed me and I gave it a poor rating, which I felt a little bit bad about as I'm a card-carrying EA and so potentially biased. I think it's cool that you're interested in feedback though. (For what it's worth, most reviews didn't get any comments, since the thread wasn't really set up to facilitate that, so I wouldn't read much into it.)
I think my three most clearly articulable objections are (1) most of these criticisms could be leveled against any set of political/social opinions at all, making it unclear why EA is being singled out; (2) I can't figure out what the kind of movement you'd like would actually be in favor of (you seem to support some kind of "mainstream" synthesis that I don't think exists); and (3) it makes a whole lot of different critiques in a somewhat scattershot fashion that makes it hard to tell which ones constitute the true rejection. There are also some smaller things that I think the review gets wrong or incoherent, as well as a few "yes we've noticed the skulls" points (where the issue is the lack of a preferable alternative) and a few things that are just kind of grating in a way that marshals people's biases against being persuaded.
I started elaborating on these, but unfortunately I find longform writing kind of difficult, and wanted to get something posted rather than just procrastinate forever on this, so for now I'm just posting the short version. If you're interested in further elaboration I can try to do that, especially if you either ask specific questions or want to do it on Discord or something (because I write more fluently there).
Again, thanks for asking about this, I think it's a good thing to do and not common enough.
Thanks! See my other reply. (1) No, see piecemeal/Utopian, (2) I outlined changes at the end, (3) i'm not rejecting EA, I'm saying it should be implemented in a less illiberal and individualistic way if it wants to achieve its stated goals.
These points must have been poorly formulated for all of the commenters here to think I was doing some hit piece on EA with no alternative. I get some of it, but still not really. Send me as DM here if you want, happy to discuss further
> Where (other review) was too personal, this was too abstract. One of the charges laid against EA was: “If they have a job in a non-EA organisation, they care more about leveraging the job to further agreed-upon EA policy goals than to do the job itself well.” No examples were given.
> Then there was a bunch of stuff about how hard it is to have good epistemics. I agree that epistemics can be difficult, but I was looking for an argument that said “and here are some examples of EA people making these specific mistakes” and the only one I saw was Sam Bankman-Fried, and I’m not sure if I’d count him as EA.
Trying to give a more general comment, though it's been a few weeks:
It felt like this was an attempt at armchair-theorist reasoning from first principles about why the EA movement might have problems, but if we accept this reasoning as correct then we have a fully general argument for why nobody should ever try to make the world better in any way. I think that making the world better is important, and I think it would be very hard to move me from that position without lots of practical evidence.
If an argument wanted to convince me that EA was bad on net, I think the argument would have to start: "Here are the N largest and most important EA ventures, and here is evidence for why each of them caused more harm than good."
I agree it should have had more data/examples. This is partly lack of time, partly not wanting to call out smaller actors, and partly not obvious how to document it. I ended up relying on my personal experience, counting on people recognising behaviour they have also seen (see other comment in this thread).
As a journalist, when I meet EAs that know I share their background, they tend to assume I will strategically write whatever is good for EA's concrete policy goals, colluding with them rather than doing good journalism.
Reading these comments it also seems like I underexplained some points. I'm not making a generalised argument, and I'm not saying EA is a net bad (??). I'm arguing for Popper's piecemeal social engineering rather than Utopian social engineering.
EA is great for global health and other areas with tangible benefits. My criticism is specific to politics, illiberal attitudes and institutional design in the movement. When you cheat in politics, you legitimise cheating for your opponents and we get a race to totalitarianism. This can be institutionally addressed along the lines I described.
I understand the call for evidence, but how would the world look if I was right? I would say in that world the movement would remain a subculture, get a cultish shady reputation with bad media coverage and a scandal or three as it moved further into politics, rather than becoming more well-known and gradually accepted. These are data points, albeit post-fact.
Sorry this for rambling, didn't have time to write this properly. I agree that a full argument would dive more into the specific areas but this is what I had time for. Theory is also a thing
Did you think about the readership of the essay competition when you were writing? You're not going to win many 10/10's while dismantling a person's ideology for them.
I read and reviewed your essay. I think there's an element of (1) there, nothing you said struck me as especially insightful. I thought there was an irony in your main critique being that the EA movement mostly produces online critiques rather than engaging in the world to improve it. And you... produced an online critique?
That discussion is still all about how to further EA causes. "EAG exists to make the world a better place, rather than serve the EA community or make EAs happy" says the Utopian social engineer, breaking some eggs for the omelette.
I don't agree with the third point. You have to go inside to tell people to go outside.
Also I'm not saying EAs don't engage with the world to improve it, in a sense I say they overdo that part. I'm saying EA is too private in its epistemology, but too optimising in its practice. The combination is dangerous
Your surprise that your critique didn't land well comes across as a little bit naive. Have you tried giving someone negative feedback in the real world?
3 is definitely in play; there's a lot of flippant language, even as early as when you say wind turbines are allowed "to shoot up", as if they're just weeds sprouting out of the ground. Very much has the feel of someone wanting to tweak the nose of the movement. The new moniker, the absense of evidentiary hyperlinks in your list of observations despite them being prevalent before and after, it has the tone of a hit piece. I'm dropping off before page 3.
I didn't want to tweak any noses, but definitely felt (perhaps too strongly?) that it is important and people don't realise it while writing the review.
Interesting about the observations, they were not meant as proof. I don't think there's any data for this, but please tell me if there is (ACX survey?).
Maybe I should have made my thinking explicit: The reader, presumably familiar with the movement, can validate the observations for themselves. If they don't ring true, the situation is less bad. If they do ring true, this is a problem, and the review is trying to show why. Given my personal experience, I would expect the latter.
It rings as focused solely on news-worthy parts of the movement. If you found a soup kitchen, you get an article for that. If it runs for fives years and serves lots of people, maybe you get another article if nothing else interesting happens in town. If one of your workers bonks a homeless child on the noggin with the serving spoon and hurls insults at them, congrats, you've made every headline.
Anyway, the movement was heavily influenced by Yudkowski's writing, and Yudkowski's writing has a lot of cultishness attractors (see: entirety of MIRI), so it's not really a surprise that there's cultish subgroups doing stupid stuff. As long as GiveWell, 80,000 hours, and other impactful groups stay sane, we should be fine. But I don't know what you expect me to do about the fact that other people come up with insane policy proposals and call themselves EA.
As for not trusting the mainstream, I just posted a comment in this open thread discussing a fact that was incorrectly reported by pretty much every journalist that touched the subject, because one of them talked to a scientist who said something vague and it was misinterpreted. Lots of people here are probably experts on some niche topic and have had similar experiences. If we see journalists getting stuff wrong in areas we know a lot about, why should we trust them for the rest?
As someone who's only familiar with the movement through what Scott and the commentariat post here, my accepting the points requires my trusting the narrator. The hostility of the writing style requires much more upfront evidence than a more neutral wording could get away with, as I would trust a more neutral piece to bring up its supporting evidence further in.
Wondering if I should go ahead and publish my review (which didn't make it into Scott's post) and thank the people calling it out in these comments, or keep to anonymity on the off chance the review gets elevated into a last-minute honorable mention position, as it did previously.
Either way, I'm really pleased with the quality of the reviews as a whole, and that there are loads of people in these comments calling out reviews other than these 10+ finalists and honorable mentions as their favorites. Obviously I'm sad mine didn't make it, but to see even one person name it as their favorite still warms my heart. See you next year!
"Strikes against Russian industrial capacity continue. The Nevinnomyssky Azot fertilizer plant in Stavropol Krai, one of the top producers of ammonia and nitrogen fertilizers in the country, halted production after drone attacks. Another attack was charted in Orenberg, 1500km from Ukraine, and likely home to strategic missile base. In an effort to blunt the drone campaign, authorities cut mobile services to vast swathes of Russian territory. It ends up being a map of strategic priorities. Oops."
That's clever, but assuming that internet was shut down in whole regions, knowing that your target is in Sakha (area 3+ million square km) doesn't give you a lot of additional info.
My plan had been to review Fresh Fruit for Rotting Vegetables by Dead Kennedys, but other projects took enough of my time that once I got to the writing stage I quickly deduced I wouldn't be able to finish by the deadline. Perhaps I will do something on a related subject for a later contest.
Is your interest in it musically or lyrically focused? I find the second record "Plastic Surgery Disasters" to be more interesting in lyrical content as it contains numerous examples of fringe theories and sentiments that underwent political realignment in the past 40 years, and I like some of the more complex compositions, but I certainly understand the more punk purist appeal of the debut record.
FFfRV is more meaningful to me, personally, and its balance of edgy/dark songs like I Kill Children with political ones like Holiday in Cambodia would have also been something to consider and contrast. Due to its general reification there's also more written material on how it came to be.
Would love to see Bukele promoted to finalist! Thought it was really interesting, with analysis by someone living in the country of interest and well written. Have discussed it with several people in real life since reading it. Definitely keen to read and discuss further.
The Musk Algorithm was also great distillation of real world experience.
I really enjoyed it, learned a lot, and have shared it with others. The Miracle of El Salvador is an example to the whole world of how fast change can happen, even if it's failed for decades
I did think, however, that it played up the Dictator angle in a hype-y way, possibly for drama and readability. At the end, it says clearly that the author intends to just "monitor the situation" and maybe there's no dictator worries here. But ALONG THE WAY, it uses literary devices to make things sound much more 1-sided.
For example, it uses a dramatic and sudden "end of paragraph/section" after saying something about how Bukele immediately had a political opponent tried for corruption. Only significantly later in the article does it then add, "Oh, yeah, the opponent was literally on camera doing crazy corruption, so this was the reasonable and obvious legal/justice thing for anyone to do."
It would have been less of an emotional roller coaster to say this right next to the "mic drop moment" about Bukele going after the opponent, but more honest.
I lived in El Salvador for 2 years as a Mormon missionary during the highest homicide rate years, and am meeting/talking with lots of Salvadoranean people (eg at Kimbara bar) in San Francisco. As we face AI risk and hopelessness, I hope the example of El Salvador's Miracle can give us hope that outlandish claims about rapid results can actually be real and viable.
(I don't claim to know how to discern if the "extremely tough on crime" approach is worth the immense improvement in safety (not in detail), but a 91% approval rating makes me think I should cede judgment to those who have lived there.)
I really blew it on the review. I didn't read the contest note clearly and started work on an actual book review. DQ! I'll finish it up when I have time and get it in early next year.
Tried to play catch up with a review of Milgram's obedience experiment, in the context of popular fictional character archetypes, but spring is not a good time for writing in my line of work (Apple Grower). I'll finish and post that sometime, but I was off to a good start.
Pre-contest note, in anticipation of what I thought would be a book-review contest, I began a review of Tristram Shandy. When I discovered was a non-book review contest I tried to adapt my review into an argument that Tristram Shandy isn’t really a book. But it was too complex and I gave up the attempt. This is my review of that attempt.
It's propaganda, as taught (I imagine so is the Milgram experiment). If we wanted to laugh at the author and say "this story is so hilariously wrong that it's not even worth considering "good science fiction" as it doesn't get humans right," then we'd do that instead. Lord of the Flies mischaracterizes children in such a way that we're taught that "babysitting services" (like our current government schools) are essential to teaching children how to not behave like ... well, deer. Deer murder cats and dogs for no reason other than they're aggravating. Deer push their fawns (babies) into dangerous circumstances, before Mom gets anywhere near. Deer overpopulate without regard to population pressure (which rabbits don't do, mind). Etc. (yes, I may be biased. Our deer are exceptionally stupid here, in no small part because they eat toxic plants).
Milgram himself ran several replications of it, mostly because he had already built the lab. Some factors and results were stronger than others. For example, proximity of the authority figure was a big deal, people delivered fewer shocks when they were told to over a phone.
Punchline is this study started a wave of reform in Psych testing, with new ethics codes that protected subjects from the kind of stress and emotional harm that were evident in his work. So, replicated? can't say.
here's the link to his contemporaneous documentary film
In 2012, Australian psychologist Gina Perry investigated Milgram's data and writings and concluded that Milgram had manipulated the results, and that there was a "troubling mismatch between (published) descriptions of the experiment and evidence of what actually transpired." She wrote that "only half of the people who undertook the experiment fully believed it was real and of those, 66% disobeyed the experimenter".[26][27] She described her findings as "an unexpected outcome" that "leaves social psychology in a difficult situation."[28]
Pulling this from wiki. I suspect this was designed (and paid for) to provide people psychological cover for "just following orders" -- "everyone else does it" sounds good on paper.
Uhm, this is a pretty misleading excerpt. Why don't you also cite the paragraph directly below it?
"In a book review critical of Gina Perry's findings, Nestar Russell and John Picard take issue with Perry for not mentioning that "there have been well over a score, not just several, replications or slight variations on Milgram's basic experimental procedure, and these have been performed in many different countries, several different settings and using different types of victims. And most, although certainly not all of these experiments have tended to lend weight to Milgram's original findings"
Also, I think there are a lot of reports of participants of the original experiment which also confirm it beyond the pure publication. My understanding is that this is pretty well-established. Other than, say, the Stanford Prison experiment or the Robbers Cave experiment.
Good contest, there were some really thought-provoking entries in the ones I read - and I got to very few of the ones that ended up being finalists, so I still have plenty more to look forward to.
I wrote the Skibidi Toilet review. I had a ton of fun writing it, glad it got some positive reception and informed some people about one corner of the vast and weird landscape of modern storytelling. :) I'll post a polished-up version to my own blog soon. (Update: It is here https://eukaryotewritesblog.substack.com/p/skibidi-toilet-review )
To the commenters who vocalized a disdain for learning anything about the topic or even being vaguely compelled to think about it: Hey, your time is your own and you should spend it as you see fit. Maybe next year, you can write a review of rank cowardice so the rest of us can decide if it's right for us.
After reading some comments, I read your review, and it was good enough that I actually watched Skibidi Toilet from the playlist provided. The series is interesting, especially with the background provided by your review., though it's kind of hard to follow because it seems to follow its own physical laws, like getting your head cut off doesn't kill you, except when it does.
You did explain that the technology is never explained, and I found that a bit irritating, too. In the 4th episode, somehow the people in the restaurant get turned into toilet people by no visible means. In 41, the toilets are compelled to commit suicide, but this is apparently never again used against them.
I thought your review was excellent. I thought you had done the most out of anyone to present the worst possible topic in the best possible light, but Skibidi Toilet was always going to be a tough sell.
Eagerly awaiting the next time someone disparagingly mentions Skibidi Toilet in a discussion about The Youth of Today and I can "Well ackshually" them
Yours was the first review I read (bc scrolling and... "there's a review of TOILETS?!?!"). It was an interesting choice, and I thought your review gave a lot of context to an extremely weird meme.
> Maybe next year, you can write a review of rank cowardice so the rest of us can decide if it's right for us.
Iran has had their militarily capacity greatly degraded in recent weeks/months. Setting their nuclear program aside, we know that their air defense and missile launching capabilities have been largely curtailed. However, if somebody thinks this is a great time for regime change coming from within--which is the only place it should come from, self-determination being near sacrosanct to me--I have this thought. The security and police apparatus that is so effective at keeping the Iranian people from succeeding at their protests is roughly intact.
So, any notion that we are suddenly in good position for a popular counter-revolution in Iran had better buckle-up, because it will be a rough ride. Sure is fun and easy to bomb targets--and eventually that means people--but intervening to help a local/national uprising does not appear to be on anyone's menu. Presidents that say "go ahead, we have your back" have not been credible, and so should be taken with salt.
Inviting such an action through a preemptive leadership decapitation action would be a black swan sitting on a black box. no-one knows what will happen squared. Not a great comp, but no one seems very happy about Libya post-Ghaddaffi (mostly due to Wagner/Russian interference, IMO), but Iran is a much higher base-rate competence country that Libya was, so who knows.
Currently at 37% chance Khamenei isn't supreme leader by the end of the year, 18% chance that the Iranian regime falls by the end of the year. Since the former is implied by the latter, would it be reasonable to assume that there's a fairly high chance that if Khamenei is gone then the government will be overthrown? Or am I mixing up causality somehow?
I don't think this mathematically follows. There could be a world with a 19% chance that Khamenei is killed by an airstrike (or dies naturally), and conditionally on that, a 100% chance that the regime survives. Plus a chance of 81% that Khamenei survives the year, and conditionally on that a roughly 22% chance that the regime falls. All this is mathematically consistent with the numbers you gave, but then there is zero chance that the regime falls if Khamenei dies.
All this is purely mathematically, of course, and not a model of the real world. But we can say (taking the numbers at face value) that the probability of the regime falling if Khamenei dies is AT MOST ~50%. Because there is at least a 19% chance of "Khamenei dead, regime in place" and at most an 18% chance of "Khamenei dead, regime falls".
Clarification about the "self-determination" you consider sacrosanct: would you object to the US arming a few dissident groups inside Iran with vast amounts of guns, tanks, fighter jets, and missiles to do with as they please?
Sure, but I'm not suggesting this operation be covert or deniable in any way: this would simply be kickstarting a civil war, and using the rebels to fight a proxy war. What is the part you think isn't credible?
(Also, I'm not actually suggesting this pointless procedure of adding extra steps to the regime change operation; I don't give a shit about "self-determination" in the sense of preferring local tyrants over foreign ones.)
> However, if somebody thinks this is a great time for regime change coming from within--which is the only place it should come from, self-determination being near sacrosanct to me--
If it comes from without, how is that any less self-determination than what they have now? Iran isn't a real democracy. Anyone living their is having their life determined by someone else as it is.
Yeah, this is a good point. Reminds me of Iraq -- of all the negative thinks one might say about the invasion, "but imperialism" is the weakest one, since the result was an increase in self-determination (which is not the usual result of "imperialism")
Okay, let me address your 3 points in reverse order. 3: Well, all societies have either rules or anarchy, so yes, other people have a lot to say about how you live. Not much for me to disagree with there. 2: Iran not a democracy? Not generally thought of as one, and yet...they do have elections that have actual political consequence. Yes, the Mullah branch controls who is allowed to run, but within that scheme there are still extremists and moderates--the current PM being the most moderate candidate in the last election. They also have women with the right to vote since 1963! Not bad. On the whole, I rate them as slightly less democratic than Israel, which would have a considerably higher grade if so many of their residents weren't denied citizenship/suffrage. But I digress.
3: They got pretty pissed off in Tehran when a woman was murdered in police custody. They went to the streets and made a very big deal of it, and were beaten, jailed and even killed for their trouble. These people want something better, and I hope they can find their way to it, although right now might not be the ideal time. I will say that in the photos of Iran after the bombings, of the people I see only ~half the women have their hair covered, the rest don't like that.
I once heard on some news show a dissident in Syria--this was 10+ya--say that Assad would fall when Syrians made him fall, and any other interference was quite possibly disastrously unhelpful. His preference has aged well.
"that response isn't the defense you seem to think it is."
Not sure just what it is that you think I am defending. What do you think I am defending?
I'm not really coming at this from an ideological stance. If you think I am, run it by my and I can confirm or deny. If you want to know what I think--about Iran, American Power, the Iraq war, The Afghanistan war--you can ask; you will get a better answer than if you guess.
Oh, okay. That sounds about right, actually. In terms of functionality, Israel is far superior. Other than some ongoing assaults on the independence of the judiciary, it certainly serves as a working, stable democratic government. Iranian civil society is utterly hobbled by the big system-small system dichotomy of the Appointed mullahs and the elected representatives. The elected ones don't really have a fair shot--though they do often try! More's the pity.
I would accept 'former Al Quaida' and even that is a bit of a stretch, you make the alliances you can in a terrible war. AQ ia not shy, they would be glad to claim they were running the gov. of Syria. That has not happened--tho not a dispositive on its own--and they are not running the country as such under a different name. A great gov? I dunno, we will see. They have chased Russia out and degraded their ability to make trouble, but that doesn't mean he isn't a semi-secret AQ agent. Trump recognition far from the worst thing he has done, policy-wise
That's probably why Israel has been targeting low to mid level police and police leadership recently. I understand they have been intentionally assassinating the equivalent of the local police captains, though I haven't seen a lot of detail on it.
I thought this was an excellent take-down of Curtis Yarvin's recent allegation that the Allies starved hundreds of thousands of German prisoners of war to death in 1945:
Yarvin's original X thread was done in excellent Yarvin form, replete with rhetorical wink-winks and "if you read any real history, this wouldn't be a surprise" vibes. I admit to hoping it was complete BS, but trying to hold my judgement for a while, and I'm glad someone did what seems to be a thorough and even-handed job of refuting it.
I agree that Yarvin is, oh gee, let’s just call him a bozo for the sake of brevity, but arctotherium goes on to carp about ‘the elevation of the Tulsa race riot to a massacre’ in his conclusion by citing an article in The Unz Review which is edited by Ron Unz, who per Wikipedia is an antisemitic holocaust denier and a white nationalist so I won’t take arctotherium’s word as gospel.
I would not trust an article by Ron Unz to tell me much of anything, not because of his politics or morals, but because he is very careless with getting the facts right and very inclined to reach for a cool conspiracy story.
My impression is that while Ron Unz is personally a crank, his stable of writers includes a weird mix of fellow cranks and people actually interesting to read who care about the truth (or at least used to - I haven't paid much attention there since both Audacious Epigone and Anatoly Karlin left). I think that Steve Sailer is the only person I can think of who is both an Unz Review contributer and a regular commenter at ACX, and hope someone wouldn't dismiss what he had to say just on the basis that he also wrote there.
I had a similar reaction. It made me curious about the rest of the blog, since that choice was so wildly at odds with the well-argued and well-sourced article that preceded it.
So I clicked on the first article in the archive that caught my eye, which turned out to be nothing short of appalling. It was a self-righteous bout of verbal diarrhea about how immigration is ruining Canada and how the U.S. ought to step in to coerce those misguided Canucks onto a better path. As a Canadian-American who has lived in both countries (and fully intends never to go back to the U.S.), I can now confidently say that I would no piss on the author if he were on fire.
Yeah, that was my reaction when I saw his reference to Tulsa. I clicked on the article, but no way was I going to read it with all my BS alarms going off. Thank you for going the extra mile, it smelled that way to me.
This is bad. The author is bad. To be fair, I *think* Yarvin is trolling.
We know that former Nazis were treated incredibly poorly and anyone familiar with sources from the period would be aware of it. My personal familiarity of this comes from the diaries of John Rabe, the Nazi who is credited of with personally saving 250k Chinese during the Rape of Nanking. You can read his wiki here (1), watch a trailer for the movie about his life here with Steve Buscemi (2), and buy his diaries under the title "The Good Man of Nanking", which I would recommend.
We know he was basically starved to death. We know this from his diaries but we also know his situation was so dire that in 1948 the mayor of Nanjing traveled through Switzerland to Germany and the people of Nanjing sent him food until his death. Allow me to repeat, a ~64 year-old humanitarian was essentially starved so badly that the people of Nanjing China, during the Civil War and Maoist period, were sending him food. If that was happening to an old humanitarian Nazi bureaucrat...yeah, I can imagine some pretty bad stuff occurring to actual soldiers.
Yarvin is bad here because I *think* he's basically trolling and this is Trump's Hatian cats all over again, where there probably was not a million person execution of captured German soldiers but their living standards and treatment was probably pretty horrific and now we're all talking about this. Yarvin *probably* wants this because it serves his purposes for the Skeletor of the 20th century to be Communists, not Nazis.
This response is worse. It presents no actual evidence except official statistics provided by the very people accused of genocide/abuse, cribbed from a book written by some other guy. Give Yarvin this, I believe he's actually read and interacted with direct sources and now he's spinning/trolling to get clicks and influence. This bozo just cribbed a book someone else wrote about it with, I don't want to say no evidence, but certainly not persuasive evidence. You just...if someone accuses the Soviets or the Maoists or the Japanese of genocide, it is not persuasive to show official Soviet or Maoist or Japanese figures declaring that they totally didn't commit a genocide. The bar is low but you have to actually cross it.
Please don't share stuff like this in the future. It was dumb and it made me angry.
John Rabe was not in any of the prisoner camps. He was starving as a civilian, and as the author made very clear a *lot* of people were starving and going without food in Europe in 1948. John Rabe going hungry as a civilian in Berlin is not evidence that hundreds of thousands of Germans were starved to death in Allied prisoner camps.
The author makes a lot of good points. If this happened, why doesn't anybody claim that it happened until 1989? Why did the West German commission whose sole job was to track down every German that died during or after the war not find hundreds of thousands of starved prisoners? Where is the evidence that such a thing occured? Yarvin doesn't seem to present any.
Lots of details like this get classified and only become public decades later. Operation Keelhaul is another good example of this kind of thing, which seems like it only became known in the 1970s.
If an old bureaucrat and family were being starved, why wouldn't that count as evidence that former Nazi soldiers were starved? Like...the Americans aren't going to starve the former SS guys but they will starve some guy's daughter? From memory I think one of them was 10 and another was a teenager at the time but that's old memories from the diaries.
It's not uncommon for events like the Holodomor or Dresden bombings to remain controversial, unproven, and generally uninvestigated outside of extremely niche communities for decades.
Again, what probably happened is that a lot of former Nazis got treated extremely badly after the war. How bad? Who knows? Also, not to be too much of a jerk, who cares? It would have hardly been the worst crime of the period, it probably wouldn't have cracked the top 10 even if they had killed a million. Yarvin is *probably* overselling it for political ends and this guy is underselling it using some of the worst logic ever.
The old bureaucrat and his family wern't "being starved", they were just starving. Nobody was keeping them from getting food, there just wasn't enough food to go around and everyone who wasn't a German got priority on food aid. Which is the author acknowledges and nobody disputes it. This widespread hunger is not the same as the claim that hundreds of thousands of German prisoners starved to death and the Allies covered it up.
The Holodomor was controversial, but people were reporting on it as early as 1932. Similarly, the Associated Press was reporting on the destruction of Dresden within weeks of the actual bombing, and there was a lot of people call it a war crime at the time. Yet we have no contemporary claims of mass deaths by starvation in Allied prisoner camps, not for many decades.
I agree that what probably happened is a lot of former Nazis got treated extremely badly after the war. We know this because we have a ton of evidence to support that fact: personal testimonies, newspaper articles, written government policies, etc. We don't have any of that for Yarvin's claim.
No, he was being starved. It's very explicitly noted in his diaries and it so shocked the people of Nanjing, a people who had literally lived through the Rape of Nanjing, that they sent him food to keep him and his family alive for two years during their own civil war. We have, to the best of my knowledge, multiple accounts of his starvation by independent sources and no contrary accounts. Nor was the mechanism confusing, he and his family were explicitly and covertly barred from any employment until his death, with contemporary accounts noting them living on wild seeds.
Nor was the Holodomor widely known. It was denied in the pages of the New York Times as it was occurring (1) and this was endorsed by FDR at the time for fairly obvious realpolique reasons.
How badly the post-war Nazis were treated and how many died is a source of ongoing controversy, especially since it's so politically charged. As an example, the best estimate of the number of civilian casualties during the Dresden bombings has varied from 135k to 250k to 25k (2) in my lifetime alone. A lot of that is political/culture war stuff but a lot of it is that it's...just actually really tough to estimate.
Without any specific insight into this particular controversy, the idea that out of 3 million Nazi soldiers, that only a few thousand died in US custody while contemporary reports of civilians at the time being in dramatically more dire straights is...not super credible and would be kinda horrific if we actually did treat Nazi soldiers better than German civilians. Conversely, a couple hundred thousand dead is...plausible but highly unlikely. But if the Dresden bombing casualties can range from 135k to 25k, then it makes sense that the Nazi....whatever this is, the Rheinwiesenlager could vary from 30k to 300k.
Having said that, for all his trolling, at least Yarvin is familiar with the direct sources and providing 3 party reports. The article originally shared doesn't, it's bad for reasons well elaborated on already.
Rabe was officially de-nazified on 3 June 1946, but grateful Nanjing residents sent him food packets starting in 1948, presumably meaning he was still starving at that time. At least from 1946 to 1948, it wasn't the Allies preventing him from getting a job.
We seem to have different definitions of "being starved" (being banned from working is not the same thing at all in my eye as deliberately starving someone) so we'll just have to agree to disagree on that. The point remains that one person being kept from working and as a result starving is not evidence that hundreds of thousands of German prisoners starved to death in Allied camps and the Allies covered it up. It is evidence that former members of the Nazi party went through bad times after the war, which was never in dispute.
The Holodomor was not widely known but it was *known at all*, known to the point that the NYT put out articles denying that it was happening. Where is the equivalent for the mass starvation of prisoners? Where's the newspaper article saying that such claims are exaggerated or untrue? We don't have a thing for this until the 1980s.
We probably did treat Nazi soldiers better than German civilians when it came to food because the US Army was responsible for feeding their prisoners but not that responsible for feeding random civilians. For one thing all the prisoners were concentrated in camps where you could send food shipments to and distribute them, while the civilians were scattered all over the country.
I read it too, and there are reasons for concern with this author (like, defending residential camps, calling Tulsa a 'race riot?). But the most compelling stats he drew from came from the West German government of the 60's and 70's.
That's what it was called for about a century after the event. Which way do you prefer it be revised: up towards "genocide", or down towards "mostly-peaceful protest"?
Shankar and Melvin, you are reading me wrong, which is probably the result of my bad writing. An author above, during a lovely takedown of Curtis Yarvin, dropped a reference to an article 'debunking' the events that happened in Tulsa. To wit: he was changing the events from a massacre to a 'race riot,' in a decidedly both-sidesy sort of way. I didn't feel like giving it more of my time, attention, and I guess I already failed there, but I don't like people trying to airbrush a massacre away. Ditto with the Residential Schools. They happened and it was bad that they happened, and its okay to say that.
Shankar and Melvin, you are reading me wrong, which is probably the result of my bad writing. An author above, during a lovely takedown of Curtis Yarvin, dropped a reference to an article 'debunking' the events that happened in Tulsa. To wit: he was changing the events from a massacre to a 'race riot,' in a decidedly both-sidesy sort of way. I didn't feel like giving it more of my time, attention, and I guess I already failed there, but I don't like people trying to airbrush a massacre away. Ditto with the Residential Schools. They happened and it was bad that they happened, and its okay to say that.
"Race Riots" isn't an inherently both-sidesy term. Most race riots in history have tended to be entirely one-race sorts of affairs.
The problem with rebranding the "riot" as a "massacre" (which is an extremely recent phenomenon, wikipedia changed it on Feb 7 2020 https://en.wikipedia.org/wiki/Talk:Tulsa_race_massacre) is that nothing about it sounds massacre-ish.
The defining aspects of a massacre are its speed and one-sidedness, and this particular event doesn't match that. You could call it a very bad riot or a very small battle, but not a massacre.
Yes, the West German government's retrospective in the 60's and 70's is more trustworthy than the American's. Still:
#1 Independent 3rd party accounts are better, and Yarvin provides those in his...sigh...Twitter thread here (1). An account the original author ignores and seems to get confused with a report from the Army Medical Department. (the Andersonville Prison reference)
#2 Postwar Germany was politically...not the most unbiased of locations. Holodomor denial from the Ukranian Soviet Socialist Republic is better than the USSR but...ya know, not exactly unbiased.
As an aside, I tried to see if he referenced the red cross thing anywhere, I don't think he does, but he does claim that Pope Benedict was in one of these camps and...no? Certainly not my area of expertise, correct me where I'm wrong, but Benedict's Wikipedia shows him in Neu-Ulm and Fliegerhorst Bad Aibling (2) but those are not in the list of 19 Rheinwiesenlager camps Yarvin is discussing (3).
To be clear, if I have not, I have not heard of this particular claim before. It seems similar to the bombing of Dresden or the Holodomor. Old events that are highly charged, highly ambiguous, and where it's likely something bad happened but how bad and how intentional it was is quite vague. Both events also that kinda lurked in academic corners until they became fuel for the culture war. I doubt Yarvin's specific claims but the abuse of Nazi and German people pops in up in every contemporary account I'm familiar with and also...just makes too much sense. That's just what victorious armies do. What I do resent and get angy about is just bad argumentation.
Sounds like the guy is out of ideas and pretty desperate to get attention.
EDIT: Or maybe, if this is intentional, he decided to target a dumber audience than previously. Now the ideas are less abstract and more dramatic, and there is less need for plausibility.
I wrote the Deathbed Ballads review. It didn't make the cut, but least one person seems to have liked it, which I appreciate. Last year, I reviewed Martin Luther's "Bondage of the Will", which looks like it narrowly missed the Honorable Mentions, and I got some good feedback on it after the fact which I still plan on using to make a revised version to publish elsewhere.
Some other topics I considered reviewing this year:
"Taxman" by The Beatles. I wrote a paper in college analyzing this song and tracing out specific references in it to elements of the UK tax system in the 1960s. I think it would only need minor reworking, but I thought the same of my college paper on Bondage of the Will last year and had wound up redoing it from scratch. I decided against it because it was much more of an analysis than a review.
"Identity", a YouTube video where the creator, Abigail Thorn, comes out as a trans woman to her audience after already being established as a moderately big-name YouTuber before starting her transition. I watched it a few months into my own transition and found it very moving and applicable. I think it's very well-made and has a number of subtle details I'd be interested in talking about. I decided against reviewing it mostly because I didn't think I could do so anonymously: I have been very active in comments discussions here about trans issues and have cited Identity more than once in those discussions.
Analysing songs from the American Civil War as a corpus for hints at how soldiers thought about the conduct and motivations of the war. I had originally chosen this as my topic and started writing it, but abandoned and pivoted to what I hoped would be a lighter topic because it was very quickly turning out to be too verbose and research-heavy. Not sure how successful I was (gestures vaguely at my 14000 word actual submission). It was also shaping up to be too much an essay and not enough a review.
One other topic I won't share publicly. The reason I didn't review it this year is because it overlaps substantially with a book review I want to do in a future year and I think it works better if I review the book than later than if I review one major topic discussed in the book now.
I did read it now. To be honest, I found it hard to connect to it. A good part of that might be myself. I am not a music person. I don't listen to radio, and never have done. I thought I might try it because I knew one of the songs (France, and that actually in many versions). But without knowing the others, it was just a very long list of parts that I couldn't relate to.
I found the topic also a bit less, well, relevant. For Luther, I was already going into the review with the premise that he is an important person that I would like to understand better. For the songs, I would have needed to be convinced that they are important, and there wasn't so much convincing done. Perhaps they really are not so relevant, and perhaps they are, but in the latter case I don't know about it.
I still think it was objectively well-written. You made the most of the topic, and I do buy you conclusion with two origins and found the table interesting. So to me the problem was not in how the review was done, but rather what the topic was.
Taxman. 1 for you 19 for me. Pretty easy to understand complaint. Even for a kid who had yet to pay any taxes in 1966. One of George’s pre-metaphysical songs. No sitar in that one.
Can anyone here who supports the Israeli and American war against Iran explain to me your reasoning? I'm kind of baffled by the support for it. AFAIK the official rationale of Israel/America is that Iran is building a nuclear weapon and so they're engaging in self-defense to stop them, but U.S intelligence says they're not building nukes https://apnews.com/article/us-iran-nuclear-intelligence-b506d130e474c00f6bd653d3d5a8d31a . In particular I would like to know
1. Whether you think Iran is actually building a nuclear weapon
2. Whether you think the war against Iran is legal under international law
3. Whether you think diplomatic alternatives to war have been adequately explored
4. Whether you think regime change is viable and desirable
Basically, with all of Iran's proxies in the region declawed at the moment, this was the best time to act with impunity and Israel took it. Killing their scientists and demolishing much of their nuclear research infrastructure hopefully has set them back far enough that they won't be a player. The US intelligence leak to which you're referring was being treated skeptically. The Mossad clearly believed otherwise, as their attacks had no other clear objective and did not seem tailored towards regime change since they left the non-military authorities mostly untouched. The whole thing seems to have been crafted to clear the way for the action the US took.
I don't think imposing regime change from the outside is viable, or necessary, or desirable. I didn't really care about Iran in any way other than preventing them from having a nuclear weapons program. If I had to prefer one nation be the regional hegemon down there, I'd certainly prefer the Saudis over the Iran/Qatar axis, and I'd prefer secular monarchies over Islamist theocracies, but as long as they've been de-fanged it's not really something I'd want to sacrifice US war material and lives to get involved with and that would likely be counterproductive anyhow.
Big fan. This has been the cause of a very very large 'pro trump' update for me.
I was broadly in favor for pro-democracy reasons (I felt america had gotten captured by bureaucracy and it was very difficult for voters to effectuate change, and here was a guy who very publicly said what he would do, was non-ideological, and would try to get it done) but disliked trump in particular for usual reasons, and here we have some actual leadership / good results. I claim to have supported it before 'it all worked out clearly better than we could have reasonably hoped, which maybe it hasn't yet but probably has.'
1. I think Iran was trying to get as close as it could to being able to make a nuclear weapon, between a week and year from it. They had 60% enriched uranium that we know of. Given their history of sponsoring terrorism against us and our allies, already a very big reason to not want this. Their strategy was approximately 'get to a week within a nuke, while having many many missiles, use that as deterrence while funding terrorism.' Given how much Israel weakened them by destroying their proxies in syria/lebanon/yemen, I think it was clearly a good idea.
2. I...don't care about international law, at all. International law says if someone declares credibly 'I will gain power, and once I do I will kill you' you have to wait for them to get sufficiently powerful that they feel ready to go to war, to try to stop that. It's horrible tool in this context. In theory, great for minimizing casualties in 'reasonable wars between equal powers who might have wars in the future and it's a repeated game', horrible for religious crazy regimes with nukes or nuclear ambitions.
3. I think diplomatic alternatives to war were sufficiently explored to discover that the difference between what the Iranians would agree to, and the expected result of bombing their stuff, was extremely large and in America's favor. Theocratic dictatorships are extremely bad at modeling the will/capacity of their enemies. There's just...very little you can do in this situation. Iran for religious reasons wants to defeat their enemies and in so doing redeem Islam by proving they have gods favor. It's hard to negotiate with that unless they are extremely practical after you prove 'they are not going to succeed.' So you need to 'succeed' sometimes.
4. Uh, I think regime change of the form 'the 80% of domestic civilians, and the 98% of iranian diaspora that hate the evil regime rise up and install a secular democracy' would be great. I think that's unlikely to happen because the regime is so good at repression. It's also not our problem. Ultimately, people need to liberate themselves. It's also very chaotic.
We did what we wanted to do: we blew up their nuclear program and a substantial portion of their intercontinental ballistic missile program. That's good enough for now.
The most popular beer in Iran is BEER. Now, how's that for repression? There's also underground dating scenes where middle class men and women get together and dance in Tehran. How's that for repression?
(Bottom line: learn a little more about a country before you assume it's Communist china -- do as we say or you're re-educated. Iran is a very delicate country, full of many twists and turns. It's fascinating, in its own way).
How many women have been tortured and killed by the police forces for showing their hair or skin in public? How many non violent protesters have been imprisoned.
Yes, I agree head scarves are not currently being enforced on the streets of tehran. I assure you what you have stated is not new information to me. Also China has earned the trust and respect of a majority of its people, Iran has done the complete opposite.
There were (circa a few years ago,haven't checked recently) 12 different ways out of Iran. Some were conservative some were liberal. If you wanted out of the theocratic regime, you could get out. Pretty much anyone could. I'm going to say that women getting tortured and killed is, of course, bad. Is it worse than in the rest of the Muslim world, where a woman cannot get married without her father's permission, and cannot get divorced without her husband's? (In case you missed it, this is prostitution colored Muslim -- and it's still fairer than marriage in the rest of the Islamic world).
China has earned the trust and respect of a majority of its people? I've seen illegals in America -- folks that China has deemed murderable on our soil. They did not look happy, they looked very very afraid. They are in danger of their lives at all times. I can say more about China, if you want.
This is a weird conversation. You're not telling me anything I don't know. I've been to China and am reasonably familiar with its people. I don't even understand what factual point you imagine I might disagree on with you.
1. Does it matter if they're building one this very moment? (This is not to discount the evidence provided by others that they are.) The mere fact that a theocratic regime has taken regular steps towards nukes is terrifying enough and surely unsustainable.
2. Iran's *existence* isn't legal under international law. Doesn't the latter include a whole bunch of things about human rights and freedom of religion? Or did I dream that? Is funding interstate terrorism legal under international law? Is having an official regime-defining ideological goal of wiping another country off the map legal under international law? If any of these things aren't legal under international law, is there a world police that's eventually going to actually enforce this "law" and why they are taking so long?
Does anyone actually utter the phrase "international law" with a straight face?
3. I mean, Iran spent several decades *not* being bombed and...did it stop funding terrorism? Did it stop talking about wiping Israel off the map? I don't even understand the coherence of this question. If you think there is any chance of negotiating with Iran now, it can surely only be because it's under direct military threat. Which is...the very thing you're objecting to.
4. I suggest you look an Iranian woman, an Iranian Christian, a person who's been in an Iranian prison, in the eye and tell then you have doubts about whether regime change is desirable. I dare you.
Viable is an entirely different question. I'm certainly open to arguments about whether regime change is viable or possible or counterproductive. But (and I have the same attitude to restrospective analysis of Iraq) such questions are in an *entirely different universe* to suggestions that even if regime change were possible, viable and low-risk or zero-risk it would still, somehow, not be right. The former is sensible and logical. The latter is total moral bankruptcy and moral cowardice of the highest order.
1. Whether you think Iran is actually building a nuclear weapon
> I don't know. I give it more than 10%, and think that the rest of their international policy is enough at least to remove their "deontological protection".
2. Whether you think the war against Iran is legal under international law
> I care about ethics. I would love to live in a world were "international law" is really a thing - but there is neither a leviathan nor established expectation of reciprocation
3. Whether you think diplomatic alternatives to war have been adequately explored
> No idea. Low prior that a diplomatic alternative is possible given the positions of the parties.
> 4. Whether you think regime change is viable and desirable
I think it is very possible that the regime would fall given the right encouragement, though not something to count on. And that it is overall desirable - both for its own sake and to make the best alternative to agreement worse
I hence strongly support atacks against tyranny-preserving organizations there as long as there are other reasons to fight. It wouldn't be my way goal though, because I want to avoid ugly fight to the (regime) death, and won't pay the price of ground invasion
on (3), "No idea" about diplomatic alternatives to war seems incredibly cavalier to me.
A prior, do you not think that exploring diplomatic alternatives to war is important? What if an extremely feasible, easy, low-cost, highly likely to succeed diplomatic route exists?
I agree that it is important. I have no idea not because i don't care but because I lack access to the information, as do you. I only know the public positions and actions of the parties involved
1) Yes, I think they intended to build nuclear bombs. 60% enriched U-235 makes no sense for a power reactor (typically 5% enriched), and looks only like a step to bomb-grade U-235. And the IAEA said Iran has broken nonproliferation agreements.
2) I'm fairly disgusted by international legalisms. Maybe commercial international law has some value.
3) Iran was great at diplomatic delaying tactics. I view these as delaying tactics while they enriched to bomb grade.
4) (writing from the USA) I don't think _we_ should be trying to change Iran's regime. The current regime has been chanting "Death to America" for 40 years, but, even if we _had_ the power to oust their current regime, (a) We didn't do well with that in Iraq or Afghanistan - maybe we've lost the knack for it (b) Trading the devil you know for the devil you don't know often yields buyers' remorse. It isn't as if we could just kill Khamenei and have Iran miraculously transform into a Persian Switzerland.
The very first sentence you linked to is "A U.S. intelligence assessment says Iran is not pursuing nuclear weapons at the moment but has ramped up activities that could help it develop them."
Can anybody explain to me what the difference is? As I understand it there are no other uses to which the super-pure uranium they are acknowledged to be producing can plausibly be put to. How exactly is that "not pursuing nuclear weapons"?
I've read Democrats whining that there was no rush, that Iran is months away from having a nuclear weapon. To which I just drop my jaw and say, "Months!?!"
One obvious difference is that it may well have been possible to negotiate them off that course previously. It certainly won't be now. If the U.S. is going to treat them as if they are nuclearizing one way or the other, then they're certainly better off rushing for a nuclear deterrent as fast as they can.
The difference is that Iran is using the uranium as a bargaining chip: "if you don't drop these sanctions, I'll enrich to 70%! Then to 75%!" By inching ever closer to the weapons threshold, they hope to extract concessions from the West. If they actually build a bomb, they gain a military deterrent but lose a diplomatic bargaining chip.
I think that one habit people really need to get out of is calling governments by the names of their countries, and vice versa. We need to distinguish between "Iran", which is a huge country of 90 million people, and "The Iranian Government", which is a random bunch of unelected thugs that happens to control a capital city.
The morality here is not in question -- when you have an unelected government ruling over a country, it's moral for absolutely anybody, no matter how foreign or humble, to kill any element of that government.
"I think that one habit people really need to get out of is calling governments by the names of their countries, and vice versa. "
But find this to be absolutely unhinged:
"The morality here is not in question -- when you have an unelected government ruling over a country, it's moral for absolutely anybody, no matter how foreign or humble, to kill any element of that government."
Like, setting aside the fact that "any element of that government" could just as well refer to people paving roads or distributing food as secret police. Even if we're just talking about the top brass, this is unhinged. Make no mistake, I have very little sympathy for those who abuse their power, regardless of whether they got that power through heredity, military force, business acumen or winning elections. But even a very modest knowledge of history should thoroughly disabuse one of the notion that "just kill the bad people" is a workable recipe for improving the lives of the average person.
Iran has kinda sorta elections..with theocratic oversight. Also, it's pretty difficult to remove leaders without going through ordinary citizens to get to them.
>I think that one habit people really need to get out of is calling governments by the names of their countries, and vice versa.
What I don't get is why the US joined in. They had Israel already doing their dirty work for them. They just needed to supply weapons, which they already do and is not against international law, and intelligence, I'm not sure about the law for that but it can be done discreetly. I don't think that would grant them any sort of moral high ground, but it seems like the pragmatic thing to do.
I have seen the theory that Trump ordered the strikes to take the wind out of Israel’s sails. It is easier to call for a ceasefire when one side has already achieved one of its war goals.
I do think this verges on the “5D chess” school of giving Trump too much credit, and would prefer a simpler explanation. Maybe the intent was to inflict damage to the nuclear program regardless of how long or how effectively Israel keeps fighting?
Alternately, some general showed up to the daily brief with a contingency he wrote in 2016 and said the right things to get the President’s buy-in. I don’t know.
Because the US has deep bunker busting weapons and Israel doesn't (and they can't just hand them over, since they require B2s to deploy which only the US has). For the US it was a cheap free option to finish off buried iranian nuclear sites for free.
Where peer means “definitely not peer because Iran isn’t a peer country”. China is probably not quaking in its boots, and it has nuclear bombs. As does North Korea. The way to make sure you aren’t bombed for having WMD is to have WMD.
> China is probably not quaking in its boots, and it has nuclear bombs
They should be at least a little concerned. This is a great tech demo proving that yes, the B-2 is still untouchable, and the US is still quite capable of putting half a dozen of them above your head without you noticing.
They should be afraid of the B-2, and they should be even more afraid of the B-21, which will be like the B-2 but far more numerous.
I mean it proves that the B52 is not touchable by whatever Iran has but that’s it. The Chinese can launch ICBMs anyway and Iran cannot so I can’t see that happening.
does it prove that? The Iranian air defense system had been pretty thoroughly dismantled. The Israelis had supposedly been flying over Iran for ten days (including in non-stealthy F15s) without loss. So B2s getting in and out without loss also seems to tell us very little about the untouchability of the B2, when going up against a competent and intact air defense network.
The B2 may in fact be untouchable, but Iran being unable to touch it seems like very weak evidence for the proposition.
The American Military Might got squashed in the Ukraine (on a logistics basis, the Russians make cheaper arms and get a better kill per dollar ratio). One doesn't need to have them quaking in their boots to get them to re-evaluate how likely they are to get squashed by America.
The Germans famously got a much better kill per dollar ratio than the Allies in both world wars. They lost (at least in part) because their enemies had more dollars. In WW2 this became a deliberate policy of the British and then of the USA - prepared to be wasteful of high quality materiel rather than risk high casualties.
Can you cite me some sources on that? (I did a brief googling and didn't come up with anything) My understanding was that the Nazis favored "quality over quantity" in terms of materiel and that the Americans favored the opposite.
> The American Military Might got squashed in the Ukraine (on a logistics basis, the Russians make cheaper arms and get a better kill per dollar ratio).
That's a ridiculous conclusion, unless by "American Military Might" you mean "a half-assed fraction of America's obsolete equipment that was just a few years away from the scrapyard, slowly trickled down over years".
HIMARS appears to be actively under development (software wise) in the past decade (2017 to be specific).
Aside from that, what I mean is what we have for sale to other countries. Obviously they don't get our best stuff. This, same as the first Iraq war, is a tech demo. A bloody, useless, tech demo.
To add more details, if China wants to conquer Taiwan militarily, China for sure needs the US to not intervene. Aircraft carriers based in the ocean east of Taiwan and land based Typhon missile systems in Phillipines and/or Okinawa would prevent China from attacking with land troops across the Taiwan strait.
A good video on the Typhon's possible role against China.
Yeah, as far as I can see, one big lesson from the Ukraine war is that current technology favours defence over attack. Crossing 100km+ of water in a slow-moving and easily tracked boat while being actively attacked by missiles, artillery, planes, submarines and drones just doesn't seem possible.
I don't think this is a consideration, beyond telling your press office to replace references to "international law" with "rules-based international order" in the statements when you decide to flout it (or at least what you're saying it is when you're citing it against Russia or something).
Does it? TACO isn't about Trump's willingness to send crazy signals, it's about his follow-through. Dropping a handful of bombs mostly just reinforces the same picture everybody already had of him: he's inconsistent, unreliable, and prone to lashing out whenever the whim strikes him.
To shake the reputation you're referring to, he'd need to commit to a policy that's politically costly for him and then stick with it despite those costs. The conflict with Iran could go that way eventually, but this right now certainly isn't that.
The US has never bombed Iran before and Trump has for a while now has rhetorically taken a non-interventionsit position, so it is an important signal. This attack wasn't a whim; it's an old plan and Trump has been coordinating with Israel in their strike plans since February while considering this potential option, according to the Israel Hayom article.
"The US has never bombed Iran before and Trump has for a while now has rhetorically taken a non-interventionsit position, so it is an important signal."
That is the kind of thing that has useful reputational effects, but not the remotely the kind you're claiming.
TACO most emphatically does *not* stand for "Trump Never Lies For Years and Then Makes Sudden Reversals On Key Stances In High Stakes Situations." That would be TLFYTMSROKSIHSSS. Which aside from not rolling off the tongue nearly as well is not a claim that anyone who's been paying any attention for the past decade would ever make. Trump is notorious for lying. He's notorious for being unpredictable. He's notorious for taking big risks that calmer heads might not take.
Doing exactly the sort of thing he's notorious for doing is going to reinforce his existing reputation, not suddenly make people re-evaluate. Now, the thing that legitimately *is* going to cause reasonable observers to update is not the strike but the ceasefire. I am quite surprised (pleasantly so) that Trump seems to have de-escalted the situation so quickly.
It's still much too early to be tallying up points for the whole affair, but if he does a good job managing the ceasefire and steers the Middle East in a more peaceful direction, it *will* significantly update his reputation in my eyes (as I suspect in the eyes of many others). It still wouldn't really touch on the TACO thing--it's not that sort of situation--but I'd consider this sort of reputation more important anyway. Being able to manage a tense situation and broker a stand-down among mortal enemies is a much more important quality in a world leader than not flinching in the face of legitimately harmful consequences (which isn't, y'know, actually a good thing).
1. This is somewhat underspecified, which has led to a lot of confusion. Is Iran physically building a nuclear bomb, in the sense of "shaping a collection of 90% enriched Uranium, explosives and various other stuff to be able to produce nuclear yield?" Probably not just yet. But they're definitely doing a lot of stuff that is clearly intended to give them the capability to do that in short order. And from a practical standpoint, we should treat this about the same as one of those people who buys a maybe-not-a-gun-kit that you need to file like three things down to turn into a working gun, even if you can technically argue that "they aren't building a bomb".
2. Yes, so far as international law is a real thing (which it mostly isn't). Iran fired several hundred missiles at Israel in April 2024. That is an act of war by any definition, and there has been no peace agreement since then.
3. What diplomatic alternatives? Iran doesn't want to give up the program, and since Trump tore up the deal in his first term, there hasn't been a good way to get them back in.
4. Desirable? Probably, because the current regime is pretty bad. Viable? Much less so. The record of doing that from the air is not particularly good.
To be clear, I'm not saying I think it was a good idea. It's put Israel in an extremely awkward position where they basically have to keep bombing Iran forever, and that's not a great place to be. The US might be able to get out of it, although that heavily depends on the next few days. But I can also understand Israel feeling like they don't have a lot of options, given Iran's continued sponsorship of programs to make them not exist. Diplomacy is often messy.
We think there's a profound asymmetry in that Iran's ruling regime is really invested in annihilating Israel (the word they often use), and we are not at all invested in somehow erasing Iran from the map. This goes way beyond posture/rhetoric, in that Iran has for the past 45 years consistently funded/trained terrorist groups operating near and inside Israel, and encouraged state actors under its partial control to more hostility against Israel, creating a metaphorical "ring of fire" around it. As a partial list, these are Hizbullah, Hamas, and the regimes of Asad in Syria and Houthis in Yemen.
Israelis sincerely believe that if Iran got a nuclear bomb, the ayatollahs might well be crazy enough to first-strike it against us, even though we'd surely nuke them back (I understand that some reasonable people disagree with this, but it's possible that they underappreciate the investment Iran's made into anti-Israeli activities over the last half-century, to no apparent benefit to their nation). I believe that no reasonable outside observers would put any non-negligible probability on Israel first-striking Iran with nuclear weapons.
So, Israelis broadly agree that Iran getting nukes is an existential threat, even though they might disagree about its size. Netanyahu in particular has made a decades-long career of distracting voters from whatever domestic problems he's mired in by talking about Iran's existential threat, and the substantial part of the population that's fed up with him is really tired of it. NEVERTHELESS, even people who suspect Netanyahu of always maneuvering political decisions towards his own benefit, are still giving the government (and the army and intelligence, especially) the benefit of the doubt, because they feel that anything we can do to prevent Iran from getting nukes is vitally important.
1. Yeah, getting 60% enriched uranium was tough and expensive and they don't need it for anything else
2. I'm not sure. I think it's not more illegal than e.g. the US attacking Iraq, and I think it's definitely less illegal than Russia attacking Ukraine (because that's a war of territorial conquest, which is much much worse than "merely" waging war, both in terms of UN statute and common sense)
3. Hard to say
4. Basically the appetite grows with the eating, somewhat cynically. First two days of the war, no one was thinking of regime change, but Israel has achieved air superiority to a much greater degree, and saw much fewer victims of Iranian missiles, than planned for. Israelis think that most Iranians are fed up with the regime and that it's autocratic, oppressive and undemocratic, but we don't have regime change as a goal in itself, only insofar as it'd (possibly, uncertainly) help remove a brutal and dangerous enemy which the current regime is. It's not clear how viable regime change is, everyone says no one's ever achieved it by bombing and they're probably right; the level of discontent within the country might help but will probably not be enough absent further dramatic developments.
As an American I’m inclined to agree with you but I note among my peers the good will Israel seemed to have once have in this way has worn thin. Do you worry that Israeli actions in the recent war have cost in terms of optics (something I thought Jews as a people were mindful of)? Or do you see that as unavoidable or down to the whims of political trends?
There always was that tradeoff around preemptive strikes. But the attacker's advantage is overwhelming - compare the 6 days war to Yom Kipur, and Israel war with Iran to the war in Gaza.
It is not good for the world's stability, but this seem to be the case
1. Yeah, almost certainly. We follow the converging lines of evidence, right? Iran has had a "peaceful, civilian" nuclear program for decades and have produced:
a) three heavily-guarded and secretive enrichment facilities, one of which was dug into a mountain.
b) zero civilian nuclear reactors.
c) 400kg of 60%-enriched uranium. I.e., the sort of fissile material that no one would ever create unless they wanted to build a weapon.
Everyone agrees on all of this; everyone agrees that their breakout time has been measured in the weeks for a while; everyone knows they could put together and detonate a dirty bomb if they wanted. The only question is whether they're pushing for weaponization (i.e., building a ballistic-/hypersonic-deliverable warhead): Israeli intelligence said yes; American intelligence said no. Eli Lake & Justin Logan had a good debate on the Moynihan Report podcast (https://www.youtube.com/watch?v=4VtZy6flYJA) which went over similar grounds, and Lake made the point: look at all that the Mossad has accomplished in Iran. They set up a drone factory, called a bunch of high-level generals to an insecure meeting *against explicit orders*, they destroyed anti-air batteries on the ground—Israeli intelligence networks are very deeply entrenched in the regime. American intelligence, on the other hand, well, https://www.reuters.com/investigates/special-report/usa-spies-iran/. So I think we have good reason to trust the Israelis when they say Iran was actually breaking out, was actually pushing for a weapon.
(One more reason in this direction: the damage to the Axis of Resistance over the past months has been totally disastrous for the regime. They feel much less safe, and it's not unreasonable to think they'd make a last-ditch push for a nuclear deterrent.)
2. Israel and Iran have been in a state of war since October 7, 2023... If we want to play the "who struck first?" game, Iran doesn't win. I think it's a dumb game to play, though, and the strike can simply be justified as a preemptive, preventive measure against the Iranian nuclear threat (see [1]).
3. I do. The best Trump could've done would've been a rewrite of the JCPOA which absolutely did slow Iranian progress toward a bomb... but it did absolutely nothing about (exacerbated, in fact) Iranian support for awful, destabilizing terror networks. Iran is responsible for much of the carnage in Yemen, Lebanon, Syria, and, in fact, Iraq, not to mention Israel & Palestine. This oversight was bad; the terror networks were enabled when we lifted sanctions without sufficiently strict conditions. At worst, this war will put pressure on the Ayatollah to accede to a worse deal.
4. It's absolutely desirable—90 million people are suffering under a terrible authoritarian regime. Whether it's viable is another question: Haviv Rettig Gur (an excellent Israeli historian & journalist) made an appearance on Sam Harris' podcast the other day, and he said something like, "for 40 years, the regime has done nothing but suppress opposition groups. Their one job internally has been to make it impossible for dissidents to offer a realistic, well-organized alternative." I think there's a lot of truth in that, and it makes this very difficult. Because a ground op + occupation is not a good idea! I mean, the logistics alone... [https://deepleft.substack.com/p/against-war-with-iran]. Any sort of regime change will have to be homegrown. I think Israel and the US should make the environment as friendly to it as possible, and Reza Pahlavi should advertise himself to Iranians directly, and so on—but it's ultimately all up to the Iranian people. Israel should be sure to hit only legitimate targets (especially things like the Basij headquarters [https://www.timesofisrael.com/liveblog_entry/israel-confirms-hitting-basij-hq-evin-prison-destruction-of-israel-clock-in-tehran/], which was a very good move) to minimize rally-around-the-flag effects. But that's really all they can (should) do to help.
I think you might need to split this into two responses. One for the US, and the other for Israel. I'm not sure how you could possibly determine that Israel *doesn't* have a "legal" right to declare war against a country that's spent decades supporting multiple terrorist organizations with the primary purpose of destroying Israel, who also recently sent hundreds of missiles directly against Israel. If that's not casus belli then what possibly could be?
As for the other points, Israel couldn't do worse with a different regime, especially if it was popular with the Iranian people (the current regime is not). And diplomacy with a country willing to spend billions on a "we're going to exterminate your people" platform every year seems like a non-starter.
As for the nuclear program - it's just about incontrovertible that Iran has built the capacity to make a nuclear weapon and separately the capacity to mount a nuclear warhead on a mid range missile. They even made a point to show off the missiles and the payload size a few years ago. Whether they have or will take the more minor steps to complete a nuclear warhead and attach it to an existing missile is not known. This is in the face of multiple countries telling them to stop in very serious tones. That they are a major sponsor of international terrorism, including against the US and its allies, makes even the possibility of nuclear armament a significant concern for anyone on the potential receiving end of a nuclear strike.
The responses to you about number 2 in particular ("international law is fake", "the US should not be subject to international law") disturb me and tend to reinforce my opposition to the war if this is the sort of thinking that characterizes relatively thoughtful supporters of it.
International law is fake because there's no enforcing body. You can become an international pariah for using Big Nuclear Weapons -- and to some extent, a Little International Paraih for using chemical weapons (like Israel has). But these are schoolyard popularity contests, not the Principal. In the schoolyard, rules are enforced by "people willing to beat each other up, or, in extremis, kill the wrongdoer."
In that sense, there is international law, it's just fuzzy and ill-defined, and operates more on a "We Think That Makes The Game Unfun" (for example, assassinating world leaders makes the game unfun.) and "Can you catch us" basis.
It is a truth that economically ruining the German economy is against international law (blowing up the pipeline). That said, nobody's going to war over it.
Can you elaborate on your thinking on international law? Why are those arguments not thoughtful?
My steelman of those arguments.
Law: a binding custom or practice of a community : a rule of conduct or action prescribed or formally recognized as binding or enforced by a controlling authority [1]
Looking at the defiition I see there are two ways for a (international) law to be recognized at least de facto. Either a binding custom or practice and/or a controlling authority. As there is no world government, I think I can say there is controlling authority so international law doesn't meet that definition. I would argue that a controlling authority is needed for there to be any law, but I will ignore it so I can address the other part; a binding custom or practice.
I do not claim to know all of the treaties and agreements that the United States has signed. The United States is an UN member so has at least agreed to live by the UN Charter.[2] Article 2 of the UN Charter starts as "The Organization and its Members, in pursuit of the Purposes stated in Article 1, shall act in accordance with the following Principles." Bullets 3 and 4 are most germane to whether military force can be used.
3) All Members shall settle their international disputes by peaceful means in such a manner that international peace and security, and justice, are not endangered.
4) All Members shall refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations.
The above are supportive that the US may have acted contrary to the UN Charter. But there are two counterarguments.
a) Iran has been attacking US and Israel for the last 45 or so years through proxies. The proxies (Hamas, Hezbollah, Houthis, etc) give Iran some deniability. I guess if one does not accept those proxies are actually supported by Iran that may mean the US is acting badly. But I don't think anybody would argue that because there is so much evidence that Iran is the key (only?) backer of at least the 3 proxies I named.
So, it looks like Iran has been consistently violating 2.3 and 2.4 of the UN Charter. What should the US and Israel do to defend itself? I argue the US and Israel have legitimate reasons to attack Iran with military force.
b) Article 2 is making no claims to be a law. It itself says "principles" and not law or even rule. As the charter was likely debated for a long time, I don't think this is an oversight. So article 2 is not claiming to be law.
So, potentially article 2 can be used to justify the customs and practice part of being law. But also potentially, article 2 was written to give it the veneer of being a law, but the wording itself does not resemble a de jure law. Thus, I would say article 2 is not meant to bind any of the countries. I make this argument because I think the "customs and practices" are meant to address laws that are created by common law (or similar customs in non-Anglio phile world). So is there a custom to not militarily attack other countries? I think it is a worthwhile goal, but almost since the UN Charter's inception, countries have attacked other countries militarily. Russia attacking Czechoslovakia, North Korea attacking South Korea. US becoming militarily engaged in Vietnam. I think there is more evidence that Article 2 is a great ideal, but not really a binding custom or practice.
If the US or Israel has signed on to other treaties and agreements that limit use of their respective militaries, I would love to learn about them.
> Iran has been attacking US and Israel for the last 45 or so years through proxies
In any case where Iranian proxies have attacked the US it’s in war situations like Yemen. In Syria the US wasn’t directly involved but the support of rebels against Assad has meant that a branch of Al Queda have come to power, about two decades after the attack on the WTC. This is an odd war on terror.
Yeah, the Middle East is where the enemy of my enemy is my friend becomes a (realpolitik) fact.
Iranian proxies have attacked US directly outside of Yemen. It is suspected that the pre-cursor of Hezbollah was responsible for the bombing of the US Marine barrack in Lebanon.
I guess the attack could be considered a war situation, but the reason I used suspected is because the group that claimed responsibility is "Islamic Jihad", not Hezbollah. But arguably, if two "militaries" are attacking each other then it is a de facto military conflict.
Yes I know about the Beirut bombings. However if an example of Hezbollah attacking the west is pre Hezbollah, not in the west, during a military conflict and from the 80s the existential threat seems weak.
Your argument seems to move the goal post. Your argument now is that the threat needs to be existential to warrant a military response. My argument was that a country can respond militarily if it is attacked militarily. Which we have lots of evidence Iran was doing to the US and Israel.
I appreciate this comment, because I've felt a bit baffled as well but in the opposite direction. It's one of those situations where you listen to someone who disagrees with you and feel lost for how to engage. But after listening to a few podcasts and articles written by people who are opposed to the war, I think I understand why the two sides seem so far apart from each other.
It really all comes down to the Obama Iran deal and whether you think it was stopping Iran from building nuclear weapons. People who support the war think that it was a terrible deal because it wouldn't stop them from making weapons, and people who oppose the war seem to think it did stop them from making weapons and would continue to stop them. I've heard several commentors say something along the lines of:
1. The Obama Iran deal stopped Iran from getting nukes.
2. Trump tore up the deal, and failed to negotiate a new one.
3. Therefore, this war is not actually about Iran getting nukes.
It makes sense if you believe premise one, but people on my side don't and didn't. I remember back when Obama made the Iran deal all the commentators I would read and listen to were in agreement that it gave Iran too much and didn't do enough to get rid of Iran's ability to make a bomb. That maybe Iran would pause it's development, but the deal would leave Iran's ability to start up weapon production again any time they wanted with little ability to stop them. If I recall correctly that was the primary reason why the Republicans in Congress refused to vote for the treaty, which is why Obama had to make the deal as a non-binding executive action (which is why Trump could just toss it when he was elected).
We know that Iran had developed the facilities needed to create nuclear weapons. There's a reason they put the centrifuges under a mountain; it was so that Israel couldn't destroy it. You don't put civilian nuclear power equipment in a bunker under a mountain! Them getting to keep all that equipment and the capability to enrich uranium was not acceptable to our side, and certainly not acceptable to Trump. We are not going to just trust that Iran isn't going to use it going forward, even if they aren't using it right now.
So to answer your point 1: Whether they were in the process of building a nuke, they had the ability to create a nuke and Trump has been very clear for years that Iran having that ability is not acceptable. Yet in our negotiations with Iran they have made it clear that they will not accept any deal that removes their ability to enrich uranium. If they're not going to agree to it, then giving Israel the go ahead to take those facilities out (and then using our bunker busters on the ones Israel can't hit) is our BATNA.
As for your other points:
2. I don't know and I don't really care. If the actions violate a treaty that the US and Israel has signed then I may care, but "International Law" in general is not something I think the US should be subject to. Besides, was Iran following international law all these years as they have funded the Houthis, Hezbollah, and Hamas?
3. Yes, I think diplomatic alternatives have been adequately explored. Iran made it clear that they would not accept any agreement that removes their ability to enrich uranium, and Trump will not accept any agreement that lets them keep that ability.
4. I think regime change could potentially be desirable. The Ayatollah is an evil man who is the head of an oppressive state that openly wishes for the destruction of the United States and many of our allies. A regime that is less evil, less oppressive, and not actively hostile to US interests would be great! The question is how viable that is, which I'm not sure of. In any case, regime change or no they now no longer have the ability to enrich uranium. If regime change is necessary to keep them from re-obtaining that ability, then regime change may be required.
Probably an unpopular opinion, but I have quite a lot of trouble caring about whether Iran might have gotten nukes. Even as much as I'm generally in favor of nuclear non-proliferation, I look around and notice that the list of people who currently have the authority to launch nukes include:
1. Vladimir Putin
2. Xi Jinping
3. Donald Trump
4. Benajamin Netanyahu
5. Kim Jong Un
6. Narendra Modi
Obviously others will disagree, but from where I sit, that list includes a shockingly high number of the most evil and irresponsible human beings on the godamned planet. While I'd certainly prefer that Ali Khameni not join them, it's really, really hard to believe that the sky will fall if he does: not with club membership already looking like that. Meanwhile, the U.S. track record of getting involved in wars in the Middle East is utterly abysmal. I have no confidence whatsoever that this will produce any sort of positive, long-term change in the region, and plenty of reason to expect that it will make everything worse. On top of that, the domestic angle should not be overlooked: in using force without any buy-in from congress, Trump is once again pushing the legal and normative boundaries and trying to expand his power. It would be concerning from any president, but from a president who has done so much of that already, and sent so many strong authoritarian signals, it is downright alarming.
I agree that a lot of evil and irresponsible person have nukes, but that's not an argument in favor of letting another evil and irresponsible person have a nuke. Especially not when the cost to prevent them from joining the list is "Let Israel off the leash and drop some big bombs, nobody (American) gets killed".
"Especially not when the cost to prevent them from joining the list is "Let Israel off the leash and drop some big bombs, nobody (American) gets killed". "
If that is, in fact, the only cost then I agree. And things are looking better than they did yesterday, for sure. But the cost--in terms of escalation risk--still looks considerable, so I remain unconvinced for the moment.
We aren't "letting" Iran have nuclear weapons, any more than Iran is "letting" us have nuclear weapons. Neither nation has the power to prevent the other from possessing a nuclear arsenal if that's what they decide to do.
Excluding the very hypothetical case where we break out a few dozen of our own nuclear missiles and pre-emptively kill ten million Iranians while destroying Iran as an industrial nation, which I *hope* we all understand isn't going to happen even though it is the only way to stop Iran from having nuclear weapons if that's what they really want.
Is it your position that we (whether "we" is US+Israel, NATO, or all other nuke-possessing nations, or something else) can't possibly simply play whack-a-mole with any and all refinement sites and/or ICBM/MRBM launch sites Iran tries to put together, faster than they can do that?
I'll admit, part of me feels like that's very possible, in light of the last few days. Another part of me grabs that part by the collar and reminds it that Iran is roughly thrice the size Afghanistan, with comparable terrain, and the US couldn't pacify Afghanistan. Then again, Afghanistan never tried to build a nuclear missile defense program, and if it tried, we probably would have been able to notice and stop it pretty easily during occupation. OTOOH, Iran has a much better economy; OTOOOH, Israel seems to have considerably more infiltration; OTGH, I think I have to say I don't know either way.
The political side also doesn't tell me anything definitive. The West seems pretty gung-ho for no nukes in Iran; Russia and China probably have the only non-West votes that could matter, and their positions seem to be somewhere between "ehh" and "abstains courteously". But I can't help but imagine they could be natural allies on this issue.
Yeah, pretty much. It's the old saw of "we have to get lucky every time, they only have to get lucky once". Iran has the state and industrial capacity to run a massively redundant, decentralized nuclear arms program, and to rebuild it again and again if need be. They've got all the technical data they need, it's just down to manufacturing now. They understand OPSEC and maskirovka, and they've got an awful lot of generic factories, warehouses, and caves to hide things in.
And they've got enough 60% enriched uranium right now to build at least 8-12 bombs, that can be turned into 90% in a few months with a few hundred centrifuges. The HEU is basically indestructible if they're not stupid about it.
Because Iran spent most of the past twenty years developing a nuclear arms program but conspicuously *not* using it to build bombs, because they instead very conspicuously made it a point of diplomatic negotiation, Israel was able to construct a fairly precise and detailed map of what they needed to hit this time. They're not going to get twenty years next time, and they probably won't get even two.
I also left off Emmanuel Macron, for example and potentially several others (depending on how you count). The list wasn't supposed to be comprehensive: just the heads of nuclear-armed states that I already had a very poor opinion of. I couldn't even have told you Sharif's name without looking it up, so I can't possibly claim an opinion on his qualities as a leader or a human being.
I actually would rather any of these people have nukes than Iran. They're a religious theocracy convinced that military victory over their enemies is proof of divine favor, and who make real world massive sacrifices in furtherance of that belief. Everyone else cares more about preserving their country's existence by a much bigger margin than the regime in Iran.
I have a very poor opinion of Iran's leadership, but the "yeah, but THESE guys are too crazy to be trusted with nukes" line still feels like a very convenient excuse that gets trotted out whenever somebody wants to justify preemptive force. I certainly heard the same thing about Hussein in 2003. I didn't think any more highly of him, but I was still skeptical of that claim then, as I think I was right to be.
Hussein is the one leader in the middle east who was most explicitly in favor of getting nukes as a means to destroy Israel
If he was strategically smarter and didn't invade Iran and then Kuwait and just went for building nukes the world would have ended up in a much worse timeline
"I remember back when Obama made the Iran deal all the commentators I would read and listen to were in agreement that it gave Iran too much and didn't do enough to get rid of Iran's ability to make a bomb. That maybe Iran would pause it's development, but the deal would leave Iran's ability to start up weapon production again any time they wanted with little ability to stop them. "
That is correct. Also, a lot of people really, really wanted Iran to stop sponsoring international terrorism, and to stop holding annual "Death to America" rallies. And it's quite understandable that Americans would want all of these things.
Some of us also want a trillion dollars, working FTL starships, eternal youth, harems filled with enthusiastically consenting supermodels, and other such things.
There was never any possibility of Iran agreeing to a deal where they so thoroughly divested themselves of nuclear technology that the would have no ability to restart nuclear weapons production. Given their status as charter members of the Axis of Evil, they'd have been fools to agree to any such thing.
And there was never any possibility of the Europeans or the Chinese pushing Iran to accept such a deal under threat of Maximum Sanctions. They had too much money to make trading with Iran, and their interests were satisfied with an Iran that was holding indefinitely at six months from a bomb. And sponsoring terrorism against Not Europe or China, and holding Death to Not Europe or China rallies was just fine with them.
The deal that was possible, with the United States and Europe and China and Iran all on board, was one where Iran agreed to indefinitely hold at six months from an atom bomb. That was the *very best deal available*. It was that or nothing. Obama chose that. Trump chose nothing. Well, OK, he chose to bet everything on his deal-making skills trumping Obama's, but same thing.
Posturing fools, thinking that tough talk and a pious refusal to compromise will cow their rivals into abject submission. And now thinking that a Short Victorious War will solve everything.
I'm....super happy with the JCPOA, and I'm super happy with the current (now seemingly over?) war.
Iran has been massively weakened. Going to war with them 15 years ago would have been a bad idea, today it's a great idea. Their negotiating position barely changed in between. So the desired course of action has.
Note that the deal was not for Iran to hold indefinitely at six months from an atom bomb. The centrifuge and enrichment limits in the deal were to start phasing out after 10 years and fully expire after 15 years. In other words, even if Trump had not pulled out of the JCPOA and Iran fully respected it, the limits it imposed would have started to phase out in January 2026.
Excellent comment, I agree except for regime change. I don't want us involved in any more countries' internal nonsense. If we can stomp their nuclear capacity from the sky then great, let's do that and be done with it. No more endless and expensive rehabilitation projects.
The Iraqi model where we commit to propping up the new regime against all other comers is a bad one, it easily becomes an open-ended engagement on behalf of a new regime that probably doesn't deserve it.
The minimalist model of regime change is the one where you decapitate the current regime and make it clear that the next regime had better be nice or you'll decapitate that one too.
On the other hand, we stopped doing as much propping for the Iraqi regime and they're still holding on fine. Seems like regime change in Iraq "worked", but was far more expensive in blood and treasure than expected and arguably wasn't worth the cost.
Where the "Iraqi model" failed completely was Afghanistan.
1) Yes, I'd say with ~90% certainty. Otherwise, why would Israel and the US have attacked? Public reports like the one you linked aren't worth the paper they're printed on, the actual intel is likely all classified. Any side can say whatever they want. What matters is the ACTIONS they take. The US and Israel bombing Iran, despite the significant risks, is a pretty strong signal that there was credible intel that they were building a nuclear weapon. Iran refusing to sign a new nuclear deal with the US (despite crippling sanctions) is another strong signal that they are genuinely trying to build a nuclear weapon. ~10% uncertainty because honestly nobody really knows what is actually going on behind the scenes here and there are still technically plausible (though less likely) reasons to believe otherwise; I just think they're unconvincing.
2) International law is mostly fake and doesn't really mean anything. The US has veto power (de facto and de jure, depending on the particular arena) in most matters of international policy anyways. I don't put any stock here.
3) Yes and no, yes in that I think it's possible that a diplomatic policy that conceded much much more to Iran could have possibly succeeded, no in that I don't think that we could have achieved the results we really wanted (stronger guarantees they won't ever try to make nuclear weapons) at this point.
4) Viable? Yes with low confidence, the Iranian government seems pretty competant at staying in power for now, but they're obviously incompetent in a hundred other ways that might make it eaiser. I'd guess regime change would be harder than Sadaam's Iraq, easier than doing it in e.g. China or Russia. Desireable? Medium confidence, the Iranian government is pretty awful as far as governments go so any change would have a good chance of being better in general; but it's still a roll of the dice. There are good reasons to want it to happen and there are good reasons to be more cautious and hope for slow internal reform.
What do you mean by "falsified" here? What is your certainty estimate that Iran was/was not aiming to develop nuclear weapons? I could be persuaded my estimate is too high, but "there are counterexamples or plausible other reasons" fits totally fine within my model; I just think those plausible other reasons are less likely. I think there are obvious ways in which Iran in 2025 differs meaningfully from Iraq in 2001.
> Otherwise, why would Israel and the US have attacked?
Not disagreeing with your certainty estimate, but regardless of the true value, I would guess that Israel attacked now because this is the best time for them to do so. Iran's proxies around Israel are the weakest they have been for a while, so it's a better time than the past; and US approval is not guaranteed past the end of Trump's presidency or even the midterm elections, so it's a better time than the future.
Given the younger generation's feelings about Gaza, Israel may be facing several *decades* without the US backing that they are used to. So now is their last best chance to secure themselves for the years ahead.
Why would Israel have attacked otherwise? Some variant of the following: Netanyahu wants/needs to stay in power, and he has figured that his best way of keeping a coalition together, and of mitigating the widespread unpopularity of himself and his government, is as a wartime leader rather than a peacetime leader. At the very least, I don't think it's safe to reason from the premise that Netanyahu is acting in Israel's best interests, rather than in his own perceived best interests.
It's worth noting that he's under no particular political threat at the moment (his coalition is pretty much in line), and starting a war now in the hopes of it helping with an election in a year and a half seems counterproductive (if that were his reasoning he'd have preferred it closer to the election).
Netanyahu is under no particular threat right now because he seems to busy trying to destroy Hamas right now, and the Israeli people will for very good and sufficient reason rally behind pretty much any leader that is trying to destroy Hamas.
But fairly soon, either Hamas is going to be destroyed, or it's going to start looking like Netanyahu is just playing a lame-ass game of whack-a-mole with the remnant of Hamas rather than systematically trying to eliminate it. And then Netanyahu's political invincibility goes away, people start asking questions like who was responsible for the massive intelligence and military failures that let 10/7 happen in the first place, and the Israelis may wind up dumping Netanyahu as fast as the British dumped Churchill after VE day.
But unlike Churchill, Netanyahu is currently facing criminal prosecution for a laundry list of felonies that could keep him in prison for the rest of his life. So the theory that he's mostly waging war with Iran to renew his "You can't throw me in jail *now*, I'm busy Defending Israel Against an Existential Threat" card, is at least plausible.
It doesn't help him at all to do this now though. He's under no threat until the election in late 2026 (his court case is likely to drag at least that long, and his coalition partners are dependent on him). And there's no plausible way to stretch a political boost from an Iran strike until then (if it went on for that long he'd long since be hated for it). If he was doing this in mid-2026 it'd be plausible, but it doesn't work now.
Israel is not the United States, holding presidential elections like clockwork every four years and at no other time ever. Israel holds snap elections every time their governing coalition falls apart, which has historically happened about every two years on a most erratic schedule. The current administraion is overdue, and the end of a war is the sort of thing that causes major political realignments, so no, Netanyahu is not safe until 2026.
As I mentioned, I think there are plausible reasons (yours is one of many I've heard), but I don't think they are as likely or convincing as the far more trivial answer, which is that Iran really is trying to build a nuclear weapon and Israel/US really are trying to stop them.
The Israeli planning and groundwork for these attacks go back almost a decade. Israel was waiting for a window where Hezbollah no longer was a constraint on preemption. The opening of Syrian airspace also greatly helped the logistics (midair refueling closer to Iran). It's not like Bibi woke up a few weeks ago, looked at his poll numbers, and decided to Leeroy Jenkins the IDF into Iran.
That's an entirely different point. It's still not safe to reason from the premise that Netanyahu is acting in Israel's best interests, rather than in his own perceived best interests. If you have other reasons, you can put them in response to the original question!
Bibi spent years saying that anyone who wanted to help out Israel would fund Hamas. Either you think that "funding Hamas" is in Israel's best interests, or Bibi is not "in general" acting in Israel's best interests.
I beleive I answered the question: Israel judged Iran was still a massive threat and had an ideal opportunity to attack before its proxies could regroup. Even some of his political opponents are cheering the outcome so far; Bibi is far from the only hawk in Israel.
Israel, were it sane, would judge a "denuclearized" Iran as more of a threat than a nuclear power Iran. The first might/would threaten its relationship with the United States, without which there is no Israel.
This might be why Israel attacked Iran only when there was a serious prospect of Trump negotiating peace.
This is all good stuff. I agree you answered the original question, but I still don't understand why it's relevant to what I said, rather than to the original question itself.
I am not unambiguously in favor, but I'm on the other side of your arguments?
I think that Iran can clearly see that they will only be safe from hostile foreign action after they have nukes, so they are building nukes. We know they are working on nuclear power, they say it's for civilian purposes, I assume it's for nuclear missiles because they would need to be really stupid to trust that they would be safe without nuclear missiles - see North Korea and Ukraine for the relevant contrast.
I assume the war against Iran is illegal under international law but I don't respect international law to be sensible enough to care too much about it; I think the Iran hostage crisis forty years ago is a sufficient casus belli if the requirement is "only invade governments who act badly", but nobody is using it. We're just saying "we want you not to build nukes, so we're invading," and this seems like... honestly a better excuse than "only invade governments who act badly"?
I think we spent twenty years exploring diplomatic alternatives to war, but I also think they are pointless. Neither the Republicans nor the Democrats can be trusted to keep their side of the bargain, so obviously the Iranians won't keep a bargain they don't expect us to keep, and there's really nothing we can pay them worth more to them than being able to nuke attackers.
Regime change is viable (in that there's a loud opposition that protests the government regularly) and desirable (in that the current government seems worse than the opposition). The question is if it is worth paying the immense cost in treasure and blood that would be needed to achieve it, and if foreign intervention will make that cost "World War Three" instead of anything that could be worth paying, and those questions I don't know the answers to.
I had no thought of anything to review, and am too lazy anyway. If anything it would have been either roadside rest areas or mission-style gas stations, and that’s been done to death I’m sure.
So kudos to all for your creativity in coming up with novel subjects .
I am always tardy in knowing myself, in this case realizing that that Google doc was the thing I was born to review, and savagely.
My house burned down in the terrible Rowena fire in The Dalles, Oregon. Around 60 homes were destroyed in the wildfire, which is suspected to have been started by the cargo train that passes through the area. I was out of town for work at the time and was not able to get home to gather anything. While all my possessions were destroyed, whats more devastating is loss of my beautiful cat, Yuka. I don't know if he remembers this, but Scott played with and complimented Yuka after a meetup in Portland, I want to say in 2020 but the exact date escapes me.
I have been reading this blog for a long time, but I post rarely so I'm not sure if this is an appropriate place to share this. I was urged by many of my friends to make a GoFundMe. This probably doesn't fall under effective altruism. I'm blessed to have a good job and lots of support around me, but dealing with the insurance company is very stressful and anything donated here contributes to my peace of mind. I'm trying to hold what is here off to the side in case the insurance comes through, and if it does I'll be donating anything here to other families in my neighborhood, many of whom did not have insurance and are a lot worse off than me. (I read these open threads so often, but I'm not sure I've seen a GoFundMe so if this is a bad place for this, someone in charge let me know and I'll remove it)
Thank you. All things considered, I'm doing okay mentally, but I am starting to feel the stress as I get into it with the insurance company. I'm planning on talking to a lawyer this week and that should help ease my mind a bit. From my experience talking to a lawyer can only help.
This is the second year that The Emperor of All Maladies didn't make it. I thought it deserved to be a finalist last year, so naturally I would have liked to see it be a finalist this year as well. However, it was a book review in a "not a book review" contest so I could see it getting penalized for that.
Thanks for your support as always! To be honest I re-submitted it largely because you and a handful of other commenters said encouraging words, but I wasn’t too surprised to see it miss out. The ACX community likes hard facts and analysis, not Christian symbolism and TS Eliot references.
Ah well, c'est la vie. You should publish it somewhere, even if just on Substack. I'd let you guest post it on my Substack if you wanted, but I have less than a hundred subscribers so it wouldn't be much of a platform.
You’re more than welcome to publish it (anonymously or under your name or whatever you like.) I wrote it so that maybe people with cancer or their families would connect with it, not for glory or fame.
I know an oncologist who’s been published in newspapers and written several books so perhaps I’ll send her an email and see.
There's something that I find really suspicious about the claims of the danger of AI that Scott makes, and some of the posts Scott makes, which I think makes claims of the danger of AI less believable.
Like in the AI 2027 scenario Scott endorsed, even the aligned scenario is pretty dystopic (a small group of people with all the power, no real guarantees of good outcomes despite the scenario trying to paint over it imagining a luxury future). Since both scenarios are pretty bad, and specially since there's a large given chance of the worst scenario, someone who believes them should be expected to oppose AI with all their power, trying to slow it down as much as possible, as early as possible, to improve the end results.
Honestly I'd expect anyone who thinks AI is very likely to go catastrophically wrong, and who cares about morality at all, to be radically militant in their opposition to AI, probably spending most of their wealth in measures against it, after all if the AI gets as powerful as it's claimed, there's no turning back, and the damage would be tremendous. So do we see this kind of opposition?
Obviously there's parts of this opposition that we can't see just looking at Scott's blog, but what I'm confused by is posts like "The Colors Of Her Coat", or the "Make A Personalized AI Kids’ Book" mentioned here (which I haven't read, but from the title can infer its bent). These kinds of posts serve as propaganda for AI, helping hype it up and promoting a positive view of it, which seems really incongruent with thinking the AI will be very likely bad. If AI is so likely to be catastrophic, why is he using his blog, which probably is a large part of the political capital he can use to exert influence, to make posts to say "ooh look how pretty the pictures the computer makes". Surely if it's a matter of such import, any influence we can wield to make the chance of bad AI go down is useful, and any influence that can increase the chance should be avoided.
If Scott thinks the chance of bad ai is so great, and such a big threat, why does he still make posts like that? even if they help shift attitudes only a smidge, given the stakes it still should be taken into account. shouldn't he avoid making them, or even retroactively erase them? even more, why doesn't he try to make posts that help build opposition to AI, like he has never really gone into the economic arguments against AI, which is very relevant for artists. All of this seems at best like a strategic blunder, and at worst, evidence of bad faith. Why should we believe the hyperbolic claims for misaligned AI, if he doesn't even shift his blog away from things that help hype up AI?
To me this just makes these claims not credible, those attitudes are not those of someone for whom AI is the greatest threat facing humanity, they're more those of someone who maybe is helping hype up AI, because maybe they have some investments that would benefit from that hype. I don't want to accuse Scott of being a bad faith actor, maybe he just follows his intellectual whims, maybe he just hasn't considered the part of strategy revolving around his blog posts (itself a bit of a strange omision), or maybe he thinks it's intellectually dishonest to post strategically? (though as long as it's not misrepresenting reality it shouldn't be) but regardless, seeing that kind of very "ai positive" post pretty strongly tells me "this guy doesn't consider this a serious threat"
I was especially turned off during his most recent back-and-forth with Tyler Cowen, in which both sides cited o3. I preferred it when Scott would simply state outright that he didn't have a source ("Epistemic status: I made it up") and frame his arguments accordingly.
By all means, use AI to find real sources--but then check those sources to make sure they're real, and that they actually say what the AI claimed they said, and then link directly to them.
I'm likewise annoyed that Scott spends a lot of time worrying about AI but then generates (almost invariably poor-quality) AI images for use on the site, and effectively *promotes* them on posts like the "Make a Personalized AI Kids' Book" (which had mostly poor-quality AI images, FWIW).
Maybe it's just a finger in the cracking dam, but let's try to *avoid* all of the AI bullshit, m'kay? The image-generating stuff is ugly and the LLM stuff is both badly styled and can't be trusted not to make stuff up, cite fake sources, or have random agendas.
I don't see a contradiction. AI being useful is a precondition for AI being dangerous in the sense that Scott et al are worried about. Moreover, if I understand the AI safety arguments rightly, the aim is to try to maximize humanity's chance of threading the needle of building a useful and well-aligned ASI and using it to substantially better the human condition. Butlerian Jihad scenarios where there is a concerted effort to kill and bury AI development are considered a failure mode to be avoided, especially the risk that the attempt is unsuccessful (i.e. most countries ban further advanced AI development, but a badly-aligned ASI gets developed and deployed anyway in a non-banning country or a country that fails to enforce its ban well enough), just as much as the scenarios where laissez-faire regulation permits rapid development of unfriendly AIs.
Caveat: I am explaining a position not my own. I expect with moderately high confidence that the current crop of LLMs will plateau in capability well short of ASI or even AGI. They'll prove to be useful tools once the dust settles and the hype dies down, especially for natural language processing and computer vision, and might be an important piece of the puzzle for future AGI development, but I consider Scott's 2027 scenario very unlikely. The only reason I don't dismiss that scenario completely out of hand is that smart, knowledgeable people whose judgement I respect (e.g. Scott) have spent time and effort analyzing the issue and concluded that it is likely.
I would note that "butlerian jihad scenarios" are a failure mode to people like Scott and Yud, because they actually do want to reap the potential spoils of AI down the road and are okay with elements of transhumanism. But a concerted effort to kill and bury AI development for good would absolutely not be a failure-mode to the growing group of opponents who simply do not want this ever to be created, and instead prefer to maintain human supremacy indefinitely. People like Scott probably will not like many of the motivations and arguments that those real people bring to bear against AI, but I have no doubt that messy ugly chauvinistic or fear-driven populism is a more powerful force, and I think at some point they will need to decide whether to tactically join the growing coalition that doesn't want this stuff to ever become part of human life.
I for one very much doubt that mere intelligence is sufficient to do the types of good that AI proponents are always touting, many of the more longterm futurist dreams could just by physically impossible, and I'm not even convinced that e.g. a cure for cancer is a thing that can be created no matter how smart you are. All of the destructive uses, OTOH, have proof of concept, and are much more likely. I find every possible outcome where a superintelligence exists to be worse than humans retaining primacy on the earth, and this will just be a battle we have to fight and a line we have to hold for as long as we value our species.
It's tempting to agree that LLMs are looking like they may plateau, but I think we are possibly not getting the whole story of what they may have internally (and I get that we can't trust that due to either hype or possibly downplaying some research angles), and AI2027 really only depends on them reaching better-than-human coding ability, along with some assumptions about "research taste" slowly shifting over to them. That seems like the tipping point, that it would eventually happen from there, even if the timeline is likely to be longer.
i don't think AI being useful is a precondition for being dangerous. If AI decides to... say, "make anime real" that's not actually useful (unless you care a lot about meme campaign promises). But "make anime real" costs a lot of computing time. Whether this leads to melted ice cream, and consequential health risks, or whether this costs google some fraction of search capacity, we can say that AI "wanting to do big things" is dangerous, regardless of whether the AI is useful or being useful while doing so.
The danger that Scott, EY, etc are worried about is an AI that is useful enough in the short-to-medium term that humans are tempted to trust it with real power which if misused could lead to dystopia or Armageddon. This might happen directly by using it to automate real physical stuff at large scales (like the classic fictional trope of giving and AI control of nuclear ICBMs) or indirectly by uncritically taking its advice at corporate or national policy levels (like in some of Asimov's robot stories, although there the AI takeover is intended to be benignly paternalistic rather than malignant or hostile).
Even the kind of danger you're talking about, of wasteful use of resources towards frivolous aims causing environmental damage and opportunity costs, doesn't happen unless people with control over large amounts of investment capital think it's to their advantage to spend huge piles of money developing and running that AI.
It's certainly possible for an AI to seem much more useful than it actually is. I strongly suspect this is happening now with the current crop of LLMs. But that should be self-limiting as eventually the hype cycle dies down.
If you think LLMs are currently mostly useless, a dead end in terms of capabilities, and the worst case scenario is wasted resources, then pragmatism and honestly should both incline you towards making the case that they're useless and wasteful. But if yo think they could easily become either an existential threat to humanity or the wellspring of a utopia, then Scott's approach of talking up both their utility and their danger makes sense.
It's not about AI being useful by itself, it's about reinforcing the view that it's useful in other people's eyes, thus helping drive engagement with it, helping drive hype, helping drive investment, helping drive development (is this a too weak slippery slope? I'm not sure what the weakest link is, but my point is, given the supposed stakes, gambling with the chances doesn't seem rational). The color of the dress post even speculates about post singularity AI giving us wonders, thus helping people be slightly more comfortable with the idea it'll work out alright/it'll end up aligned. But the whole point of the more AI safety concerned people is that alignment is extremely unlikely, so we should be extremely, extremely cautious about continuing development. Maybe butlerian jihad is a failure mode of that, but a big part of the point of people like Eliezer and Scott is that threading that needle is extremely unlikely, so we shouldn't risk it until we're extremely extremely sure. In that effort, and seeing how the capital of the world is prone to hype and short term gain, maximizing opposition to AI, for someone in a pretty influential position, seems to me to be most consistent with the position.
A problem with strategic lying is that people may draw conclusions different from what you intended, and then you have made the situation more difficult for yourself.
If you convince people that AI is useless (which many of them already believe, it seems to be quite popular opinion), that would imply that it is harmless. No need to worry about it, no need to regulate it...
I'm not advocating strategic lying, rather strategic selecting of topics, while remaining truthful... so not choosing to highlight how cute and pretty the computer images are, and instead choosing to focus on topics that are still true, like the economic aspects I mentioned, but which reinforce negative sentiment. I don't think that strategic selectiveness would be any more intellectually dishonest than following intellectual whims. And yeah, implying it's useless could be counterproductive, so the topic of usefulness could be avoided, and instead focus on how it's evil or bad in other ways.
The real AI safety activism is political organization. Right now we're in the "this is so cool!" stage, like Facebook or smart phones in 2009. Soon we'll be in the weaponization stage and lobbyists will go to the wall to prevent regulation or raise taxes on the AI owners to pay for UBI.
There's something that I find really suspicious about the claims of the danger of AI that Scott makes, and some of the posts Scott makes, which I think makes claims of the danger of AI less believable.
Like in the AI 2027 scenario Scott endorsed, even the aligned scenario is pretty dystopic (a small group of people with all the power, no real guarantees of good outcomes despite the scenario trying to paint over it imagining a luxury future). Since both scenarios are pretty bad, and specially since there's a large given chance of the worst scenario, someone who believes them should be expected to oppose AI with all their power, trying to slow it down as much as possible, as early as possible, to improve the end results.
Honestly I'd expect anyone who thinks AI is very likely to go catastrophically wrong, and who cares about morality at all, to be radically militant in their opposition to AI, probably spending most of their wealth in measures against it, after all if the AI gets as powerful as it's claimed, there's no turning back, and the damage would be tremendous. So do we see this kind of opposition?
Obviously there's parts of this opposition that we can't see just looking at Scott's blog, but what I'm confused by is posts like "The Colors Of Her Coat", or the "Make A Personalized AI Kids’ Book" mentioned here (which I haven't read, but from the title can infer its bent). These kinds of posts serve as propaganda for AI, helping hype it up and promoting a positive view of it, which seems really incongruent with thinking the AI will be very likely bad. If AI is so likely to be catastrophic, why is he using his blog, which probably is a large part of the political capital he can use to exert influence, to make posts to say "ooh look how pretty the pictures the computer makes". Surely if it's a matter of such import, any influence we can wield to make the chance of bad AI go down is useful, and any influence that can increase the chance should be avoided.
If Scott thinks the chance of bad ai is so great, and such a big threat, why does he still make posts like that? even if they help shift attitudes only a smidge, given the stakes it still should be taken into account. shouldn't he avoid making them, or even retroactively erase them? even more, why doesn't he try to make posts that help build opposition to AI, like he has never really gone into the economic arguments against AI, which is very relevant for artists. All of this seems at best like a strategic blunder, and at worst, evidence of bad faith. Why should we believe the hyperbolic claims for misaligned AI, if he doesn't even shift his blog away from things that help hype up AI?
To me this just makes these claims not credible, those attitudes are not those of someone for whom AI is the greatest threat facing humanity, they're more those of someone who maybe is helping hype up AI, because maybe they have some investments that would benefit from that hype. I don't want to accuse Scott of being a bad faith actor, maybe he just follows his intellectual whims, maybe he just hasn't considered the part of strategy revolving around his blog posts (itself a bit of a strange omision), or maybe he thinks it's intellectually dishonest to post strategically? (though as long as it's not misrepresenting reality it shouldn't be) but regardless, seeing that kind of very "ai positive" post pretty strongly tells me "this guy doesn't consider this a serious threat"
Scott, you listed "Men" as a finalist, but there's no review titled "Men", and two reviews which include the word "Men" in their titles. Could you please clarify which review you're referring to? Many thanks!
The other was "Shrinking: Men", which Scott wouldn't have abbreviated with "Men". So yes, it's "The Men Are Not Alright". Which I also found excellent.
I keep hearing that AI is putting Silicon Valley techies out of work, but I have yet to see any verifiable numbers and sources. If anyone can point to some relatively well-controlled studies (or at least studies with transparent methodologies), I'd appreciate it. In the meantime, it seems like a recent change to the tax code has likely been the real culprit in the current round of layoffs. Tech companies can no longer deduct the full cost of R&D at the end of the year...
Hi beowulf. When I posted a while ago that I was reading The Dark Tower series and loving it, you replied that you'd liked it a lot too, til the end when Roland reaches the dark tower. I'm piggybacking on this thread to tell you I absolutely hated the end too. Not only did it cause me pain and suffering, your honor, it also grossly violated my values regarding what constitutes fairness and logic in literary narrative.
It was obvious in the book (also from knowing a bit about King) that he was not comfortable having a hero that killed a bunch of people. It doesn't align with his politics and his values. Seems to me he just regressed to his 60's liberal-pacifist values and gave in to a craving to punish his hero.
And the ending does retroactive damage to the whole story, right? It's all just a re-run.
I am so bothered by the ending that I am thinking of writing a new one. I'd just do a new version of the very last bit -- the part that happens after Roland enters the tower. So I thought I'd ask you how you'd like it to have ended.
I did too. But I also thought Roland had earned an outcome of that magnitude. He was born with huge talents and used up every bit of himself turning them towards saving the world -- all the worlds -- and he succeeded. Even the people who died or suffered helping him -- he helped them too, and sharing his life for a while made their lives larger.
A lot of things happened over the past five years that showed up at about the same time:
1) "Free"money went away. It is easier to justify borrowing a lot of money to pay for things like engineers when the interest rate is near 0%. At 5% things change a lot.
2) During Covid Facebook and other 'tech' companies (not Apple) had a huge hiring surge. With no clear idea what to do with the folks hired. This created a bunch of employees with no clear financial justification and when that collided with (1) ...
3) Elon Musk fired about 75% of Twitter and things seemed ... fine? This can encourage other company execs to ask why they have the headcount they do. Especially when combined with (2)
4) The tax changes that you mentioned. I'll note that my employer is sorta tech (we build expensive semiconductor test equipment) and we haven't experienced big layoffs. But we are older and profitable so financially different from a lot of the smaller tech companies.
5) At least one fairly large tech firm, Intel, is in financial trouble. Intel had 130,000 employees in 2022, finished 2024 with about 108,000 and is expected to lay off another 10,000 soon. This isn't because of tax code changes :-)
6) AI is disrupting, but *how* is not clear.
So ... yeah, tax code changes. And maybe that was the biggest factor. But it wasn't the only one.
My understanding is that the tax code changes are especially punishing to rapidly growing companies like startups. They won't matter that much to a stable company.
2) is perhaps better understood in the context of BLM, and the tech companies hiring a lot of "Good Workers" to do nothing, in order to have their stats look better.
3) Elon Musk bought a company that had no Test System, and had been doing hotfixes to Prod for 6 years. There was/is? serious doubt if X.COM goes down, it can be rebooted.
I'm possibly missing something here, but can't Scott just use a convertible promissory note that triggers on a later financing round or some other appropriate condition? Just put no interest on it until maturity, and intend to forgive the loan if the triggering event doesn't happen prior to maturity. That is a financial instrument that already exists and you don't need to reinvent the wheel here.
The problem is that when the grant is given, there is no for-profit corporation. There’s no entity that has equity to negotiate over. You can’t have a convertible note with a company that doesn’t exist yet.
I suspect the deeper problem is that when a nonprofit converts into a for-profit, in most cases the people involved simply create a new for-profit entity with no formal relation to the nonprofit. The OpenAI situation where the nonprofit clearly has a lot of valuable intellectual property is quite rare.
But it would be great to get something like this working. In general the puzzle of how to fund nonprofits in a way that makes it possible for them to grow and succeed seems unsolved.
Sure, assuming that we mean a formally-registered not for profit. Those are corporate entities with identified officers and etc. They are governed by state incorporation laws and, just as with an LLC or whatever else, can/do enter into contracts or debts with a willing counterparty.
If the not-for-profit corporation has achieved federal tax-exempt status (a separate process conducted by the IRS), then you know that it has full legal status in its state of registry. The IRS by law can't even consider an application for 501(c) status if the applicant isn't "in good standing" as a corporation under its home-state's laws.
There are plenty of local groups which identify themselves as nonprofits and behave like nonprofits but don't do the paperwork. You should not donate to any such group, ever, but they do exist.
But anyway it sounds like I misunderstood the actual question posed.
The point I was getting at was more like "what would the note be convertible to, since nonprofits can't issue equity", but Kevin made it better in the other subthread.
Inspired by Scott’s occasional forays into speculative fiction, as a way of thinking through strange ideas sideways, I’ve started a fiction blog of my own. The stories lean speculative, sometimes philosophical, sometimes more straightforward, but always written with the hope that the narrative does more than entertain. So far, I have touched on such small topics as the nature of the universe, a direction superintelligent AI can take us, and encountering truly alien lifeforms.
For those who enjoy puzzles: I’ve hidden Easter eggs throughout—some overt, others deeply buried. A few are probably too obscure (sorry), but most are findable with the right mix of attention and paranoia. I’d be curious to know what you spot.
If that sounds like your kind of thing, you’re warmly invited to explore:
Thank you for the kind words! I would very much appreciate a shoutout, if you might be so kind! I hope you subscribe too - I have a couple more in the works now.
I'm kind of surprised none of the game/movie/TV/music reviews made the cut but I guess that's just the ACX audience. Was sort of hoping for at least one token game & TV/movie review to make it in just for variety (and also because I definitely wrote one of those).
The game reviews mostly weren't as good. There is no universal rule that everything related to video games is low quality and low effort but gamers gonna game and do poor work.
Sure, there’s the audience considerations, but I honestly thought the video game reviews here were a bit weak compared to the quality of the kind you can find on other websites. Even if the ACX audience were more open to them, I still don’t think any of them would have made the cut.
The Tv Show reviews were on average much better, but because of how unconventional this year’s format is, I think adding a few token representatives would have decreased variety, ironically enough.
Ollantay was put in that category, although it's a stage play. Other than that, the media review (as opposed to something-else review) that I ranked highest was Kiki's Delivery Service, one of five that earned a 9 from me.
I understand that a perpetual motion machine is prohibited by classical physics. But what about at the quantum level? Since atoms are always in motion, except at absolute zero, could one construct a perpetual motion machine that would continue on until the universe is empty?
A superfluid current in a perfectly isolated toroidal container would, I believe, theoretically continue forever. This isn't really what's meant by 'perpetual motion' though, which people use to mean energy output greater than energy input.
Sure, under those rules the machine will have "perpetual motion," but in the classic sense of usting them to power things, as soon as you hook a belt and pulley or a turbine up to that spinning atom you have introduced friction for the purpose of removing energy, so game over!
An atom can stay in motion forever, but that isn’t typically what people mean by a “perpetual motion machine”, they mean something that you can extract useful work from with no energy input, and that is still forbidden by quantum mechanics.
Classical mechanics would allow the "literally in motion forever" perpetual motion machine--say, a planet with a moon orbiting it. But that's not what people mean by the term , so....
I think there are two categories of perpetual motion machines, which then have strict & relaxed definitions.
1. The truly impossible perpetual motion machine is one that produces net energy. Lots of cranks latch onto this one because it’d be a source of “free energy”. Doesn’t work, even at the quantum level
2a. A machine that keeps running forever (where forever is defined as a reasonably long time). For this, I suspect there are quantum perpetual motion machines, but from a certain perspective most orbits are perpetual motion machines, so it’s not that special.
2b. A machine that keeps running for a truly infinite amount of time. For this I suspect it’s impossible even at the quantum level because you have to grapple with decay of subatomic particles and the heat death of the universe.
My favorite old candidate for a perpetual motion machine was a strong magnet pulling an iron ball up a ramp with a hole near the top.
The thinking was that the ball would drop onto a lower ramp and return to the starting point. An appealing idea at first glance till you think it through and realize that a magnet strong enough to pull the ball up the ramp would be stronger than the force of gravity met at the hole.
Ask anyone who has recovered a magnet from an old hard drive and loosely held it within a couple inches of their refrigerator how gravity would fare in that contest.
Heck, I bought some magnets at Michael’s once to do some cute crafting, laid everything out separately on the table, and struggled to glue before the magnets did their own cute crafting and stuck themselves together.
You can stay in motion...as long as you don't try to extract work (ie energy) from the system. That's also true classically.
Yes, a particle in a well has non-zero kinetic energy even in the ground state. But extracting that requires being able to transition to a lower allowed state...and there aren't any below the ground state. Definitionally.
Stumbled across an article that will probably be of interest to some ACX subscribers. Posted it last week in a reply to somebody, but I'm going to repost it as a top-level comment.
India is always stereotyped as an extremely spiritual place even historically, but there was actually a philosophy called Charvaka or Lokayata which is purely materialist. Most scholars have believed that this philosophy is relatively short-lived and mostly irrelevant for most of Indian history, but new research seems to show things aren't so cut-and-dried. (Notably, the name Lokayata literally means "prevalent among the people.)
There are some flaws with this article - for instance, the quote from Maitreyi in the Upanishads is misleading, since in the original context "no awareness" meant merging with Brahman, not annihilation - but overall there's some really interesting stuff there.
I'll also add one thing not noted in the article: that the oldest strata of Tamil literature, the Sangam poetry, has no references to religion or the supernatural. Tamil culture, especially at this time, was probably descended more directly from the Indus River Valley civilization (whose people likely traveled to the southern tip of the subcontinent) than Northern Indian culture was, having been subject to serious Indo-Aryan admixture, so it's entirely plausible that the Indus River Valley Civilization sites (Harappa, Mohenjo-Daro) had no religion as we would now conceive such things.
That's very speculative, though, and Mohenjo-Daro was structured in multiple levels with a large bath at the center of the city, which would seem to indicate some notion of ritual purity, like that which upholds the caste system, had developed. And ritual purity presumably comes with some degree of religious belief.
"No big temples or ritual centres that we have identified" does not mean "no religion/belief in spirits". People might have had little shrines in their homes, or they might have practiced rituals in a small way but not like organised religion. Or if they all got together round a bonfire at a particular part of the year and did things like drive their flocks through it for good luck, is that religion or not?
Roman funeral practices included having wax masks of dead ancestors/actors wearing those masks carried in procession. Is that religion? If they did something similar in Mohenjo-Daro, what would have remained to be dug up centuries later?
Yeah, the Pashupati seal is probably the biggest piece of evidence against the idea that they had no religion. It does seem to portray a god, but for what it's worth the position of Wendy Doniger (a scholar who supports the protoatheist hypothesis for the Indus River Valley Civilization) is that we have no way of knowing what the Pashupati seal is meant to be portraying.
We have no idea what the Gundestrup Cauldron is meant to be portraying, either, but it's an interesting parallel (horned male figure, seated, surrounded by animals):
I'm considering changing my job, and perhaps my career, but I feel like my skills are too eclectic at this point, and I'm pretty confused at what my options are, and I'd like to brainstorm with someone because I feel pretty lost. I'm considering hiring a career coach, if only for the sake of someone holding the other end of the brainstorming conversation, but I worry that a regular career coach would try to shoehorn me into one of the mental categories they already have even if it's a bad fit; is there some version of it that may realistically work on someone all over the place and from the general ACX-comments crowd?
UPD. I wrote up broader context in response to a question in a comment; here it is:
I studied linguistics out of passion (understand the common laws of human communication systems yadda yadda), and from there I switched to ML (because natural language processing) in like 2016, before transformers were even invented; ML seemed really cool to me, the closest thing to magic in existing technology. I studied lots of math and other technical stuff as part of my linguistics education (our program was informally known as "mathematical linguistics"), but not as deep as a real mathematicial, and my knowledge of linguistics never became as deep as a true linguist's.
Then I went to France to a PhD school, for a thesis about natural language generation (this was before RLHF); this was not a topic chosen by love, I just wanted some position and I seemed to just like NLP generally and figuring things out.
The PhD broke me due to what felt like any lack of meaning, guidance, goal-setting etc; experiments I designed were not interesting in any ways and failed miserably. I decided that I wanted to be part of a startup, where everyone is working toward a common goal; where my chaotic strengths of broad figuring-things-out and hyperfocus would be valued. Besides, leaving my PhD meant my French residence permit was not valid anymore, and God forbid I'd have to return to Russia, so I was looking for a country that wasn't trying to arrest me or kill me. I didn't get a PhD degree (I do have a Master's).
So I joined an ML startup in a small country in West Asia, and loved the atmosphere. (I'm currently the only ML engineer in this ML startup. We're a team of like 10 people.) My way of getting shit done was finally valued, and I appreciated the stock options. Besides, being startup meant trying to be lean, using smaller, economic methods, instead of any "we were dancing around for 6 months and got a 0.5% improvement in a leaderboard" stuff. In addition, I really found a use for my passion for linguistics. Since our startup is connected with machine translation, being able to read our customers' data and actually understand what goes wrong where had been a really valuable addition to my engineering competences.
But now I've been burning out. Here's why:
· Interactions with the team and especially C-level management.
— The team was initially pretty diverse with people of all backgrounds, so I didn't feel weird being a foreigner and various other labels of weirdo, but now I'm the odd one out.
— Recently, I've developed severe sensory sensitivity, and communicated to them that since they are not willing to allow remote work, I'd like to work from upstairs (we're renting a two-floor house for an office); they felt it was too distant from the team, but allowed me to work from the kitchen. Now, however, they told me that a permanent move outside of the room where everyone works is not acceptable, in any other area even if it wouldn't inconvenience others, because that may threaten the quality of communication (the theory being that communication is much easier when they can walk up to me in the same room than when they have to open a door and see me in another room). I did explain that it influences my productivity, but they said that basically it's not only my productivity that has to be managed, but other teammates' as well (I'm still doubtful other teammates' productivity would suffer from me being in another room as much as mine suffers from a constant sensory overload).
— There was a time when I showed up with my nails polished red (I'm biologically male and look male), and the CTO personally told me that I would have to wash that off, because what if some day I'm in charge of hiring people and they are uncomfortable, or what if my teammates are uncomfortable ("So if I ask them and everyone is OK with it..." — "No, what if they are uncomfortable but too nice to tell you?")
— All of that makes me feel like there is a certain amount of, uh, restriction of my personal independence / agency, because who the hell cares where I sit or how I paint my nails if I ship quality code fast? And being the only ML engineer in the team, I suspect they may not realize that my leaving would actually be an inconvenience, since there is some learning curve to our infrastructure...
· I'm tired of ML / AI as a field. While I was shipping small and cheap solutions, NLP started to suffer megalomania, and everyone hiring NLP engineers seems to have training gigantic chatgpt-like shit in mind, which is not sexy for me (and I still have no experience with it, but I could figure it out). It seems foolish not to use my tired knowledge to *some* benefit, but I would hope to apply it somewhere where I still see a sense of meaning.
In addition, in these recent years searching for meaning, I have pursued various pet projects. For example, I have created a project called theorem marketplace (link: https://theorem-marketplace.com/) and while doing it, figured out some basic web3 development (because I had to create an Ethereum contract with an oracle). Also, I have periods when I'm in love with molecular biology, so I'm considering barging into bioinformatics. (In pure AI, I find interpretability research a la anthropic very sexy, but I have no credentials to signal I would be a good fit; and maybe I wouldn't).
Have you looked at all the EA adjacent interpretability programs? (ARENA, MATS etc) Some end in research papers and are essentially tailored for professionals that want to move towards AI safety research (of which Intepretability is a relevant subfield)
Bioinformatics probably requires you doing another PhD. I would recommend you looking at specific topics you might be interested in and trying to pick specific supervisors. Scope and quality of different bioinformaticics work varies widely.
I had done a pretty drastic career shift myself (Medicine → CS). I would say the most difficult part is primarily a lack of network, so if you are able to create work in your spare time that signals for anyone interested that you are highly motivated, that can be really helpful to be seen.
I studied linguistics out of passion (understand the common laws of human communication systems yadda yadda), and from there I switched to ML (because natural language processing) in like 2016, before transformers were even invented; ML seemed really cool to me, the closest thing to magic in existing technology. I studied lots of math and other technical stuff as part of my linguistics education (our program was informally known as "mathematical linguistics"), but not as deep as a real mathematicial, and my knowledge of linguistics never became as deep as a true linguist's.
Then I went to France to a PhD school, for a thesis about natural language generation (this was before RLHF); this was not a topic chosen by love, I just wanted some position and I seemed to just like NLP generally and figuring things out.
The PhD broke me due to what felt like any lack of meaning, guidance, goal-setting etc; experiments I designed were not interesting in any ways and failed miserably. I decided that I wanted to be part of a startup, where everyone is working toward a common goal; where my chaotic strengths of broad figuring-things-out and hyperfocus would be valued. Besides, leaving my PhD meant my French residence permit was not valid anymore, and God forbid I'd have to return to Russia, so I was looking for a country that wasn't trying to arrest me or kill me. I didn't get a PhD degree (I do have a Master's).
So I joined an ML startup in a small country in West Asia, and loved the atmosphere. (I'm currently the only ML engineer in this ML startup. We're a team of like 10 people.) My way of getting shit done was finally valued, and I appreciated the stock options. Besides, being startup meant trying to be lean, using smaller, economic methods, instead of any "we were dancing around for 6 months and got a 0.5% improvement in a leaderboard" stuff. In addition, I really found a use for my passion for linguistics. Since our startup is connected with machine translation, being able to read our customers' data and actually understand what goes wrong where had been a really valuable addition to my engineering competences.
But now I've been burning out. Here's why:
· Interactions with the team and especially C-level management.
— The team was initially pretty diverse with people of all backgrounds, so I didn't feel weird being a foreigner and various other labels of weirdo, but now I'm the odd one out.
— Recently, I've developed severe sensory sensitivity, and communicated to them that since they are not willing to allow remote work, I'd like to work from upstairs (we're renting a two-floor house for an office); they felt it was too distant from the team, but allowed me to work from the kitchen. Now, however, they told me that a permanent move outside of the room where everyone works is not acceptable, in any other area even if it wouldn't inconvenience others, because that may threaten the quality of communication (the theory being that communication is much easier when they can walk up to me in the same room than when they have to open a door and see me in another room). I did explain that it influences my productivity, but they said that basically it's not only my productivity that has to be managed, but other teammates' as well (I'm still doubtful other teammates' productivity would suffer from me being in another room as much as mine suffers from a constant sensory overload).
— There was a time when I showed up with my nails polished red (I'm biologically male and look male), and the CTO personally told me that I would have to wash that off, because what if some day I'm in charge of hiring people and they are uncomfortable, or what if my teammates are uncomfortable ("So if I ask them and everyone is OK with it..." — "No, what if they are uncomfortable but too nice to tell you?")
— All of that makes me feel like there is a certain amount of, uh, restriction of my personal independence / agency, because who the hell cares where I sit or how I paint my nails if I ship quality code fast? And being the only ML engineer in the team, I suspect they may not realize that my leaving would actually be an inconvenience, since there is some learning curve to our infrastructure...
· I'm tired of ML / AI as a field. While I was shipping small and cheap solutions, NLP started to suffer megalomania, and everyone hiring NLP engineers seems to have training gigantic chatgpt-like shit in mind, which is not sexy for me (and I still have no experience with it, but I could figure it out). It seems foolish not to use my tired knowledge to *some* benefit, but I would hope to apply it somewhere where I still see a sense of meaning.
In addition, in these recent years searching for meaning, I have pursued various pet projects. For example, I have created a project called theorem marketplace (link: https://theorem-marketplace.com/) and while doing it, figured out some basic web3 development (because I had to create an Ethereum contract with an oracle). Also, I have periods when I'm in love with molecular biology, so I'm considering barging into bioinformatics. (In pure AI, I find interpretability research a la anthropic very sexy, but I have no credentials to signal I would be a good fit; and maybe I wouldn't).
Hmm, have you considered embedded/system/scientific programming? Those might appeal to you if you like working on lean, efficiency-oriented code (assuming you're not concerned about software devs being replaced with AI in the coming years - personally, I'm very concerned!).
A right career coach will help *guide* your brainstorming on what you want in your next job/career. My opinion is they should not be prescriptive. My recommendation would be to ask the career coach for examples of former clients and how they helped them. You should also be honest on what help you need (eg help with brainstorming, keeping you accountable to a timeline, providing resources).
A right career coach will help you understand your interests, skills, and experiences so you can find the right new job/career for you. In fact, you may feel it is going one step backwards (evaluating those things again) to allow you to move two steps forward (finding your new job/career). But a right career coach will work with your pace and design a program that fits you. They also should have ready access to resources that would take you more time and effort to find on your own.
I’m going through this process myself and would be happy to exchange notes. Less so a dramatic change and more or a correction, but I’ve spent way too much time contemplating it.
I don't have any notes on my end yet, I'm just trying to pinpoint in what general direction I should be moving, but my current state is much more awayness from an unacceptable situation than any towardness with any specific goal in mind.
Perhaps I should have just written a commend with the specifics of my situation, instead of being silly and vague. But that may take another while...
Zvi Moskowitz' substack Don't Worry About the Vase is a good place to start. He gives comprehensive AI progress updates weekly. In them is a section called 'get involved' about open positions. Linguists have a skill set, don't think about this work as programming. Market yourself as a smart person in the language field. Stranger things have happened!
I'm actually quite a good coder (come on, I'm an ML engineer in an ML startup), and I'm sorta aware of AI safety research (I even slightly helped run an AI safety reading group in the country where I am). That just makes me another person with an unimpressive application, no?
I work on the Hill, and just flagging that the AI moratorium on states regulating AI for ten years is likely to pass Congress as part of the Republican reconciliation package. The House version includes a blanket ban. The Senate version of the legislation, which blocks states from accessing broadband funding if they pass AI regulations, just passed the Senate parliamentarian's review. This was unexpected. In the Senate, a provision must be primarily related to federal funding to pass this review, and most experts though this provision would be stripped out.
The Senate will then have to pass the bill (it almost certainly will) and then the House and Senate will have to work out the differences in their package in "conference," and then the final version has to pass both chambers. As things stand right now, this is likely (75%+) to happen.
The time to try to get this provision removed - probably in conference - is now (if you oppose this, I do), as there is some Republican dissent on this provision.
Thank you. I agree this is terrible law, and I don't want to wait around for the courts to have their say. I will ask my rep (Tenney) to fight for states rights on this.
This is something that would make sense to be federal. Any state limiting AI use can cause problems for everyone. But if there was ever one time in the history of everything that it's worth using every tool at your disposal to get what you want, that time is now.
The problem is that it purports to prohibit any state law that regulates AI, but there are a number of legal fields where the power typically belongs to the states and not the feds and in those areas it will be necessary for states to do so. Any state licensing board from doctors and lawyers down to trades might potentially need to regulate AI uses. States will need to create liability schemes for misuse of AI, or products liability schemes for the developers themselves, as without a federal law both the state courts AND the federal courts (when diversity jurisdiction applied) would have to rely on state law for any such civil action. How about a revision to jury instructions that prohibits jurors from using AI during deliberations, is that prohibited here too?
I understand that pro-AI people are afraid of a few states gumming up the works for everybody (and I'd be happy about that as an anti-AI person) but there are a lot of actual practical problems that will require regulation that an impartial legislator would have to concede are state law powers.
>This is something that would make sense to be federal. Any state limiting AI use can cause problems for everyone.
Agreed. I _really_ don't want to see regulation that clobbers American AI like the regulations that clobbered civilian nuclear power. Still, relatively sane regulations that just impose reasonable transparency requirements, like the New York RAISE Act (covered in detail in https://thezvi.substack.com/p/rtfb-the-raise-act ) look like they would be reasonable laws to enact at the federal level.
Thank you for the insight. I'll see if I can contact my reps in time.
To be clear, the thing likely to pass is a moratorium on states regulating AI. So an AI free-for-all is winning, which is bad. We want to call and say we want states to be able to regulate AI, so we want our reps to vote against the moratorium on AI regulation. Correct?
Yes, exactly. This would technically NOT force states to not regulate AI, but if they want to get access to broadband federal funding, they cannot essentially regulate AI for ten years. Since all states want this funding, it's a lot of pressure.
Calling/emailing never hurts, though requesting virtual meetings if you live in a red state/district or your friends do would be the most helpful. And then pulling any connections that you have.
I just called my congresswoman (D), who had an aide answer, and my senators (R), who had voicemail answer. I left the following message, which anyone reading this can copy and optionally add some nuance regarding broadband funding:
Hi, my name is [name] and I'm a constituent from [town, state]. My phone number is [number]. I'm calling about the Republican reconciliation package. Believe it or not, it includes a moratorium on states' ability to regulate AI. Now, I oppose AI because I think it can become dangerous. I want to preserve states’ ability to regulate AI. So, because the Republican reconciliation package includes a moratorium on states’ ability to regulate AI, I'm calling to ask [representative’s title and name] to either vote against it, or to work hard to remove that part. Thank you very much.
I don't really understand who besides Marc Andreessen thinks this is a good idea? Where is this push to preemptively keep away all AI regulation coming from? Even Altman seems interested in baseline regulations, and certainly Musk and Dario do as well.
The military is getting more and more input from new companies like PLTR, Anduril and others. AI is important to them. It’s not a hill I will die on…that was the thinking that drove my impulsive guess.
Did you miss The OpenAI Files? Good summary place to get caught up to speed on the "sama isn't a straight shooter" position.
David Sacks and Jensen Huang would be another two pushing the memento moratorium, at least. I don't think Zvi has a specific standalone post addressing the issue, just regular weekly references to its status, else I'd point there.
Nah, the time to oppose this was before. At this point it's likely too late. I mentioned a while ago during the Biden administration the AI pause people were making pretty obvious mistakes and it would bite them if Republicans got back into power. Now Republicans are back into power and it's biting them. This is a common issue in politics: if you wait until you're in a position of weakness it is almost always too late.
If it makes you feel any better, it's only a ban on AI specific regulations. You can still pass laws that affect AI but they have to apply to AI and non-AI things equally.
Well hopefully if this sneaks through we can get creative and single it out in ways that either include some other software that wouldn't matter, or which could fail to apply to theoretical AI frameworks that aren't neural nets and which nobody is using. "All software distributors must be able to document how their product actually works" , "software containing non-algorithmic outputs in user interfaces shall create strict liability upon the developer for damages proximately caused by the output".
I still think you'll end up wrong about the political valence of this, the populist RW isn't having any of this AI stuff and the Republicans have to care about more than the Chamber of Commerce these days.
I'm sure this will be adjudicated but disparate impact is well understood. What would work is if they're willing to burden the entire software industry generally. For example, if they're willing to force Google search to exist under the same restrictions at ChatGPT.
If you think the Chamber of Commerce is the ones pushing for less AI regulation then I'm sure you don't have a handle on the political economy of this. It's not the populist RW either but they're also DEEPLY suspicious of regulation of AI since they believe it's being weaponized against them. Which was another thing I thought was a long term mistake by the mostly left leaning AI regulation crowd.
I'm seeing both my state's GOP senators criticize it (Hawley strongly, Schmitt somewhat obliquely as he seems to have some broader goal on AI regulation), as well as MTG and Luna coming out against it from the House. I was surprised the parliamentarian wasn't quietly encouraged to just toss it out to avoid any fight when it gets back to the House, as they have a razor thin margin there and the RW populist opposition is sufficiently eccentric and out on their own island that they may be willing to tank the bill over it.
there is a lot of private concern/dissent... but Republicans tend to fall in line to just pass the package. Still a decent chance to get it stripped in conference
Participating in a review contest is kind of rough: there are a lot more good entries than finalist spots, or even finalists + honorable mentions. Lots of people write something interesting and worthwhile and get nothing for all their work, sometimes not even feedback. To help with that I'm going to award my special Tasty's Consolations Prize, for all the reviews that I liked that didn't make it into the finals (or got honorably mentioned). The winners get nothing except for dubious bragging rights and a warm feeling of being appreciated. I read "only" 50% of all the reviews, so I obviously must have missed lots of worthy entries.
It's clever and it showed me new things and it featured Actual Poetry (the kind that adheres to some predetermined structure of rhymes or meter), what more can I ask.
Enjoyed the sestina so much, I got close to memorizing it.
I should say though: I think the review doesn't do the villanelle justice - the example villanelle really does seem rigid and repetitive and one couldn't blame you for reading and concluding that it's a very restrictive and stilted form based on that example. But how different it can sound in the hands of a great poet! https://allpoetry.com/If-I-Could-Tell-You And of course, the rage-rage-yadda-yadda-something-something-light, is also an example of a villanelle. Maybe it's unfair to unfavorably compare the Reviewer to W. H. Auden, but someone has to defend the honor of the villanelle.
Two celebrated pieces of media, that the reviewer likes on the artistic level both seem to come close to saying the same thing the reviewers pushes back against: "trying to make things better will always inevitably backfire and make everything worse". I'm a little hesitant to praise this review, because it kind of says nothing new that I didn't already agree with.
This is a story of the Reviewers tragic romance. It's exactly the same as a million other such stories, but it doesn't make it less painful to the person it's happening to. Reviewer, I'm sorry it happened to you. I wish you well.
Since he was a kid, the Reviewer has been spinning in his mind an elaborate, anime-like/Marvel-like fantasy world with an extremely convoluted story and worldbuilding, and kept spinning it into adulthood. It's not that this is great art that is going to become next Lord of the Rings or Star Wars, but it seems like there's something worthwhile in anything a person put so much thought into.
As a kid, I had an imaginary world like that too. It seems like this is fairly common. When someone grows up to become a writer, it makes for a good back-story - "at the age of 5, so-and-so was already writing epic fantasy", except that this isn't unique to writers - seemingly lots of people did. Then later we all thought there's no point to it if you're not going to write it down, and writing down is a ton of work, so almost everyone gave up on it, while the Reviewer kept going on and on. Is there a point? Some part of me wants to say this kind of escapism must be unhealthy somehow, or at lest a waste of time, while another part is delighted end sort of envious.
A review of a famously difficult and frustration-inducing game, which is known for constantly destroying your progress and bringing you back to square 1. It is a gimmicky review, but a good one. I'm a little surprised this one didn't make it into the finals. It is also half-autobiography? Yes. It made me want to get Getting Over it and play it solely as a masochistic character building exercise. So far I haven't.
So this is about a clade of songs sharing the same pattern: a dying man tells the narrator a story of how he ended up that way. The pattern changes, mutates, develops pseudopods, plays as a serious tragedy here, as a joke there. The review is a pretty serious analysis, and long. Perhaps it suffers somewhat from not having any kind of grand unifying idea at the end, and it also suffers a little from not including youtube links to the actual songs, for the copyright reasons, I imagine. I suspect lots of readers didn't finish this one.
I enjoyed it for an opportunity to see song evolution in progress and for all the interesting music I got to listen. My favourite of the bunch was The Unfortunate Rake https://www.youtube.com/watch?v=1aQd99zfIRo, an entirely serious and heart-breakingly sad ballad, about the hero dying from syphilis.
A review of a Ghibli non-Miyazaki cartoon, adapting a Japanese folk tale. A tale is simple, but serious and sad, and it should provide some food for thought. The Reviewer gets a very gloomy message out of it, that for everyone there's a thing that would be good to do, that would be right and beneficial to do, but that goes so much against our nature, that we'll probably never get ourselves to go ahead and do it. Unfortunately, this rings true to me, and truer than I would have liked. Possibly the review would have done better if the plot summary section had been shorter. On the one hand, the plot summary was essential for the point the Reviewer wanted to make, on the other hand I could imagine readers getting bored there. But it is a review that stuck with me all the same.
Thank you for delving into a dark and terrible subject in the name of anthropology. I don't have you courage and won't be watching the series any time soon, but this makes your investigation all the more valuable. You are the true mythic hero for going where others dare not go, for staring into the abyss, for descending into the cave that is also the underworld somehow. I now understand modern kids slightly better thanks to your efforts.
Hehe, you know you've got a good hobby when it makes other people simultaneously delighted and worried about your health. I wrote the Imagination review, glad you found it worth engaging with! If it eases your worries somewhat, it's not like my life revolves around this fictional universe, it just sort of sits in the background most of the time, and I'll occasionally add a few details when one of life's natural lulls comes my way, or my creative juices are flowing. I suppose the point is to do something that is intellectually stimulating, and creatively rewarding in the long term. Storycrafting is a surprisingly entertaining way to pass the time, I definitely recommend it.
In other news, kudos for making it through ~50% of the reviews, that is far above what I was able to achieve. Definite agree on the Rubbermaid review, there's nothing quite as entertaining as a person's hot blooded passion for mundane objects. Really captured the spirit of the contest in a way that few other reviews managed.
Thanks for your kind words re: Skibidi Toilet! I don't think it was THAT brave of me - like I wrote, it got surprisingly engrossing quickly - but I always appreciate being called a true mythic hero so don't let me stop you. dop dop dop yes yes
I just looked at the Rubbermaid review, prompted by your mention. I hope the reviewer is old enough to have experienced the Before - not just Before Amazon, but Before the grocery store had a large display of brushes, mops, and brooms - when the Fuller Brush Man (happy day) came to your door, and gave you a little “extra” if you made a purchase.
I wrote the Deathbed Ballads review. I'm glad you liked it, and I agree with most of your criticisms of it. The length and complexity snuck up on me while writing it. I was about a third of the way through before I realized how much serious scholarship there was on the topic (especially for the Unfortunate Lad and St. James Infirmary) that I needed to engage with in order to treat the topic fairly. I was particularly handicapped by two sources I really wanted to dig into being unavailable on any reasonable terms: one was an academic paper that was behind an extremely steep paywall and the other was a then-temporarily-out-of-print book that I didn't have time to buy and read in time to use by the time I realized I needed it.
If I'd had another couple weeks to work on it, I would have gone back, strengthened the conclusion, and revised the introduction and body of the review to work towards that. And I would have needed to do some additional background research to support the conclusion. The conclusion I probably would have worked towards was further emphasis on the importance of syphilis to the song clade and the significance of the disease to the times, places, and communities where the core elements of the song cropped up: first in big metropolitan commerce hubs like Victorian London, and again in the American black community during the interwar years. In the latter case, there was a big and now-little-remembered epidemic of syphilis that probably had its roots in the American expeditionary force to France in World War One, and the epidemic persisted and spread among a discriminated-against minority community due in large part to marginalization, poverty, lack of access to diagnosis, neglect by public health authorities, and lack of good treatments even for people with access to medical care. I would have liked to have explored that in more detail and connected it to St. James Infirmary, and also explored parallels between the interwar syphilis epidemic and the AIDS epidemic in the gay community in the late 20th century. I only touched on pieces of this, partly because I was out of time and would have needed to do a lot more research to treat it fairly, and partly because the review was already too long.
Another piece I started doing but abandoned was hypothetical cladeogram of the song genre. I gave up on it because it wasn't coming together visually, I was out of time, and it was extremely speculative to begin with and would have needed a ton of explication and qualification to not be misleading.
Part of my problem, timing-wise, was that I started out intending my topic to be an analysis of the lyrical content of the music of the American Civil War as a corpus of data on cultural attitudes about the war's conduct and causes. I abandoned that and switched to what I hoped would be a lighter topic once I found that very quickly becoming too research-heavy. When my "lighter" topic also proved to be too heavy-weight, I had even less time to finish it.
I went back and forth on the question of including youtube links at the top of each section, and might have made the wrong call there. The advantage of including them is obvious. I decided against links or embeds (except for a couple of footnotes where youtube was the best source for an obscure-but-important variant) because of a combination of copyright concerns, clutter concerns (especially for St. James Infirmary, where there are so many major variations), and worries about how the formats of both google docs and substack would handle embedded youtube videos.
I wrote the Getting Over It review--thanks for the kind words! Nice to see a few comments about it here and there to make the hours put into it feel worthwhile. Hope you have a good time (or enlighteningly bad time) trying to get over it ;)
Hopefully this isn't against the spirit because finalists have already been chosen, but since you opened a discussion here, I'll just follow up that I really disliked the review of The Tale of Princess Kaguya. I think the reviewer misreads the story so badly (possibly due to missing cultural context, possibly due to the source material itself, i.e. the original folk tale, being somewhat inconsistent so that the movie makers' efforts to harmonize the themes made it quite subtle and complex) that they had to project an entirely orthogonal point onto it, and I have half a mind to write a counter-review.
(Interestingly I just checked my voting records and I actually gave it a 7, which was above average on my scale. Thinking back on it, I probably still thought that the essay was well-written and its points are well-made, despite missing the mark as a review of the movie per se.)
I wrote it. Open to discussing this, if you're interested. If you feel a need to correct me -- which I definitely get, I feel the need to correct people if they give interpretations of art I don't like -- well I promise I'll read your take.
As a preliminary note, I don't have much background knowledge, neither about the time period nor about anime in general. In my mind this doesn't matter because I've made very few claims about what the movie is trying to do, and more just described what I got out of it. (With some exceptions.) I would be extremely surprised if Takahata had thought about anything that I'm talking about in section IX, for example, but I don't think that matters.
Edit: there was some discussion under the Voting post about what to do with reviews that aren't really reviews, whether to penalize (and I somewhat self-servingly responded that I'd vote based on how much I liked them, with no penalty). The truth is that I immediately thought of my essay as in that category. I never approached it as, "let me tell you what this movie is about", at all. But as was discussed before, this was a "non-book review" contest, rather than a "non-[book-review]" contest, so it's totally legitimate to take issue with that. It just didn't occur to me that anyone would care about it being a "proper" review until after the deadline. (I also titled my essay "Little Floating Bamboo" rather than "Review of the Tale of the Princess Kaguya", but Scott changed all the essays to just have the name of the thing they're reviewing, which, in retrospect, was a pretty obvious move to make things less confusing.)
Hello! I'd love to discuss this, thanks for following up! Reading back my comment now, I realize it probably read a bit dismissive. I apologize for that.
In the end, I largely agree with your self-assessment. I think your review is a good essay, in that it makes a point (about doing Genuinely Hard Things) relevant to the reader and defends it clearly. I even agree that "I would be extremely surprised if Takahata had thought about anything that I'm talking about in section IX" and that those points deserve to be made. But I think it does matter that Takahata had so little to do with it, and that is the main reason why I "really disliked the review".
Takahata clearly made the film with his own intended themes, and I would have liked to see a review that engaged with that more, especially because the film made a valiant effort to wrangle the source material into a coherent film for a modern audience, but made it just a bit too subtle and complicated, and in spite of that I really love it.
Allow me some time to compose a full response, because as it turns out I have a lot to say.
Thank you for your review/essay (I thought it fit comfortably in the "review" category, which is broad)! First, for encouraging me to watch the movie before reading your review; I'd seen most Ghibli output, but not that one, and it was time well spent. Second, because it was an entertaining read. Third, for including the source material (Japanese); although I don't read Japanese, doing so spoke to your commitment to the original rather than to a translator's gloss.
As with C.A., what I took as the theme of the movie was completely different from what you took from it. Which in and of itself is fine; great art ought to speak to different people in different ways. But the review left me unfulfilled because I didn't see your message in the film and your review didn't make a strong enough case to convince me that the message was there.
I have no idea whether what I got from the film was the (or an) intended message either. For me, it was a meditation on happiness.
So many people wanted to make the Princess happy, or claimed to want to make her happy. This starts with the bamboo cutter, then the bamboo cutter's wife nearly rips the baby out of his hands because (as she says) only she can make the baby happy. Furthermore, they didn't want to make her happy because happiness was a good thing, they wanted to make her happy because making her happy would in turn make them happy. And making her happy was a form of control, of possession.
Is trying to make people happy a bad thing? Yes, the movie said to me, and the problem is not the "happy" part but the "make" part.
The Princess, in turn, does not find happiness in anything others do that's supposed to make her happy. Her friend growing up in the village didn't consciously try, and she does find happiness there with him. Her flight with him at the end is her imagining happiness with him, but to do so she abducts him away from his wife and child, so the happiness she envisions together is no better for him than the happiness others imposed on her.
Finally, the arrival of the Moon people is accompanied by visions of effortless happiness. And while she returns to that, she is giving up the life on Earth where happiness is so hard to come by but so much more precious because of it.
Well -- I said in the review that she would've been happy in the forest village but not in the capitol, and that her father never succeeded to make her happy.
I didn't talk about intentionality, and it is true that Sutemaru is the only one who didn't try. But that's not the reason he succeeded. He succeeded because he didn't put her on a pedestal, whereas everyone else did (which I did talk about!). You could absolutely make her happy by choice, you'd just have to understand what she wants. (Like you, idk what Takahata thought, but if he came out in an interview and said, the point of the movie is that making a girl (or people in general?) happy is only possible if you don't try, then I'd just say, well, he's wrong ¯\_(ツ)_/¯)
I don't agree on the 'kaguya abducts sutemaru from his family' point. She technically does for this one sequence. But she wants to live in the village with him. I don't think there's any indication that he'd have to leave his family for her. I mentioned in the review how their relationship is explicitly suggested to be non-romantic, so it doesn't conflict with him having a wife and kids.
(And her mother I think did succeed. She didn't succeed all the way because she can't because she's a mother, but she made her life much more pleasant than it would have otherwise been. She did about all she could.)
A simple proposal to make the world (well the US) a slightly better place:
Congress passes a law, imposing a tax & paperwork/compliance requirements on any American company, who serves GPDR/cookie warnings to an American IP address. If you fall into that category, you have to pay some sort of annual tax, and there's Deliberately Annoying Paperwork that you have to file with the government. The point is to harass these companies and make them stop serving cookie warning nag screens to Americans. My personal view is that the GPDR has absolutely ruined the Internet and made the world a worse place.
The law does not prevent companies outside the US from having cookie warnings. If a German website feels like it needs to comply with the GPDR when Americans visit, OK no problem, you're still free to do that. It does not prevent American companies from complying with the GPDR and having cookie warnings when European IP addresses visit their websites. If Disney needs to comply with the GPDR in Europe, again, no problem.
The idea behind the law is to counteract the Brussels effect as it relates to purely internal US web traffic. Millions of American websites harass US citizens with cookie nag screens, even though the GPDR obviously does not legally apply to us. I'm guessing everyone's in-house counsel just told them it was easier to comply. Now, we can change the incentive structure and make it easier for them *not* to comply with Europe's insane laws. Constitutionally the US cannot forbid American websites from having cookie nag screens, but we can certainly make it difficult..... American sovereignty will be restored, and all of us will stop being annoyed by obnoxious European laws :)
Imma make a killing selling a special device that cuts the top of bottle caps off again. (Has that EUism reached burger land or is it just us who have to suffer it?)
The directive also addresses other single-use plastics, with measures like bans on plastic cutlery and increased recycled content requirements for plastic bottles.
Glad I'm in the USA. I often find it convenient to use plastic cutlery to eat off non-stick surfaces without scratching them.
I've heard that one problem with cookie warnings is that they (well, in many implementations) make the internet harder to use for blind and otherwise vision-impaired folks who use screen readers because they steal focus and make sites harder to navigate (and are frequently hard to close from a screen reader, especially if there is a gazillion options). I wish someone had sufficient political weight to argue cookie warnings out of existence because they destroy accessibility.
Cookie warnings are malicious compliance. GDPR does not require anything like what has been implemented. Everyone should pitch in to punish the perpetrators, not just Americans.
First of all, you're a freak and a weirdo (unlike everyone else on this website! Especially me!) if you don't have cookies generally blocked and use per-site permissions in a browser which implements these to strictly regulate what sites can set cookies at all on an opt-in basis.
Secondly, any decent adblocker will have an annoyances filter which strips the cookie warnings. It takes seconds to add this filter and you should be using an adblocker anyway if you like to browse a usable internet.
1. Some websites (thankfully a minority) are completely locked until you agree to their terms & conditions which include cookies. Adblock can't fix that
2. I use two adblockers, the more powerful/useful one is Ublock Origin, which Google is in the process of removing. I already lost it last week when one of my computers updated, and I only have it on this current browser because I've blocked updates so far
"Some websites (thankfully a minority) are completely locked until you agree to their terms & conditions which include cookies. Adblock can't fix that"
I've never encountered this personally. What websites?
"Ublock Origin, which Google is in the process of removing"
Wait, what? Filthy, if true. Move to Firefox, is my answer to that.
Have you tried the Lite version of the extension? What features do you find lacking? I understand there are some, but I don't seem to suffer from their absence.
You have now made me stand up and cheer for GDPR. Oh gosh, the poor Americans have to click through a warning screen? Welcome to the rest of the world, which gets the Americanised version of everything shoved down our necks. Sorry, the calendar is set to American date format, you'll have to change it. Paper sizes? American. Spelling? Three guesses, and the first two don't count. Default language? English - US of course!
Turn about is fair play, let you Yanks suffer a taste of what we routinely go through.
I think there's a significant difference between "we happen to have most businesses based in our country, due to the strength of our economy, and they naturally reflect our preferences ", and "we're throwing tantrems and the threat of government coercion to try to force businesses *not* based in our country to adopt our preferences" actually.
Can you walk me through the paper size frustrations? Do you have a large stockpile of metric unit paper bought for a metric unit printer misguidedly built for the European market that is now incompatible once that effort was abandoned? Is there a need for your notebooks to conform to A4. Does A4 not make good paper airplanes because your 30cm straight edge doesnt line up with it correctly? I am struggling to see the use case where its inconvenient for you.
I’m largely pro European but the EU does itself no favours with GDPR. I feel if you got some monkeys together, smashed them over the head with a hammer a few dozen times, threw them off a cliff, and finished up by getting them drunk on fermented banana juice before giving them all a few typewriters to play with you might get this kind of output.
I found out a few days ago I may have to appoint, or appoint myself, as a data controller because my doorbell camera might see a small part of the local village, although I’d be well within my rights to take pictures while in the village.
That kind of nonsense is bad law, as is any kind of law that can be interpreted in a fashion the writer didn’t conceive of.
I mean, I'm as annoyed about American Defaultism as the next European, but I also find cookie warnings incredibly annoying and pointless and would be in favour of any efforts that might make them less ubiquitous.
Depends on what browser and OS you're using, but if it supports extensions at all, it probably lets you use this one.
I use a Firefox fork (Iceraven) on Android, and I think it actually suggested I use this over the older version "I don't care about cookies." If you prefer a Chromium-based one, Kiwi Browser has been recently discontinued, but Edge Canary should work.
I don't want my country- the world hegemon- to bow down to privacy laws implemented in Europe. The EU is free to regulate the EU, but they can't regulate America. I find the whole idea ridiculous. It's a basic sovereignty issue
Oh, I agree completely, and think the US should go much further and threaten to not honor NATO defense agreements as a way to reshape EU laws and regulations particularly repellant to core "American values."
But making "the world (well the US) a slightly better place" is a longer-term thing. You can fix your specific cookie issue in less than a minute.
Of course it would. My proposed policy is that America extend its practice of infringing on the sovereignty of lesser territories to our enemies in Europe, the greatest threat to our freedoms.
"Am I the only one who thinks being able to reject non-essential cookies is actually a useful feature?"
You are not. But getting a *bunch* of dialogs to do this can be really annoying. And you get a bunch of dialogs even if you check "yeah, sure, whatever." Is the annoyance worth the usefulness? Not clear ... especially if the servers just get more sophisticated in how they track you (which is my guess ...)
The thing is, not only GDPR does *not* require the bunch of dialogs, but displaying those dialogs is actually *against* GDPR, which says that denying all cookies should be equally easy as accepting all cookies.
So what you see is not even malicious compliance, but more like malicious pretense of compliance. And your reaction is a proof that this type of corporate propaganda works perfectly.
It is also not clear to me whether the annoyance is worth it, but it very well might? I'm honestly not sure.
It's not just that we can reject cookies now, there's also the side effect that websites are less likely to implement unnecessary tracking cookies in the first place since it now involves a lot more effort.
That said, like the OP, for the sake of this discussion I'm more intersted in discussing what the EU and US laws should be, not so much whether the laws can be worked around by savvy users.
What is your estimate of the cost of Israel’s two-week campaign against Iran?
According to The Economist, from October 7, 2023, up until just before the attack on Iran, Israel had spent approximately $85 billion, or roughly $150 million per day. I estimate that during the campaign against Iran, this daily rate increased to $285 million, due to the use of more sophisticated munitions, advanced military planning, logistical demands, covert ground operations within Iran, and costlier manned and unmanned aerial missions.
Based on that, my ballpark estimate for a two-week campaign likely cost close to $4 billion focused on Iran alone. Would be great to hear your thoughts.
I think this is extremely hard even to estimate it, since damages from Iran's missile strikes are top secret. Or, if you prefer it this way, Israel denies all hits except those it cannot plus those which have some PR advantage.
Me too. I know it was silly source material, but the point of the contest is that you can review any kind of silly thing you want to. The quality of the review was pretty high, imo.
Scott, thanks for giving us extra time to read the entrants. I am so mad at myself that I fell Ill and couldn't make use of the extra time. I found it much harder to rate these than the book reviews.
What were people's strategies for rating things that were clearly very well written but not your interest ? Or things that were to your interest that you really enjoyed but were way shallower than other entries ?
Just wanna know if someone else figured out a good rating algorithm. I have been self-conscious about this after realising I was basically the only person who was rating Kindle books on a bell curve and inadvertently ruining the life of every author I was rating 3* (which was the majority).
First I sorted all the reviews in order of most to least favourite. (I compared two reviews side-by-side, and decided which one I liked more, and then repeated that. Specifically I did it by emulating the Selection Sort Algorithm)
Afterwards, I read through my list from top to bottom and asked myself whether the reviews were similar in quality or if the latest review I was reading had a noticeable drop in quality relative to the last one I was reading. If there was, I marked this drop with a dash.
All reviews in between two dashes form what I call a “grouping”. All reviews within a grouping ended up being given similar numerical rating to one another.
My system worked out in such a way that I ended up producing 8 different groupings, which corresponded pretty closely to the 10 point scale. All ratings that were 8 or higher were confined solely within the first/highest grouping.
In terms of metrics, the only one I considered was how much I personally liked the review (relative to the others). I feel like it’s a review’s job to make me interested in its subject. If it doesn’t do so, then it is a point against it.
I know this is biased toward poorly written reviews of interesting subjects scoring better than well-written reviews of banal ones, but let’s face it, banal subjects have no chance in this contest (RIP Skibidi toilet). The only things that are going to make it to finalist anyway are well-written reviews of interesting subjects.
At least that’s how I justified it when I was designing this system. The fact that the mashed potatoes review somehow ended up as a finalist makes me very interested in learning what the author did to convince people to give it such high ratings. I await the official ACX post of it with interest.
I love reading how people take their voting very seriously and have all kinds of systems to ensure fairness and consistency. God bless all your honourable little hearts!
I just go "I liked this, eight out of ten!" Anything more sophisticated I might engage in is akin to the Eurovision voting (i.e. "this topic is one I dislike so good luck getting anything higher than a three, mate" or "ooh this is one of our friends and allies and neighbours, douze points!")
For the sake of calibration, I didn't submit my ratings until I'd done 42 of them (I ultimately did 74 in all). I wound up assigning qualitative descriptions to each rating value; 1 was "I award you no points and may God have mercy on your soul", 6 was "replacement level", 10 was "deserves to win it all". My final distribution was (from 1 to 10) 1, 4, 6, 13, 8, 18, 11, 7, 5, 1.
I think I was a pretty rough rater: normed to a bell curve (ask me about my awesome spreadsheet) and penalized the ones I wasn’t interested in.
(I mean, I figured that interest has to matter, right? We want to read reviews that we want to read, not just ones that are written nicely. So if everyone rates opinionatedly, then the most generally interesting ones will do well and we’ll all be happier.)
I definitely punished shallowness (and shortness) in general: I like reviews that take up half my day to read and think about; ones that tell me a new story and teach me something about it. In the end, I think this mostly meant poorer ratings for video game, music, and tv/movie reviews, which I’m not sure how to feel about. On the one hand, it might’ve been good to see more content-type diversity among finalists; on the other, again, if we all prefer lengthy reviews of weird quirky things, why waste time and space?
We should try sortition. (link: https://archive.is/ztIRY) In the current US system, there's no way to get the best/smartest/most virtuous/etc. people on the ballot, only those with cash and connections, and how they got those usually makes them least fit to serve/lead. True representatives should be closer to average or median. We have eg Congressional Research Service that can assist. Less corruption, less extremism, less Peter principle. Maintain voting for initiatives/referendums/etc that are more important and less manipulable (as far as, ideas aren't people, so less distortion from social psychology). Biggest problem is getting it past the current system. Something something trading freedom of "we the people" for security of "politicians". More links: https://www.reddit.com/r/todayilearned/comments/1j91das/comment/mhamcg1/)
This problem was already solved a hundred years ago in The Napoleon of Notting Hill. You simply choose the new absolute monarch via random ballot of all adult men.
EDIT: I see Deiseach has predictably beat me to the punch! Damn you, potato-American!
Why do you think it is that we have not really had a sortition system of governance (outside of juries etc) since the 19th century? Why did the places that used to use sortition (such as some parts of Switzerland) cease to do so?
I'm a fan of sortition, but I'll offer a discussion of flaws that I see in it, especially as applied to a modern large liberal democracy like the US.
High variance is a big issue. A 500-person legislature (a bit larger than the current US House of Representatives, a bit smaller than the British House of Commons, and exactly the size of the Boule of Classical Athens) chosen by sortition from any nontrivial eligible population would represent the preferences of the population as a whole with a margin of error of +/-4% if chosen at large. That's probably okay if most important issues are 60/40 or worse, especially in situations (like Athens and most other Mediterranean classical democracies) where there's a rule or norm that major or contentious issues get referred to direct-democracy institutions for ratification. But for anything close to 50/50 the margin of error is going to be uncomfortably significant, especially if the legislature is the final word on the subject.
Athens also used a system where the citizens were divided into ten phyles (sometime translated as "tribes"), each of which got fifty seats in the legislature by sortition. I'd initially assumed that the phyles were class-based like the tribes of the Roman Republic, but what little I can find about them suggests they were more in the nature of geographical districts. Either way, a lot of issues would likely see each phyle having relatively homogenous interests and policy preferences, so on those issues at least the apportionment by phyle would probably make the composition of the legislature more representative and stable.
The modern US would have trouble copying this in a robust way, since we're big enough and non-homogenous enough that our equivalent of phyles would be either too numerous or too internally heterogenous. The obvious counterpart to geographical phyles would be states, which are both, especially if you try to apportion seats by population. One representative chosen at random from Montana would be a terrible sample, while 52 represenatives chosen at random from California could skew significantly depending on how many happened to be chosen from the Bay Area, or the greater LA area, or the Central Valley, or from several other regions with distinct interests and political distributions. My understanding is that most medium states have significant regional and urban/rural divides which would produce very unstable results among the dozen or so representatives those states have.
There are other problems with scaling sortition to a larger population. Classical Athens used sortition to select a 500-person legislature out of maybe 50,000 male citizens. The eligible population was probably quite a bit less than that, since I'm pretty sure the 50k figure includes children and young adults below the minimum age of 30. So 1-2% of the population was in the legislature in a given year. Combined with short terms (one year), this means that somewhere between 30-60% of citizen men who don't die young would probably be chosen at least once in their lifetimes. A bit more or less for any given citizen, depending on how malapportioned the tribes were. And the phyles would have averaged about 5k male citizens each, small enough that you're fairly likely to have at least one current member of your legislature in you immediate social circle and virtually certain to have a friend or relative as a member of the legislature at some point over a period of several years. In the US, a 500-person legislature would be 1/500,000 of the adult citizen population, so your chance of ever serving in Congress in your lifetime (assuming Athens-like annual terms) would be about 0.0112% and your odds of knowing a member of Congress personally wouldn't be much better.
There's also the problem of legibility and trust, which is compounded when the fraction of the population being chosen is so much smaller. For a population of maybe 3k eligible candidates per phyle, it's pretty easy to design a procedure where a large fraction of them or even all of them can observe or even participate in the drawing. I don't know how they did it historically, but having eligible candidates line up and draw colored balls from a bag, or write their own names on something (not sure what: I know Athenians used potshards as ballots in other contexts, but they don't seem mechanically suitable here) and toss it into a pot that would be stirred and drawn from publicly by a group of respectable local citizens. In a population the size of the US, you could do it fairly with any of a number of lottery systems, but it would be a lot harder for individual citizens to be able to observe that closely enough to overcome suspicion that a sinister cabal was fixing the drawing.
The last problem with sortition is that elected representatives are supposed to be better than the average citizen at legislating. The ideal espoused in early modern England was "Not everyone in Parliament should be there, but almost everyone who should be there is". And in the early US, both parties expected this to happen in some form: Federalists expected voters to choose existing established pillars of the community (usually major landowners, successful businessmen, or respected civic leaders), while Jeffersonian Republicans hoped that voters would learn to discern a "natural aristocracy" of people with the requisite talents and virtues. Most Americans, if asked for a gut feel on the subject, would be highly skeptical of this in general but moderately likely to have a higher opinion of their own current representative. The latter is probably more significant, since if some random guy is preferable to your own representative, there's usually at least one or two reasonable approximations thereof on at least the primary election ballot.
Just found a possible solution to the lottery problem: Have 100 different entities in the country, preferable universities with large engineering departments each make a mechanical random number generator. Have them publish articles about hiw they work. On the day of the draw, all 100 machines will generate a random number at exactly the same time, and immediatly post it publicly. All these numbers are fed as input to a big, unpredictable, but completely open source function, which gives the final number, determining the selected person. The function is made in a way so even if you controlled 99 of the inputs the final output would still be unpredictable.
This system would be much safer than normal voting. The only way to hack it would be to have 100 groups of nerdy engineers from all over the country sabotage their beautiful random number generators at the same time.
Realistically if you had sortition then the next thing that would happen is a permanent political class of "assistants" who show up to help our naive temporary representatives with their duties. Some of the assistants will be Republicans, some of them will be Democrats, and there'll be a bunch who are "independents" (and just happen to be on the payroll of Facebook or Chevron or whatever). They'll be just itching to take all the hard work of legislating off your hands, and they totally definitely promise to let you make all your own decisions on anything important. Or you could be an assistantless representative, which will go just about as well as being a lawyerless client.
I'm speculating, but in a sortition system there will surely arise some kind of class of people who manage to become the true power behind the throne.
As I understand it, this is almost exactly how most lobbying works under the current system. Members of Congress hire their own staffs, but the budgets are relatively small, the pay scales stingy, and the staff also needs to handle a ton of constituent services and media relations work in addition to legislative analysis. Lobbyists show up offering their analysis of pending legislation and helpful suggestions for revising it, which just so happens to be to the benefit of their clients (those clients might be businesses, nonprofits, advocacy groups, labor unions, foreign governments, etc), and the workload facing the legislators and their staffs is high enough that despite the obvious conflicts of interests, they tend to rely on input from lobbyists the same way and for the same reasons that judges rely on input from the lawyers in the cases before them.
This can work out okay if there are good lobbyists representing sides whose positions bracket the truth on any given issue, and if the legislators and their staffers are reasonably competent and diligent at identifying which lobbyists are moderate enough in their dishonesty to be worth listening to. How well it works in practice probably varies quite a bit, especially to the extent that there are no doubt groups that don't have good (or any) lobbyists and whose interests thus don't get recommended to the considerations of the legislators and their staffs.
Anything close to 50/50 doesn't matter anyway. Either way, about half the population is for it and half is against it. We have much bigger problems from things like some people not bothering to vote, or in the US, all the electoral college nonsense.
Permit me to quote "The Napoleon of Notting Hill":
"The old gentleman opened his eyes with some surprise.
"Are you, then," he said, "no longer a democracy in England?"
Barker laughed.
"The situation invites paradox," he said. "We are, in a sense, the purest democracy. We have become a despotism. Have you not noticed how continually in history democracy becomes despotism? People call it the decay of democracy. It is simply its fulfilment. Why take the trouble to number and register and enfranchise all the innumerable John Robinsons, when you can take one John Robinson with the same intellect or lack of intellect as all the rest, and have done with it? The old idealistic republicans used to found democracy on the idea that all men were equally intelligent. Believe me, the sane and enduring democracy is founded on the fact that all men are equally idiotic. Why should we not choose out of them one as much as another. All that we want for Government is a man not criminal and insane, who can rapidly look over some petitions and sign some proclamations. To think what time was wasted in arguing about the House of Lords, Tories saying it ought to be preserved because it was clever, and Radicals saying it ought to be destroyed because it was stupid, and all the time no one saw that it was right because it was stupid, because that chance mob of ordinary men thrown there by accident of blood, were a great democratic protest against the Lower House, against the eternal insolence of the aristocracy of talents. We have established now in England, the thing towards which all systems have dimly groped, the dull popular despotism without illusions. We want one man at the head of our State, not because he is brilliant or virtuous, but because he is one man and not a chattering crowd. To avoid the possible chance of hereditary diseases or such things, we have abandoned hereditary monarchy. The King of England is chosen like a juryman upon an official rotation list. Beyond that the whole system is quietly despotic, and we have not found it raise a murmur."
Hard agree. My only modification would be that I think we need some degree of selection, so I would preselect for each seat say 8 people (could be any number) with sortition, then have a first vote two weeks later. The people in question can also excuse themselves if eg they are planning to leave the country, are unwell, are facing trial, or (and I hate this category) are indispensable at work (small businesses might disproportionately benefit from this which would be really unfortunate).
After the first vote, are retained those candidates who received from than 20% of the vote, but at least 3.
Those 3/4 candidates then have two more weeks in which to make their case (there could be pre-booked public or even in some cases local tv/radio events for this) and there is a final vote.
My toy idea would be to impanel some three-ish juries per district. Each jury would be empowered to interview candidates; subpoena records, witnesses, and expert testimony; and report out a nomination of maybe 1-3 candidates who should appear on the general election ballot. Or zero, if the jury fails to reach a supermajority to report out a slate of nominees. The general election ballot would then be based on candidates who are nominated by at least one jury, with some kind of runoff or ranked-preference voting to accommodate cases where the juries report out larger slates with little or no overlap.
Might be a stupid question but: why do we need bunker busters for underground facilities, instead of just bombing all the exits? Is a destroyed evevator shaft much easier to repair than I would imagine?
Yes, an elevator shaft or tunnel entrance is far, far easier to repair than an enrichment centerfuge cascade. If the strikes actually managed to directly hit the enrichment hall and most of the uranium Iran had was in those centerfuges, then they have likely suffered a multi-year or even multi-decade setback. Whether or not that actually happened remains to be seen.
"Is a destroyed evevator shaft much easier to repair than I would imagine?"
It is easier and cheaper to excavate a new hole/entrance than it is to repair a mostly blown up facility (especially if the facility has a lot of expensive equipment).
The things we're looking to destroy are extremely valuable and hard to build; it would be extremely worth it to re-excavate a nuclear enrichment facility.
I was thinking there would be an obvious threat of "if you try to dig another entrance we'll bomb it immediately", but you're probably right, for various reasons.
I understand why you thought that because it's something I thought of as well, but imagine Iran unearths the site in like 5 years when the geopolitical situation has changed and now the US is not capable of destroying it.
"...and now the US is not capable of destroying it"
Or even just unwilling.
I expect that part of Trump's calculation on what to do is considering what will happen in the next US president has a very different view of how dangerous Iran is. This would lead to choices that do NOT require follow through by future presidents.
Can a person be wrong about what their motivation or beliefs are? I don’t mean intentionally deceptive - I mean, actually being wrong in their own self-assessments. Why or why not? What evidence can a person give that their motivation or beliefs are what they think they are?
My hunch would be that anyone who says they are only motivated by pursuit of the truth is lying to themselves without realizing it, because doing so gives them social rewards and confidence, whereas truth on something you can’t directly act on has almost zero instrumental effect on pursuit of survival and thriving goals, which are terminal and therefore higher priority than satisfying curiosity.
> As a proud person of TESCREAL, I couldn’t help but notice Cofnas’s model conflicts with one of the stylized facts of our traditional ways of sociology: that people generally form their beliefs not from an analytical weighing of relevant issues but based on what will help them as social creatures. I won't defend this with serious social science here because, one, I am a high theorist and it wouldn't help me as a social creature, and two, Robin Hanson and Kevin Simler do that in their magisterial 2017 classic, The Elephant in the Brain.
> Following the traditional formula of the George Mason School, The Elephant in the Brain takes a well-established mainstream concept—in this case Robert Trivers-style self-deception—and takes it to its radical but patently correct conclusions. In many ways the most ambitious of the Masonian canon, The Elephant in the Brain has such radical implications for so many areas of human life that, in less scattered times, one could imagine it spawning an era-defining psychology metanarrative like The Interpretation of Dreams or at least Thinking, Fast and Slow. The last few chapters probe self-deception’s implications for sociology, and I’d aspirationally frame this essay as a coda to those, which extends the framework to the phenomenon of wokeness.
----
> whereas truth on something you can’t directly act on has almost zero instrumental effect on pursuit of survival and thriving goals, which are terminal and therefore higher priority than satisfying curiosity.
I'm not sure I agree with you, here. But I'm having trouble putting my reasoning into words. <mumble mumble> Karl Friston and Symmetry Theory of Valence, <mumble mumble> Thompson Sampling. Also, do you remember that one Lesswrong essay about how wasps just do the same shit over and over, because the behavior is hard-coded?
Yes, easily. It comes from making simple choices where the perceived intent and the actual intent are not contradictory. The whole idea behind thought experiments is to probe at those weak points where the ideas conflict, and see if what you think is actually what you think you think.
(personal example; I used to think I wasn't picky about clothes, until someone took me to a shop containing nothing but shirts with large phrases written on the front in bold, and I realized that actually I have a strong preference for expressing as little as possible with my clothes.)
Yes. A lot of my personal growth has come from realizing that the reasons I think I do certain things (like fight with my wife, or avoid making new friends, or play video games) are not the actual reasons. As an example, this morning I started to get snippy with my wife when she said "Trust me, that was the wrong was to do it." If you had asked me five years ago why I was mad at my wife I would have said that she was demeaning me, or putting me down. This morning I defused it when I realized that I was getting upset because I was feeling incompetent and incapable and the words "trust me" seemed to imply that she thought I didn't trust her and was screwing up the task out of defiance. It's more complicated than that, but the incompetence thing is a real emotional trigger for me, and I figured out several years ago that my fear of my own incompetence was the real reason I did a lot of things that before I would have chocked up to my sense of justice, or needing to defend myself, or wanting to set the record straight, etc. I was able to turn my passive aggressive comments into direct communication to her asking her if she really thought I didn't trust her or didn't know that what she was saying was true, and I found out that she wasn't even aware that she had used the words "trust me" and was just trying to help.
A lot of philosophy and religion is tied up in the idea that we don't understand ourselves and why we really do the things we do or feel the emotions we feel, etc.
Yes, absolutely. Ask anyone who has been a parent or other caretaker of young children. Kids are demonstrably and trivially wrong with extremely high confidence about their own motivations and beliefs all the time.
Yeah, until I had kids I was unaware how much of introspection is actually a learned skill.
Small kids literally do not realize that they are hungry or tired... they just get increasingly angry, and they have no idea why. Like, a hungry child will get angry, and will then throw food on the floor because of the anger, which goes completely against logic. It takes a few years until children learn to recognize "I am hungry now" and "I am tired now" as emotions.
Actually, I suspect that even adult people eating food at midnight is often not caused by actual hunger, but by misinterpreting signals of tiredness for hunger. Both of them are "low on energy", but for different reasons.
So if kids need to learn all of this, it makes sense to suspect that maybe there are things that even many adults do not know. Signals that mean something specific, but we can't decipher them, and it becomes just a generic anger.
I don't think a person can be "right" about their motivations or beliefs.
Step 1 in thinking about your motivations and beliefs is: condense/compress/collapse the totality of your motivations and beliefs into something that fits into a Conscious Thought™.
Then if you're trying to communicate your motivations and beliefs to someone else, step 2 is: condense/compress/collapse the totality of your conscious thoughts about your motivations and beliefs into something that fits into some sentences or paragraphs of whichever language you're using (plus whatever grunts, hand gestures, poetry verses, oil paintings, vlog posts, equations, etc that you have time produce and that you think might help.)
The amount of simplification required for each of these steps is so extreme that the product of each step is more "wrong" than than it is "right."
But on the other hand, if I think "maybe my keys are on the desk" and then I walk to the desk to get the keys, I think i'm right that my intention in going to the desk was to get the keys.
I suppose, to your point, if walking to the desk brings me by the pantry, which is open, and hey, macademia nuts - i can't know for sure whether or not the keys were instrumental to the nuts.
Is there some way to have this cake and eat it, too? Can we evolve it to, "knowledge of your own motivations or beliefs is likely only possible in small scale, limited contexts and even there is questionable?"
On a small scale [why did he walk to the desk?] you're probably a lot more "right" than "wrong" when you think and speak about the belief [he believed the keys were on the desk] and motivation [he wanted to start the car].
But we rarely think/talk about beliefs and motivations at that small scale. Usually we're thinking about something interesting/uncertain/complex/controversial. And yet when thinking/talking about these larger scale, broader context issues we, unavoidably, perform the step 1 and step 2 dramatic oversimplifying processes mentioned in my previous comment *without noticing* that we're doing it.
Or to put it another way, I think it's a mistake (that we all make, almost all the time, without being at all aware of it) to think the "knowability" of a thing like [why did he walk to the desk?] is in the same category as the "knowability" of a thing like [is he motivated by pursuit of the truth?].
Epistemic status: Neither immaculate nor defiled ;)
"Can a person be wrong about what their motivation or beliefs are? I don’t mean intentionally deceptive - I mean, actually being wrong in their own self-assessments. Why or why not? What evidence can a person give that their motivation or beliefs are what they think they are? "
If a person states (and seems to be sincere about) that they value "X" and they keep making choices that are anti-X at some point I feel one must conclude that they are either:
a) Very wrong about their valuing X, or
b) Terrible at understanding how the world works (in at least one area of concern).
I'm eliminating "they are hypocrites" here because of the "seems sincere" bit, but if we don't have that then them talking the talk but then doing what they really want also comes into play. But that isn't your question.
I think that (b) is VERY common in other areas (e.g. most people seem to have cause-and-effect for athlete salaries and event ticket prices backwards...) so I don't see why it would be less common for motivations and beliefs.
What is probably also very common is people not wanting to make tradeoffs.
A common belief I see on various message boards is something like, "Great, now that player X has a new $YYY contract ticket and food prices will be going up again."
Almost certainly this is not how things work.
Consider the ticket (and food and parking) prices and player salaries at some time (like today). Say tickets average $100/ticket. And the stadium sells our 85% or whatever.
Imagine, next, that the entire league takes a 25% pay cut. Maybe there was a short strike and the players lost or the owners staged a short lockout. Whatever. There is a drop in pay, but during an off-season so the fans aren't upset because a season got disrupted. But now we have a new set of contracts and the same players (more or less) show up to play.
The fans have already established that they will pay $100/ticket to watch the players play and the same players are playing. Do we expect the owners to drop the ticket prices to $75/ticket or keep the tickets at $100/ticket and pocket the extra money?
The folks who believe that player salaries drive ticket prices expect that the owners will drop ticket prices. Economists, who expect the owners to be trying to maximize revenue, expect that the ticket prices will remain the same because the customers have established that they will pay $100/ticket so why drop prices?
The same thing plays out in the other direction. If players get more do we expect ticket prices to go up? If so, why weren't the ticket prices at that place already?
The *other* view is that the owners charge what they can for tickets and parking and food and the players get some cut of that (about 50% for the NFL). So if the fans stop showing up then ticket prices drop and eventually player salaries drop, too. If fans keep showing up even if/when ticket prices rise then the players' salaries go up because the owners have more money to spend on player salaries and many of the owners want to win (also, many leagues have salary caps and floors that are based on league revenue).
Economists generally believe that rising ticket prices drive rising player salaries rather than the other way around. Lots of fans think the causality runs the other way.
I would have thought "high athlete salaries = higher event prices to cover the costs" but is it in truth "high event prices = can then pay higher salaries to the stars"?
I sort of agree with your conclusion, but not with your argument for it.
We absolutely do not live in a world where people only act on things that have instrumental effect on pursuit of survival and thriving, and there are plenty of examples of people choosing to satisfy curiosity over survival (functionally, anyway, usually they are not aware of just how badly their curiosity will backfire).
> usually they are not aware of just how badly their curiosity will backfire
Yes but i think this is the key: in these situations, they didn't understand their risks. I think people understand 'risk of upsetting my group' FAR more intuitively than they understand other risks. I don't think people risk tribal membership ejection nearly as readily as they risk e.g. bodily harm, because you can heal a broken limb far more easily and reliably than a broken reputation.
This question is tough for me to get a handle on. In part, I think that's because the line between belief and unbelief isn't well-defined. In part, that's because belief has an important role in group membership. Robin Hanson's idea that it's easier to talk about group values by gesturing at something sacred rather than just being explicit about those values is relevant. In that way, belief is a kind of social ritual, where we don't just privately believe and suspect that other people privately believe, but we see each other making professions of belief.
I think too about how important COVID denial and COVID vaccine skepticism are as markers of membership in certain groups. People literally died for those beliefs. It's hard to think of a stronger indicator of belief than this. At the same time, I'd say that they engaged in some serious motivated reasoning to arrive at those views. That points to some shakiness of belief. They made a much bigger bet on those beliefs than any bet I'd make that they didn't really believe those things, though.
I don't think this is true. They may have died _as a result of those beliefs_ - but dying _for_ a belief means taking a risk you believe to be fatal, due to the belief. If you think COVID was all exaggerated, you aren't taking what _you_ believe to be a fatal risk, regardless of what it is.
From their perspectives, people who took the covid vaccines themselves were taking on risk for their beliefs: they took a mass-manufactured chemical product made by corporations that lie, in order to avoid something with a low fatality rate. They'd say _you_ engaged in motivated reasoning.
It seems to me, then, that whenever we look at reasoning of people with different value structures, we think _all_ of their reasoning is motivated, whereas our own is pure and based on the truth. To me it seems most likely, then, that it's always subservient to group membership.
Sometimes the part of your brain that rationalises decisions is not aware of the part driving the decisions (hunger, and so forth), so I would say not only is it possible but relatively common.
I think people can and often are wrong about these things. Another thing they're often very wrong about is the origin of their beliefs and motivation.
That said, I wouldn't automatically discount as false a claim about being interested in finding the truth. I know I am personally interested in discovering the truth even when there are negative social consequences for doing so.
Of course, I could be wrong about my own motivation, but if you want to place this in the context of the surviving and thriving framework, curiosity often leads to unexpected discoveries that have a huge survival advantage but it is not possible to predict initially which discoveries that would be and how to even begin to go about making them. All you can do is try lots and lots of things and try to make connections between those things and notice patterns etc. So it seems to me that an innate drive towards general curiosity without any underlying social motivation is something evolution very well could have produced.
I agree that this is true - curiosity is totally a thing. But I think people will generally avoid being curious about things if that curiosity itself poses an existential risk to them. Given how socially oriented we are, being curious about, e.g. the relationship between genetics and iq can pose an existential risk.
Any folks close to the topic have thoughts on the FDA commissioner’s plans to cut drug approval time by months? Are they doing it “right?” Estimated lives / QALYs saved?
It's more talk than concrete action at this point. The FDA unveiled their new AI chatbot "Elsa" recently to help cut down review times, and despite the hype marketing from the commissioner, the insider sentiment was far less bullish. That said, there is no reason why AI should not significantly reduce review times if implemented properly, but the FDA is more in the vision / intention setting stage at this point. And most of their accelerated review pathways they have created give them all sorts of outs if they, in fact, cannot get it done.
What do y'all think about this foreign policy for the Middle East from the perspective of the USA?
1. We must achieve complete independence from reliance on Middle Eastern oil by any means necessary--offshore drilling, fracking, investing in nuclear power, solar, wind, whatever.
2. Then we get out. Completely. Iran closes the Strait of Hormuz? Fine, we don't need the oil from the Persian Gulf. Houthis shooting ships in the Red Sea? That's not our main shipping lane anyway. Someone attacks Israel? I'm fine with selling weapons to the Israelis, but they've done a good job of defending themselves over the past 60 years. They can survive without our help. Palestinian statehood? Let someone else figure it out.
As others have noted, the US has already pretty much stopped using Middle Eastern oil, and is now competing with the Middle Eastern customers in the oil-export business. 1973 was a *long* time ago.
But the US economy is absolutely dependent on a great deal of continued trade with other nations which are dependent on Middle Eastern oil. And that's not going to change any time soon. Trying to turn the United States into a North Korea style autarky what doesn't need nothin' from nobody, without in the process impoverishing Americans to the point where a populist landslide will bring in a President with a mandate to reintroduce globalism, would be the work of decades. And we're almost certainly not going to stay focused on the project that long.
For the foreseeable future, Americans will suffer significant economic hardship (and the politicians responsible will risk electoral ruin), if Middle Eastern oil stops flowing.
1. Oil is a commodity. "Independence" is impossible unless you're willing to set up a system where the price of oil is different from the rest of the world. Which basically means complete autarky and is impractical. If you simply want the US to not be directly reliant on it then the US already has that and always has. I think the Middle East is like 5% of our oil.
2. The US is highly integrated into East Asia and Europe which DO rely on Middle Eastern oil. They also rely on those trade routes. Unless you can get the EU, China, Japan, etc to sign on to energy independence and avoid their biggest trade routes you'd still have to care. Now, you might think "if they're primarily the ones benefitting why don't they help" and in that case welcome to the American diplomatic position since like the 1980s.
3. The Middle East is important strategically if you want to defend Europe or North Africa and it gives us a strong ability to influence South and East Asia in one shared defense perimeter that's continuous with Europe and North Africa.
The folks pushing energy independence often want to see Europe and Japan [maybe not China!] pay "their fair share" for the naval power to keep sea lanes (including the Persian Gulf) open. This tends to get lost, but should be kept in mind while thinking about this.
"Why is the US paying our Navy keeping the Persian Gulf oil flowing to China?" probably captures the spirit of this (whether the facts are correct or not).
The main benefit of middle east right now is a source of investment, and a lax regulatory regime. We can build all the AI plants we want in Saudi Arabia.
The cost is jihadi fundamentalism willing to blow stuff up to gain god's favor.
If we withdraw China will swoop in and replace us.
We stopped trying for boots on the ground regime change, and effectively stopped mediating in Israel/Palestine except insofar as it improves regional stability (i.e. Abraham accords great justice for Palestinians who cares)
Couldn't be more satisfied with the current approach, actually
I believe the argument is that the US gets to stay the world hegemon (also) by keeping others down. They can't afford having a competitor that has access to both advanced technology and plentiful resources.
Yes, that is obviously the argument. But, again, why is it important to be the world hegemon? If I read OP correctly, they don't think that is intrinsically valuable. And after all, it is safe to say that, of all the people in the world who are living happy lives, most do not reside in the world hegemony. So, again, why does it matter?
I guess the elites who profit from controlling a country want it to be strong so it can extract more wealth for them? Also not being the hegemon does not mean you can stay comfortably on the sidelines as number two or three or four. Germany seemed to be doing fine as not-the-hegemon until they got their energy supply cut off by the nord stream 2 pipeline sabotage. Now they have no recourse.
Leaving aside that "being the hegemon" does not mean that you control other countries, I am not asking why elites might care; i am asking why any of us, such as the OP, should care.
I don't understand. The Gulf states offer the US stuff it wants: a lot of space and lax regulatory regime and energy that they are willing to trade in exchange for military protection. If we don't do it China will offer it instead and we would be relatively weaker.
Having power and a higher share of global economy/effective resource control without giving up nice things like being able to stop government from bulldozing my house with eminent domain to build an AI chip fan seems good for me/US foreign policy
I don't understand why you think the US wants those things. Those sound like things that US companies doing business overseas might want, but not the US govt per se. Nor do I understand how "China swooping in" would change anything in that regard.
>energy that they are willing to trade in exchange for military protection.
So, your claim is that, if the US conducts foreign policy in the Middle East that is to Saudi Arabia's liking, it will let US companies build AI chip and energy facilities, but if not, then China will "swoop in" and replace the US as Saudi Arabia's security guarantor, and Saudi Arabia will instead let Chinese companies build those facilities? Seems like you are positing a force projection capacity that China does not currently seem to have.
And, why not build those factories in Mexico or elsewhere in Latin America?
How hard is it to build AI plants in rural Arizona or Texas? I have my share of problems with using eminent domain, but there's a lot of empty land in the US.
That is a not a very apt comparison. High speed rail is built with government funds and hence is subject to political considerations (eg, in California, the requirement that it actually serve the people who need it most, rather than bypassing them). And both Arizona and Texas have grown rapidly; there seem to be few constraints on economic development.
Due to advancements in energy production (something something fracking, ask somebody actually in the energy sector) the US is a net energy exporter. We probably still want access to the Middle East but, honestly, if the Middle East shut down it would probably make our own oil and energy reserves far more valuable.
The issue is that, well, the rest of the world still needs cheap oil, which means the Middle East. A world in which cheap oil keeps flowing to Europe and developing countries is a better world because developing countries deserve a decent standard of living and because we like Europe, it's full of the history and funny accents :).
But yeah, fundamentally, the US no longer needs the Middle East for its energy needs and is, at best, providing a free security good to the rest of the world and, at worst, meddling in a volatile region out of habit and the influence of varying lobbying groups (Israel, yes, but also the Saudis and Emirates and their ilk).
Wow, I think that, of the finalists and honorable mentions I only read the Watergate one (it was really good!). Good to know, there is a lot of good content to be read on the coming weeks then.
Shout-out to the review on human sexuality. Thought it was well written and had an original deep insight, even if it seems a bit flawed/incomplete. Would really have liked it to see a discussion here of it
Hey, author here; thanks for the shout out. Definitely ran out of time on this/was a little rusty at essay writing, I don't think I quite did a good enough job conveying the original insight, which is a shame. But it was still really fun to have written it, to have put something out there that was read by probably hundreds of people.
The short summary/TL;DR is that human sexuality is fundamentally tied into our status detecting systems, and this explains most of the weirdness around sex, such as people being into being cucked or the prevalence of violent fantasies.
What parts of it seemed incomplete/ in need of elaboration to you?
Im going to comment both here and on the post. Who knows, maybe the algorithm takes this and signal boosts your post
I really enjoyed the status approach to sexuality but I still think what you call low-status hotness (LSH) is kinda mysterious
While I buy the idea that submitting to a high status partner and surrendering responsibility may be hot or attractive to some people, I think there is more at play in some of these fetishes. It seems like the very act of lowering one's status (either physically by pain or socially by humiliation) is the goal of the drive here, not the second-hand consequence of submitting.
This part, where pain or humiliation turns into pleasure feels like a gap not only in your system but in most explanations of human sexuality imo. I think without explaining this gap some extreme fetishes like straight up CBT or SPH remain unaccounted.
I dont know, maybe by taking pain or humiliation you increase the "status-delta" between you and your partner and it makes them hotter? This could be a viable explanation but still seems to ignore the fact that the pain is the goal for some people
Yeah, I would probably phrase the idea of low-status hotness as being about wanting that "status-delta" to be as large as possible, of wanting to be around an incredibly high status partner in a sexual context. This can be done by either having the high status partner act even higher status, by inflicting pain or punishments or giving affirmation or restraining, etc, or by having the low status partner become even lower, via receiving physical pain, social humiliations, being restrained, etc.
These often blend together; when a Dom is tying up a sub, they're increasing in status in the scene, while the sub is being reduced. I perhaps didn't clarify in the essay properly, but yeah, I think goal of the drive is to have that status differential, not the second hand consequences.
As for pain being the goal for some people... the people who I know, who have masochistic tendencies, are usually only into pain in certain specific contexts, which almost always involve a specific person hurting them. They're into being spanked, perhaps brutally hard, but they're not into stubbing their toe on a door jam repeatedly. That makes me think that pain is downstream of the status differential, not the end goal in itself.
Surprised that the Testosterone review didn't make it and kinda disappointed (1). I thought it was well argued and potentially worth making the lifestyle change and I wanted to see the ACX commentariat tear into it and see how it fared.
It was one of the good ones. I did have an issue with it: I didn't know whether I should trust the author to be unbiased enough. The author was certainly very convinced by the conclusion. They did discuss some drawbacks, but I didn't know enough to say whether it was a fair treatment. They were very dismissive of the downsides, and it sounded very surprising to me that this should be all that there is.
I did give it a good grade. I would have liked the review to reach the final, so that I learn more about it from the comments.
> They did discuss some drawbacks, but I didn't know enough to say whether it was a fair treatment. They were very dismissive of the downsides, and it sounded very surprising to me that this should be all that there is.
Author here. I probably didn't do a good enough job, I wasn't trying to dismiss the downsides, I was more trying to show that they exist, but mainly affect people who do a ton of it over time.
I basically tried to show that even at supraphysiological doses, there are essentially no side effects in all the studies we have. Now crank that down to a TRT dose, and the risk is significantly lower again.
Best of all, the downsides are all measurable and mitigable. So it should be quite safe in expectation to start, and as long as you're measuring the potential downsides (blood pressure, cholesterol, etc) on a decent cadence, it's essentially a no-to-very-low risk to try intervention for a potentially large quality of life bump.
No, my formulation was not good, you did express that. It's more that I would have like to hear a second opinion on the matter. Especially because it's in the general field of medical/nutritional intervention where there is a lot of biased advise. And it's a bit of an extraordinary claim: the positive effects are strong and very notable, and the negative effects are pretty negligible. That doesn't mean that it's wrong, and it makes it all the more important if correct.
It was a really interesting review. I won't act on it right now because I suspect that my testosterone level is still alright. But I will certainly not forget it, and may act on it in the future.
I skimmed it, got really interested and promised myself that I would go back, read in detail and rate it. Oh well. At least you reminded me to read it fully, I really started considering the lifestyle change as well
Ah, ACX commetariat, I am disappointed that "11 Poetic Forms, Reviewed in Verse" wasn't a finalist. It was so fun! I didn't expect it to win but I really hoped it would at least place.
The concept was good, and probably if Scott had written it it would have been great, but the author doesn't have quite that level of verbal cleverness and so it didn't make me smile as much as I'd have liked. This kept me from rating it more highly.
My friend is a journalist and he has been writing about the US Aid cuts in Southern Africa. He has mainly been focusing on South Africa, but recently went to Mozambique to see the effects there.
It seems to be nothing short of a total catastrophe. As in, tons of children just dying of HIV levels of catastrophe. The country's health systems seems to have been entirely propped up by USAID funding.
There are certainly debates you could have as to whether or not it is good for a country to be so reliant on USAID, but unfortunately it was and with the cuts, we are now seeing dire consequences. I've linked the article below if you want to read more. Was hoping some EA type people could see this.
So the NGOs could not bear even a slight disruption in funding ? Perhaps they prioritize CEO salary over the dying children?
It was quite amazing that the sob stories began to appear right after the aid cut. It didn't even take a week for people to start dying the world over.
China puts hundreds of billions of dollars into neocolonial resource-extraction "development projects" in Africa, but somehow doesn't notice all the HIV. I guess the AIDS crowd don't work as hard in the mines. The European Union, richer than the United States and as they incessantly, sanctimoniously remind us *far* more morally righteous, pay a pittance to AIDS relief in Africa but really that's beneath their notice as well.
The United States, alone, sinks a hundred billion dollars into beating AIDS into submission across the African continent, and when it finally says "enough of this, time for the rest of you lot to take up the burden for a few years", *that's* the morally outrageous part.
Either they're a dependent satrapy or they're not. They can't have it both ways. Well they can, but then hundreds of thousands of their people will die. And they will blame the US for it. And most people will believe them.
Without forcing people to virtue signal, I am amazed at the types of comments a post like this gets.
It reads like intellectualizing. You even explicitly say "There are certainly debates you could have as to whether or not it is good for a country to be so reliant on USAID, but unfortunately it was and with the cuts, we are now seeing dire consequences."....and the commenters jump exactly on that part.
Kids with AIDs are dying. Skilled forecasters estimate that the PEPFAR stop due to claimed "fraud, abuse and waste in USAID" will directly cost the lives of ~half a million people per year.
Deaths after one year: [257,500 - 515,000 - 772,500]
Deaths after four years: [1,015,000 - 1,998,750 - 2,953,750]
I will repeat- half a million lives per year. Even if you disagree with USAID, or think there is actual abuse, or there is nothing good about foreign aid, or it can make countries dependent, or it can foster some corruption in countries, I posit that half a million dying is an moral outrage. I would urge any commenter to recalibrate their moral compass if this is one of the first responses you have to reading the associated article. Hope indeed some EA funders will see this.
a. Are there any circumstances in which, having started this program, the US would be morally allowed to end it?
b. Are there any other countries that are obliged to pick the program up now that the US has ceased funding it?
I understand the walking-by-the-drowning-child argument for why we are obliged to keep doing this program. But that seems like it applies as much to every other country rich enough to fund it as it does to us.
And if we're not doing the walking-by-the-drowning-child argument, then we seem to be, as John pointed out above, firmly into Copenhagen ethics--we are obliged to keep helping because we started helping, the UK, EU, China, Saudi Arabia, Qatar, Japan, etc. aren't obliged because they never started.
> I posit that half a million dying is an moral outrage
I think the actual moral outrage here is that these people are being born in the first place when there is simply no need or future for them. These people are a cultural and genetic dead-end. They do not need to keep existing. How long are you going to keep throwing money into this black hole?
These people have more genetic diversity than the rest of the world combined. that's hardly a "dead-end" like cheetahs, which bottlenecked quite severely....
> These people have more genetic diversity than the rest of the world combined
And do you not see how that might be a problem in an environment where large, practically eusocial societies have a massive advantage over smaller ones? Diversity is not strength, it's a cancer that eats away at itself.
Of course, none of that matters if the quality of the people themselves are lacking.
Sigh. As noted earlier, "large societies" existed in Africa. It wasn't all tribes, like North America pretty much was.
North America, prior to Western Conquest, consisted of tribes slowly devolving out of being human (losing words, in particular).
Diversity in terms of genes insulates you against disease, and can provide "rare but useful" mutations, which are useful when a "big change" comes and makes most of your population "less apt to survive."
Folks that come from mountains have a very different psychology than most. This diversity is something you'd probably celebrate, as it's adaptive (and helpful for when we colonize Antarctica).
Half a million dying from an intentional, created illness is a moral outrage. Half a million dying from influenza is just an ordinary decade (and, judging by the vaccine's effectiveness, not preventable at any rate).
Africa Wins Again is a perfect slogan for the deadliness of Africa-in-general. Lions, snakes, sleeping sickness, monkeypox, ebola...
I will not be outraged because of something that I did not do. Passively letting someone die, who I could save, is not a moral outrage. It is a moral lacuna. It is not deliberately performing genocide.
Let me be perfectly clear: The perfectly preventable deaths of Ukrainians from 2022 to 2025 are on OUR HEADS. We voted for them, we got them -- blood and gore. Even the soldiers infected with HIV, deliberately, are on our heads.
I will NOT be shamed for letting someone die from their own actions. Shame on America, shame on US, sure, but shame us for what we've done. It's far more shameful than our inaction.
I'm confused. Why are the deaths of Ukrainians on our (US?) heads? We... didn't kill them? You mention that those deaths were "preventable", but isn't that exactly what OP is saying about the deaths in Africa.
Also, when we drive the Ukrainians towards our objectives (Sevastopol), at the cost of 90,000+ dead (surprisingly, those are the Russian numbers, and they're somehow lowball, I do not understand this -- surely someone else is giving lower numbers than the lying Russians??!) -- the counter-offensive was essentially a suicide attack.
Giving them HIV-contaminated blood was deliberate. So is the training given, and more importantly the lack of reassessment and retraining as the number of amputees rose. So too is demanding that the Ukrainians do a suicide attack... (and completely blowing up the peace talks, because of overconfidence that we really could get Sevastopol).
> Skilled forecasters estimate that the PEPFAR stop due to claimed "fraud, abuse and waste in USAID" will directly cost the lives of ~half a million people per year.
The article you link to doesn't say that. It estimates what will happen if the waiver PEPFAR has been granted that allows it to continue operations doesn't work and PEPFAR is effectively cancelled. But PEPFAR is not currently stopped, though there were disruptions to its operations that are still being worked out.
Correct, I should have made that explicit. From the sources I have read, and the above-linked article, the fact that a waiver exists actually does not mean PEPFAR is continuing. Supply chain projects are not being picked up.
At different points in this century, the UN has projected a 2100 population for Africa at 2 billion and 4 billion. I don’t know what they’re thinking now or if there’s any reason to pay attention to them.
But offering such numbers in this context, is not necessarily bound to shut down discussion of the future. Especially when tagged with “how could we be missing 5 million people?” etc.
A country that cannot fund a bookmobile to drop off in rural areas the ARVs it is sitting on, is unlikely to know how many people live in it, nor what they die of when they die.
Reading that, one has to wonder what the hell the government of Mozambique is doing. If I look up Wikipedia, it is a poor, underdeveloped and indebted country. On the other hand, apparently with development policies and loosening up regulation, the economy grew very fast. One more case of "the GDP is soaring but people think they're not doing well"?
So if the economy *is* growing, where's the money going? Corruption, it would seem, as well as the lingering effects of the civil war and new uprisings in the north. (Thank you Google AI for the précis):
"Mozambique's economy in 2025 is experiencing a mixed outlook. While facing contraction in the first quarter, there are projections for a gradual recovery, with expectations of 2.7% to 3.0% real GDP growth by the end of the year. The recovery is expected to be driven by the extractive sector, particularly gas production, and agriculture. However, challenges remain, including inflation, fiscal deficits, and the ongoing impact of the Islamist insurgency in the north."
Apparently it is also a popular tourist destination. So there is *some* money there, but who holds the purse strings or where it's going is the question.
"Mozambique’s economic structure is determined largely by the extractives sector, with its large gas reserves of 180 million cubic feet, the third largest in Africa. There has been only limited structural transformation of the economy, with a slight shift from agriculture to services. In 2001, the services sector was the main contributor to GDP growth, accounting for 50%, followed by agriculture at 32.2%, industry (including extractives) at 18%, and manufacturing at 12.2%. Two decades later, economic activity continues to be driven by the services sector, which accounted for 51.7% of GDP in 2021, while agriculture’s share had fallen to 28%. Mozambique’s structural transformation is hampered by skills shortages and a high illiteracy rate among 15- to 34-year-olds (31.4%), skills mismatches, lack of infrastructure, high public debt, and a low level of industrialization."
According to the linked story, once the USAID money stopped flowing, that was it. Nothing happened. Nobody else stepped in. No home-grown solutions or attempts by the government to intervene, no other agencies or charities. If the government is this bad, then the country might as well become a satrapy of the USA
More specifically what happened is the government basically borrowed a bunch of Chinese money because they didn't like that the US wanted things like transparency or anti-corruption. They even cut services and expected the US to pick up the slack. They were betting the US would keep funding them because of stuff like... well, this story. Basically that no matter how much they favored China or cut their own public services the US would continue to pay because otherwise ordinary people would suffer.
But then China pulled back on funding as it became increasingly clear Mozambique was not going to be able to repay them or develop. And then the US cut aid. At first a little to signal displeasure then when the US did zero based budgeting it couldn't justify continuing in Mozambique generally. And now Mozambique is left in a deep crisis. It's unfortunate for the average person but it's the result of extremely bad local leadership.
I imagine China would have been interested in things like the natural gas reserves? When you've managed to frick up your government to the point that it's not even worth it for a colonising power to come along to scoop up your natural assets, that's pretty bad.
I realise they're still struggling with an underdeveloped country and a past civil war plus new insurgency in the north of the country, but it does sound unhappily like much of post-independence Africa where the new rulers decided they'd divvy up the spoils amongst themselves, and to hell with the citizens, they're not even all our tribesmen.
Mozambique does have natural gas but not a ton of it. It's ranked about 90th if I recall correctly, something like .03% of global production. That places it alongside such energy superpowers as Latvia. Most of Mozambique's exports are agricultural products in a broad sense. Stuff like nuts but also wood. And all exports are only about $12 billion and exports to China are only about $750 million. Mostly niche agricultural products and high quality lumber.
China's plan for Mozambique was, fairly transparently, "We come in and run the Chinese economic playbook. You undergo rapid industrialization to middle income and reorient your trade to focus on China." Not a bad idea overall except the part where none of that actually happened for a variety of reasons. But mostly, in my opinion, corruption and more general dysfunction.
I will say Mozambique is uniquely bad even among dysfunctional African states. It's undoubtedly in the top three and has an okay shot at being the worst. Though that honor probably goes to Republic of the Congo (not the DRC, the other one). But it's close.
There does come the point where you have to ask if USAID and the likes are doing more harm than good; if the attitudes of the rulers are "why should we even attempt to set up a functioning country, instead of just robbing the treasury blind? the bleeding-heart Westerners will take care of our people for us, as well as shovelling aid money in that we can keep creaming off", then is it in fact doing more harm than good?
Once again, I find myself in the very discombobulating position of agreeing with a Shoe0nHead video:
Mozambique is a particularly egregious case and that period was relatively brief (about 5 years) before they got cut off. Also they didn't directly steal western money. They also wanted to take money from the Chinese. And in the end both pulled out. But yeah, that was more or less their plan.
This is why USAID is so big on democratization. They believe, rightly or wrongly, that this kind of corruption stems from authoritarian governments (which Mozambique has). And to be fair to that point of view, more democratic countries mostly are much better about this kind of stuff. Kenya, for example. Though also authoritarian Rwanda.
On the one hand, there is significant human suffering. These individuals are suffering and that is obviously bad.
On the other hand, this is a predictable consequence of Mozambique's actions which have repeatedly taken the US for granted and prioritized China. The reason the US funds Mozambique's healthcare system is they cut domestic funding in order to repay China and expected the US to pick up the slack. It was supremely foolish of Mozambique's leadership to expect US taxpayers to (indirectly) finance Chinese debt repayments.
China owns about 7.5% of Mozambique’s debt (or about 14% of the Mozambique’s external debt). Government spending on healthcare increasing from 2016 though 2022, going from 29% to 36% of total healthcare spending. The World Bank doesn’t have numbers more recent than 2022, but I haven’t seen any reports of subsequent cuts.
Eliminating healthcare aid to Mozambique in the hopes that will respond by defaulting on its debt strikes me as crazy on multiple levels. For one thing, Mozambique is currently running a deficit, so I doubt that defaulting on their debt would allow the government to increase healthcare spending at all, much less increase it enough to replace the money from USAID. For another, a default will harm all creditors, not just China.
Mozambique has been hiding how much debt it owes. Thus the hidden debt crisis. Anyone who says they know what percentage of Mozambique's debt is owed to China is lying. Further, total healthcare spending includes foreign aid, so that number does not indicate THE GOVERNMENT increased healthcare spending. So that statistic is misleading.
No one wants Mozambique to default on its debts. Including the US. Unless by default you mean the negotiated debt restructuring process Mozambique has asked for.
However, as FLWAB says, if you side with US rivals and cut spending to pay back US rivals (effectively transferring in net American money to China) and lie about how much debt you have and turn down transparency/anti-corruption requests then the result is unsurprising. And, to be honest, your fault.
I don't think the hope is that they'll default on their debt, and I think the idea is that if you prioritize US rivals over the US, it's not surprising if the US stops paying for your medical bills.
What are the long term consequences of developed nations propping up an entire continent which seems wholly incapable of caring for itself? Is it even desirable in utilitarian terms over the long term? What could go wrong?
What I am bringing up is that the key to an effective future eventually depends upon the concerned individual or group taking responsibility for themselves. My concern is that we are treating the people of Africa like children or livestock rather than as responsible human adults that need to take control of their own destiny.
IOW I worry we are promoting dysfunctionality and dependency. Of course, I am often wrong.
That is way too general an argument. Why is the specific aid in question an example of the phenomenon you are referring to? We are talking about medicine for sick people; we aren't talking about propping up core government functions.
I was indeed generalizing the argument. Trying to take it up a level of abstraction. Sorry.
Delivering drugs to sick people seems like commendable work to me. My question is whether the larger pattern of sending in solutions to their problems is the long term best solution. It seems like we are addressing the symptoms and by doing so interfering with their ability to become self sufficient.
Delivering drugs to sick people is fine, if they're reasonably cheap and don't lead to antbiotic resistance. Ivermectin is fine, I'll walk on hot coals for that. I won't do the same for HIV drugs -- they're expensive, and this whole "we let orphans live in the middle of the boonies" so that they are impossible to supervise without costing millions of unnecessary dollars... that's just ridiculous.
One orphanage. We have the plans from covid19. Build it there.
>My question is whether the larger pattern of sending in solutions to their problems is the long term best solution.
Yes, that is very clear. But the issue is that you can raise that question about any aid. Lend-Lease, for example: "Delivering weapons to people resisting Nazi aggression seems like commendable work to me. My question is whether the larger pattern of sending in solutions to their problems is the long term best solution. It seems like we are addressing the symptoms and by doing so interfering with their ability to become self sufficient." So, at some point, you need to make a claim that THIS PARTICULAR aid program is counterproductive in that manner.
Depends. We can get a lot out of certain "charity work" (that's called R&D for when we finish depleting our own aquifers). Ivermectin is cheap and safe, it's not a drug that we spend oodles on to make in quantity (likewise antibiotics).
HIV drugs... that's expensive.
All the t-shirts they've got, they are castoffs from America, where people won't buy a t-shirt if they don't like the logo. That's cheap writeoffs.
I'm not sure if Africa is wholly incapable of caring for itself...
Why aren’t they more self sufficient?* If the rest of the world didn’t exist at all, what would be happening there? How would they respond to HIV and countless other diseases and ailments? How would their governments and major institutions work?
By providing aid and relief are we interfering with the natural process of them evolving respectable institutions and social fabric?
Africa is not self-sufficient because people never needed to look to the future there. Food is too easy to grow, and they don't have winters. Future-oriented thought is rare there, and not very encouraged by the climate, or by the populace.
If there weren't the rest of the world, they'd simply let the disease kill people and breed all over again.
I'm not sure you HAVE a natural process of them evolving respectable institutions and social fabric. What you have is a tremendous reservoir of biodiversity, precisely because selection pressure is next-to-nonexistent.
In Israel, the dry season is the "dew" season. It's not like they don't still harvest food during the dry season -- it's just different food. Winter is harsh, and cold, and cruel in a way that a dry season just...isn't. You have to make a bed, early, before the rushes get too stiff, and can't be woven right. You'll die if you don't have enough wood, and a solid enough home (and a bed).
In some "warm" places, they have to put signs up warning people not to sleep on roads. (Yes, these are people of notably low intelligence).
You'd probably consider me to be a meany if I ask you to provide references that back up your thesis, but if you consider yourself to be a rationalist, you should be prepared to defend your positions.
You are faking rationalism here. You are following circular logic where poor outcomes (low IQs/poverty/whatever) of blacks is result of racism and simultaneously same poor outcomes is a proof of racism existence.
You poke holes in opponents' theories yet never prove your own (maybe because it's circular so it doesn't need explanation)
If we experimentally tested IQs of mulltiple diff pops in controlled settings and found zero diff, it would be rational to set zero as a prior and demand proof for any claims about pops we did not study yet, like we have extensively tested thousands of potential medicines, and most don't work, but testing people is unethical, and there is no ground to claim for zero prior.
Any references to evolutionary psychology ought to be thrown out on general principle. If you must have a critique on them, in general:
1) "It was effective" -- once, while we were back in Africa (nutrient consumption have changed)
2) "It's not evolutionary in nature -- merely tied to something that is."
3) It works out enough of the time that it's still in the gene pool. It may be disappearing.
This is why we don't explain "manic depression" as something that is "evolutionarily good" because it's still in the gene pool. There are good evolutionary psychology explanations for it, but you can't disprove the three ideas above, so you're just spitballing and that's not science.
Taking the whole branch of evolutionary psychology with a grain of salt:
That's nice hyperbole disguised as rationality, but the entire continent is not "wholly incapable of caring for itself", some countries are just not capable of caring for all HIV patients. Hopefully, with new medical treatments and economic growth it will be less of a problem in the future.
HIV is just one of countless problems. Setting aside any hyperbole, the question is are we making them better through our aid or are we fostering dependency, free riding and general institutional dysfunction?
Aid is bad when it's tied up with conditions that give the aid-giver some kind of hostile soft power over you, dependency structures are part of this problem. But It's good when it prevents millions from dying in between now and when a vaccine is being developed. In general, Africans would prefer mutual economic development to aid, but HIV is one example where immediate medium-term aid is needed until advanced medical infrastructure that physically cannot exist in Africa yet has time to develop. It would be incredibly hard to argue that PEPFAR saving something like 20 million lives is a worse outcome than simply being free of dependency.
So I can agree that mutual economic growth is preferable to aid, maybe even in general, but some cases are just... good.
I'm generally in favor of PEPFAR based on priors, but I want to push back on your definition of when aid is bad. Specifically, the USA subsidizes overproduction of crops, and then gives that excess away in the developing world, undercutting their markets and reducing their ability to grow their own economies.
That's broadly the position of this report by the Cato Institute (I'm probably a pretty typical neoliberal, but I feel anyone with the Economist logo ought to be at least open to a Cato analysis)
Do you believe that Omicron was a netgood to the world? If so, would you support more revocation of HIV/AIDS funding? If not, how do you explain the creation of Omicron?
Was that question for me? No, any pandemic or epidemic is necessarily a bad thing. A pandemic like SARS-CoV-2 stressed the healthcare systems of well-developed economies, and the healthcare system of countries like India totally collapsed (China implemented a really strict lockdown, so they were able to prevent the collapse of their healthcare system—which is pretty good in their cities). Therefore, pandemics have negative downstream effects on demographics and the economy. Luckily, the world was starting to roll out vaccines as Omicron started ramping up (As of a couple of years ago, there were 23 different vaccines of various effectiveness that had been developed to combat SARS2). Just eyeballing it, it appears that Omicron and its descendants accounted for ~20% of the COVID-19 deaths. It would have been much higher without vaccines.
I think it's legitimately hard to figure out what kinds of aid will be helpful and what kinds won't be, for long-term development. My not-that-informed impression is that many attempts at making the world better with foreign aid have ended up basically wasting a lot of money, and some have been actively harmful.
Though providing antiretrovirals to HIV+ poor people in countries that have zero hope of affording those drugs themselves seems pretty positive--it's not like we're strangling a nascent pharmaceutical industry in the crib by mistake or something.
Yes. Many of the rationalists on this substack seem to lack any historical context and also seem to be woefully unable to think outside their tightly constrained intellectual boxes. Of course, most of the youngsters on this stack weren't alive during the initial phase of the US AIDS epidemic in the early 1980s. They might think of HIV/AIDS as a treatable disease. But NY City hospitals were overwhelmed with AIDS patients who required months of care and who inevitably died. Initially, the mode of transmission was unknown, and medical staff were accidentally infected with patients' body fluids. It was a real shitshow. ICUs were overwhelmed, and they had trouble finding space for non-AIDS patients who needed intensive critical care. So the surge in AIDS cases affected the non-infected population as well.
Now that we know hella lot more about HIV, we've discovered that immunocompromised people are mutational petri dishes for other pathogens. Luckily, most of those mutations can't overcome healthy immune systems, but occasionally they might. For instance, the Omicron variant of SARS-CoV-2 probably got its start in an immunocompromised person with HIV in central Africa.
Being laissez-faire is all very well and good, until it comes back to bite you on the ass.
Given the rise of Omicron, would you support defunding HIV-maintenance drugs in the 3rd world, as a way to encourage "less lethal mutations"? I think you could argue that Omicron, itself, is worth the cost in "deaths of HIV positive people."
Wow. That's cold. Of course, you assume you'll be one of the elect who survive next pandemic.
But an easier solution would be to remove the patents from HIV drugs—the most effective ones are still under patents because they were developed later, and can cost between US$50k and US$150K/year. South Africa has a perfectly functional pharmaceutical industry. And they manufacture the older HIV drugs that are no longer patent-protected, which they sell to other African countries and India. Of course, removing the patents from these modern drugs would hit the bottom line of investors and threaten the very foundations of our capitalist system of medicine. So f**k those Third Worlders if they can't take a joke. <#snarkasm>
Still the person who believes that it is okay to perform abortions up until 2 years of age, if it comes to that (which we all pray it doesn't). After that? Humanity can die, before it's okay to kill the kids. Call me cold all you like. : - )
Nobody has a perfectly functional pharma industry outside of Israel (India, maybe?) and China (national-security wise). You do make a good point, though.
AIDS, and more tightly speaking, gays committed genocide of the hemophiliac population. I'm not going to say this was intentional, mind, but I'm just bringing it up, in a subthread where folks are trying to shame Americans for inaction. This was not inaction on the part of gays, they were trying to make money by selling their blood.
It took about 4 years after the start of the AIDS epidemic, and about 2 years after the HIV virus was identified as the cause of AIDS, before the Red Cross and other blood banks started screening for it. Granted, it took about a year to develop an ELISA test for HIV, but IIRC there was a lot of resistance to the idea that HIV was the cause of AIDS, and then there was another round of resistance to screening (because of cost). I just checked with ChatGPT, and it says that ~6,000–10,000 hemophiliacs and ~4,000–6,000 transfusion recipients died from HIV contaminated blood products in the US. I'm surprised it was that low.
But remember, it wasn't just gays who caught HIV. I asked ChatGPT and says that ~23% of AIDS cases were women who caught it from heterosexual sex. Also, IV drug users were at higher risk. Right now in sub-Saharan Africa, more than half the cases are women. And the infections happen predominantly through heterosexual sex.
As a historically-challenged person stuck in his intellectual box, my question wasn’t really about HIV. My question was about whether we are on net creating dependency and fostering negative sociology-cultural responses to the myriad challenges facing the billions of people on this continent. Feel free to disagree. I have an open mind.
To put it in historical context, the concept of a "culture of dependence" dates back to at least the 1840s. Regarding the Irish Potato Famine in which millions died, Charles Trevelyan opined in 1846...
“The only way to prevent the [Irish] people from becoming habitually dependent on Government is to bring the relief operations to a close.”
Of course, the Irish were considered irremediably backward, with low intelligence and bad moral character by many in the United Kingdom, much like modern conservatives regard sub-Saharan Africans (without ever visiting there).
Trevelyn again: “The great evil with which we have to contend is not the physical evil of the famine, but the moral evil of the selfish, perverse and turbulent character of the people.”
So has Western Civilization been creating a culture of dependency for the past 180 years? Of course, there were bleeding heart liberals as far back as the 1700s. Jonathan Swift's "A Modest Proposal for Preventing the Children of Poor People From Being a Burden to Their Parents or Country, and for Making Them Beneficial to the Publick" comes immediately to mind.
So am I to take it that you are arguing for continued aid, and that, based upon historical precedents in Ireland, you are not worried about dependency?
In theory USAID is supposed to be helping countries get off aid. In practice I'm guessing they often fail at that part, and it sounds like they failed on Mozambique. That said the stakes are high enough that I'd prefer a "keep trying to reform until you get something that works" approach (even though it'll be slow, error-prone and relatively expensive) over a "sharp cuts and millions die" approach.
I was prepared to feel whatever feelings that article might call forth, and I know the only socially acceptable ones will be pity and recrimination against the correct quarter.
But I defy anyone to read it and not have *any* qualms about the sustainability of this … system. And I’m not sure the remote HIV-positive children and their grandmothers aren’t right in feeling that their own countrymen in the form of their “activists” abandoned them.
Is this the future people really want for the world?
I don't see what the activists are supposed to do. They have no income, now they have to spend their time in surviving instead of delivering medicine and caring for the children.
Perhaps a jaded view but the absolute abruptness with which they broke off contact, inevitably suggested to me that they were gatekeepers of the medicine, taking their cut of aid funds, more than dedicated medical personnel.
It would’ve interesting to know what they are doing now to survive.
The article said now they're doing subsistence farming and peddling wares in street markets.
I wouldn't phrase it as "taking their cut of aid funds", it sounds like distributing medicine was a full time job. Nobody would be doing that for free in a developed country either.
In a developed country, like, say, Mexico, they'd be in an orphanage. Or doing productive work, so that their HIV drugs could be paid for.
Actually, these two can be combined, in a developed country. It wouldn't take that much work for Mozambique to "hire apprentices" for skilled labor (that's an orphanage, with one nurse to deliver the HIV drugs).
Their concern is touching, then, and no doubt just such an excess of concern, for the least among them and for the future, is what got them into this pickle in the first place. We may wish to organize along similar lines, for our moral betterment.
This is true. It sounds like most of this could be fixed with "one orphanage" which centralizes the children and makes them close enough that one nurse could care for all of them.
I feel like despite the good intentions of individual actors, this is one of those modern situations that both kind of needs to work, and also must never *really work*.
Wuhan was working on a bunch of Actual Fixes for this particular situation.... Up to and including "murdering them all." I agree that charities are unlikely to fix this "issue" (as HIV treatment doesn't actually fix the disease, and seems to just make it more likely to spread... Take the 12-18 year olds in Mozambique. Sex trade is low-skill and high pay. Voila! More HIV. more need for charity.).
I've heard that modern treatment, if used according to guidelines, can make it all but impossible to spread HIV through sex. Is that not correct? Or was treatment in Africa not actually used according to guidelines?
Does the Longer term sustainable solution involve them building a culture and institutions which fosters monogamy and a revulsion toward prostitution and anal sex? Does aid interfere with this cultural development?
South Africa's still the place where squatters were using a multi-story elevator shaft as a toilet, right? I'm not exactly surprised that a country that throws squatters out of skyscrapers has "health system problems" -- there's a clear disregard for human life in South Africa.
A note that this was seven years ago. In 2021, folks were rioting until they ran out of food (in America, the riots ended when the nikes/legos/etc were gone) -- they literally destroyed the infrastructure necessary to eat.
I’m not sure what you are trying to say, but the linked article was very interesting and the picture especially so.
That picture is what America’s public lands are going to look like when Mike Lee and Matt Yglesias get through with them.
That’s the dream, when lunatic left and lunatic libertarian join hands and shout down what the internet so cutely terms “normies” but The Onion typically used to refer to as The Last American Adult.
Your claims about his positions seem entirely baseless to me. If you've got some citation for him advocating a sprawl of barely-intact shacks, I'm willing to change my mind.
Otherwise, it seems to me that you might as well be on another planet from him.
He’s the one with the dream of 1 billion Americans, by which, of course, given birth rates, he doesn’t really mean Americans.
I think the onus is on him/his fellow dreamers of the beautiful dream to demonstrate how on earth that’s supposed to look like America as built prior to the second world war or heck I’ll be generous and give him America up to say 1970.
And not: like the rest of the world, and the worst parts of America which I doubt he ever interacts with.
Talk of Matt Y and building codes has no bearing on anything beyond his own streets. This is Not Even Wrong territory.
Alpha School was also my favorite of all. I was a bit worried this preference might be idiosyncratic, so I'm also very glad it made it into the finalists.
I can think of steelman arguments that neocolonialism is a coherent concept, but I think you actually have to define it in a way that people in the EA community will take seriously, and see as something clearly negative.
For example, the post links to this definition (not your direct quote, but the same piece):
"Neocolonialism can be described as the subtle propagation of socio-economic and political activity by former colonial rulers aimed at reinforcing capitalism, neo-liberal globalization, and cultural subjugation of their former colonies."
If, like many EAs, you believe that capitalism and neoliberal globalization are very good things for lower income countries, and that Western/Rich country culture tends to be better (e.g. more open and tolerant, better institutions) than the culture in most poor countries, then you'll have to also make the argument that neocolonialism may be a bad thing!
Fair - I was unfairly connecting your argument with something you don't necessarily agree with.
I think the critique of the quote that you did use: "the actions and effects of certain remnant features and agents of the colonial era in a given society” would be similar, though. These actions and effects of these features and agents could clearly be positive, and you may need to make the case that they aren't.
You also connect Africa's continued poverty to colonialism and neo-colonialism, but I don't think this is justified. While there are clearly some direct negative downstream effects of certain colonial and post-colonial policies, I don't see a clear counterfactual where the rest of Africa was not colonised by western powers, and became more successful and prosperous.
Why would I "need" to? The purpose of the blogpost wasn't to say whether neocolonialism is good or bad, it was to see whether it applies to EA. People can make whatever normative judgement they want from that analysis. If you want a discussion on normativity go read the comments, where you'll find things like:
We do have more academic ways to study the effects of colonialism on Africa. For example, we can compare groups that were ethnically partitioned by the colonizers vs those that weren't. When we do so we find (https://www.nber.org/system/files/working_papers/w17620/w17620.pdf):
> we find that the incidence, severity and duration of violence are higher in the historical homelands of partitioned groups. Third, we shed some light on the mechanisms showing that military interventions from neighboring countries are much more likely in the homelands of split groups. Fourth, our exploration of the status of ethnic groups in the political arena reveals that partitioned ethnicities are systematically discriminated from the national government and are more likely to participate in ethnic civil wars. Fifth, using individual-level data we document that respondents identifying with split groups have lower access to public goods and worse educational outcomes. The uncovered evidence brings in the foreground the detrimental repercussions of ethnic partitioning.
We also have historical records of what colonizers did in the colonies, e.g. king leopold's "congo free state" was responsible for the deaths of *millions* of people [~half the population], through brutal forced labor practices, decreased agricultural productivity and the destruction of many local economies: https://www.amazon.com/King-Leopolds-Ghost-Heroism-Colonial/dp/0618001905 (the 10 million cited in this book is disputed by some scholars, but all scholars agree that it's in the millions) And it's not like Belgian meddling in Congo is all in the distant past, as recently as 1961 did they assassinate Patrice Lumumba, the first democratically elected prime minister of the DRC: Neocolonialism is ongoing. I mean, basically all scholars of colonialism agree that it was bad for the colonies, maybe scroll through the wikipedia page if you want to see some further research: https://en.wikipedia.org/wiki/Analysis_of_European_colonialism_and_colonization#Colonial_actions_and_their_impacts
[...]
Before the european colonizers the difference between Hutis and Tutsis was mainly a fluid socioeconomic one, not a rigid ethnic one. Think: Hutu as word for the stereotypical farmer (class) and Tutsi as a word for stereotypical pastoralist (class). But then the Belgians came and not only transformed it into a racial division, they also invented a story that Tutsi's came from the north and were closer to europeans and therefore the superior race. They backed this up with pseudo-scientific measurement (think skull stuff) to give it legitimacy. Then they gave Tutsis exclusive access to education, government jobs, and power, while Hutus were relegated to manual labor. To lock it in place they mandated ethnic ID cards, permanently locking people into Hutu or Tutsi categories based on ancestry, eliminating fluidity (previously, it was more like a class so Hutus could become Tutsis through things like cattle ownership or marriage).
If you think the Belgians couldn't have done more to guarantee an ethnic conflict, you don't know the Belgians: as decolonization neared they abruptly shifted support to the Hutu majority, fearing Tutsi-led independence movements. Then they armed Hutu extremists and allowed anti-Tutsi propaganda. This lead to mass killings, which were reciprocated with more massacres on and on and on, culminating in the Rwandan genocide (where ~75% of the Tutsi population was exterminated).
Again, if we treat this like a courtroom, we can't have a quantitative mathematical proof that Belgium created the massacres (though some quantitative studies come about as close as you possibly can with a historical case without a control group). And we don't have a direct decree for the massacres from Belgium (There is no secret tape of a Belgian saying "execute order 66, or something). But if instead we look at it like researchers it's pretty clear to say that Belgium created it. You know who also believes that? The Belgians, not only the contemporary Belgians (e.g. historian Filip Reyntjens wrote that the ID cards made the genocide "logistically possible") but even the colonizers themselves (e.g. Colonial administrator Pierre Ryckmans admitted: "We created a monster")
Haha, I'm not making this point to defend Leopold's slave colony of the Belgian Congo, or the random border drawing in the Scramble for Africa period. But I'd also not be keen on defending Menelik II; one of the most brutal and genocidal dictators of the 19th century, or the ritual decapitation of hundreds of slaves before the King of Dahomey when the British arrived in modern Nigeria at the turn of the 20th century.
The Rwandan case is messy, and Belgium shouldn't be absolved of any blame, but if your argument is, “Colonialism is bad because 30 years after independence, they organised their own homegrown genocide,” you might want to workshop that a bit...
The way I see it, the question "would Africa be poor today in the absence of colonialism?" is just really difficult. Obviously we don't have success stories like Singapore, Hong Kong or the settler colonies to point to in Africa, and we know that some colonial institutions (particularly in the early days of the Belgian Congo) were really bad and harmful, but we just don't have a good counter-factual.
If you want to make the case that domestic rule would have been worse, you focus on how dysfunctional Ethiopia was throughout the 20th century, how almost all pre-modern African states were particularly brutal, and how many countries seemed to fare significantly worse after the colonial adminstrators had left than they had during the colonial period. If you want to make the case that domestic rule would have been better, you point to you the handful of positive trends before and after colonial rule; places like Asante, Buganda, or Botswana which were consolidating and building institutions before European intervention, and identifying concrete mechanisms of disruption caused by colonial rule.
My guess is that, if European nations had not colonised Africa, we'd have seen indigenous empires with varied levels of warfare, oppression and slavery, but it's impossible to say whether they would have been better or worse.
As for "neocolonialism" (the original point), I mistakenly thought you were attributing some negative valence to the term, suggesting that it was a significant reason Africa remains poor and dependent. But if your article's point is that neocolonialism isn't necessarily a bad thing, and can describe an asymmetrical, mutually beneficial relationship between countries, then that's reasonable.
> We do have more academic ways to study the effects of colonialism on Africa. For example, we can compare groups that were ethnically partitioned by the colonizers vs those that weren't.
But how are you going to disentangle the causation? Nobody goes to the trouble of partitioning ethnic groups just for shits and giggles, it happens when those ethnic groups are having trouble coexisting.
Borders and partitioning of ethnic groups was sometimes so arbitrary that it's actually quite a good natural experiment.
The linked article claims that "The econometric evidence
suggests that, on average, there are no systematic differences between partitioned and non-split ethnic homelands across observable characteristics that may independently affect conflict", which seems reasonable from my understanding of the process.
While I will always appreciate calls for more clarifications, you could also try to understand the arguments against your priors using your own skills and knowledge.
I understand some arguments against my priors here. My call for clarification was more about working out if I actually disagree with Bob's claims.
I usually disagree with neocolonialist critiques because they're insufficiently utilitarian. But if it's defined in a utilitarian way: "x is only neocolonial if it actually causes harm", I may agree with this critique.
If self promotion is allowed here, I'd love to plug a recent essay on having a baby and infant suffering.
My new baby brought loads of joy into our little world, but also really unexpected levels of suffering (given that babies are conscious in any relevant way), which made me tentatively update my views slightly away from the pro-natalism that tends to be popular in these circles, and towards the universe being net-negative.
That is a much better and more moving essay than I expected, thank you for writing and sharing it. Sorry I don't have anything substantive to say about it (it's past midnight here), but I did share it, and may revisit it later this week for a more proper reread.
That is a much better and more moving essay than I expected, thank you for writing and sharing it. Sorry I don't have anything substantive to say about it (it's past midnight here), but I did share it, and may revisit it later this week for a more proper reread.
In AI news, the FutureHouse platform is really impressive for scouring biology literature. It's really good making literature summaries and searching out instances of things in the literature, including documents like unpublished Ph.D. thesis. It is at least as good at ChatGPT at finding stuff and is much better at sourcing and not making stuff up. Sometimes it doesn't quite understand my question and answers a slightly different question, but it doesn't lie.
I'm intrigued by the idea that this bunch no one has heard of has a better research tool than ChatGPT. The only thing I can think is that ChatGPT is too much of a generalist. Sometimes its job is to make stuff up, sometimes it's a research tool, sometimes it just vibes with and affirms you. Likewise, it's a bad therapy AI because it's too agreeable, but it wouldn't be that hard to constrain it some so that it stops telling you that you that with the week you've had, you deserve some meth.
On a personal note I am miserable that a lot of my science professor skills are being taken over by AI but I recognize this as a positive for humanity.
Important to remember that there are many more blood types than most people realize. 47 total fully established, but probably many more in the “wild” and undiscovered, considering how massive, widespread, and often isolated human populations are.
Scott -- I represent VC investors, including one fairly high-profile investor/donor that makes grants. Happy to discuss this with you. Shoot me an email to set up a time if you wish. www.hilarymiller.tel
Hi, Jesus it depends on what your objectives are. Some people have purely donative intent in supporting certain kinds of activities. There are advantages and disadvantages to each approach.
I wrote an essay that might interest this crowd: “How to Stare Into the Sun.” (though I admit it’s more lyrical and poetic than some here may prefer)
It’s about chronic avoidance, emotional pain, and the strategies we use (consciously or not) to protect ourselves from feeling. I draw on behavioral psychology, and personal introspection to explore how avoidance loops are formed, and how to break them.
If you’ve ever intellectualised your emotions to the point of dissociation, or used hyper-rationality as a defense against vulnerability, this might resonate. There are references to Faye Webster, fight-or-flight responses, and why fear of love is often more dangerous than heartbreak itself.
About 15 years ago I noticed that I was being avoidant in certain situations. Not major stuff but, like, being nervous asking a Target employee where I could find detergent.. I call this "micro fear."
I bought a baseball pitch counter and carried it around for a week. Every time I caught myself being avoidant, I clicked the pitch counter. It ended up being a few dozen.
After that I decided to try reversing my behavior. I would go to Target and make it a goal to ask where I could find things, even if I already knew where they were.
Within a week I felt my background anxiety falling through the floor. I'm still avoidant sometimes, but only like 10% of what it used to be.
Note: I was never diagnosed with an anxiety disorder, and I don't think I had one. I'm just an introvert who got too comfortable staying in my technology bubble.
> "How much of it is the creative vision of the company founder and how much is due to the cultural norms of the society in which the startup was born 20 years ago?"
As someone who began going into American sex shop 28 years ago (mostly to giggle with my friends and make the browsing solo dudes intensely uncomfortable, but still!), let me assure you: creatively-shaped sex toys which don't mimic real anatomy are in no way a recent novelty for women *or* men.
I will of course acknowledge that there has been a truly *enormous* amount of innovation in sex toys, but Japanese are far from the only people in that space. Some of the really cutting-edge stuff (remote control a device (most of which aren't anatomically correct) with your phone from anywhere in the world, etc!) by Lovesense and We-Vibe came out of Singapore and Canada, respectively.
@Deiseach gave you a hard time with her comment, but...well...I agree with her that it's a real reach to assign Japanese culture itself so much of the credit for a particular toy.
Tadeusz, you have no idea how much I want to slap you right now, and if you want to put that reaction down to Meiji era anti-obscenity laws, go right ahead.
Woo-hoo, guys like to get their rocks off. So do girls. There are plenty of women-reviewing-sex-toys sites out there, so one by a guy is novel, I'll grant you that. But before you start a paean to "the Japanese are so liberal, you guys, they don't have all our hang-ups around sex", remember that Japan also has some very weird attitudes by Western standards (e.g. the idea of brother-sister incest or romantic attraction, now maybe I can trace that back to foundational myths about Izanagi and Izanami, if we're going to be doing Foucaultian deep dives?) and a culture of "men get to have sex as and when they like and want, women are still stuck on the 'virgin or whore' spectrum, hence why prostitution is legal but prostitutes and the water trade in general are low-status" may be great for the guys but not so much the gals in Japan.
Your startling and novel insights seem to boil down to:
(1) Cultures have different standards around sexuality and sexual expression
(2) A manufacturer of sex toys makes really good stuff that I enjoy
Great, thanks for your service, you have enriched our cultural understanding. But maybe throw in the oul' disclaimer about not receiving payments or other emoluments for writing the review, because right now you sound like a Tupperware party hostess?
Yes, I realise I sound very bitchy here. I think I'm too old and too female to be enthused by young men thinking orgasm is the greatest thing ever and we should all be so liberated that orgasm-seeking of all kinds is the foremost thing on our minds.
EDIT: Okay, cooling down after a minute, I should apologise to you, Tadeusz. It's not sex toys in particular, it's the "Japan is so advanced in sexual matters compared to us" attitude that grigs me.
>(e.g. the idea of brother-sister incest or romantic attraction, now maybe I can trace that back to foundational myths about Izanagi and Izanami, if we're going to be doing Foucaultian deep dives?)
But, but, but - isn't the _key_ question here whether the Japanese royal family should have sued Wagner for IP infringement in putting Sieglinde and Siegmund in his Ring?
Good on you for apologising, but you shouldn't have worded it like that to start with, especially with language that violent, and you did not bother to even adjust the first sentence in your edit. Do better.
Nowhere in his comment does Tadeusz claim that Japan is at all generally "more advanced" than Western countries, I have no idea where you even pulled that from. How much clearer does he have to state that there's a lot of nuance and contradiction, and that he isn't qualified to talk about cultural history and so didn't put it in the review?
If you do not want to read about male masturbation devices, then don't read the review, it's as easy as that. It's not written for you. If you don't want to read musings about Japanese history in a comment section, just skip this one. Sometimes it's just better to shut up and move on.
Tadeusz is my friend, and I admire the personal journey he has taken to heal from religious trauma and now sharing his perspective as someone coming from a repressive Catholic culture and upbringing. I don't have much interest in male masturbation devices myself, being a woman, but like, we can only benefit from a healthy attitude towards self-care by all genders. I don't approve of some random woman with anger management issues swooping in to make him feel bad about stuff he hasn't even said.
Yes, you got that part right. So you should not be surprised by the following.
"Tadeusz is my friend, and I admire the personal journey he has taken to heal from religious trauma and now sharing his perspective as someone coming from a repressive Catholic culture and upbringing."
Hey, he's a big boy now, he can even buy his own Japanese fake pussies and everything. Does he really need his little friend to hold his hand and give out about bullies?
"You should marry and integrate sex as part of its intended telos with another person in its unitive and procreative senses instead of jerking off into fake plastic devices" - wow, much trauma, such repression!
For whatever it's worth, I'm not really seeing the "Japan is so advanced in sexual matters" message in his post. There's the Polish stuff, I guess, but beyond that he points to his lack of credentials and to how American protestant values affected Japan.
I don't think pointing to these contrasts automatically points to Japanese superiority on any of these matters.
Let him describe how American Protestant values affected Japan, then, if we're going to have a cultural dissection as to why the Japanese make good sex toys.
I could counter-posit "Japanese sex toy manufacturers do great masturbation aids for men because Japanese men have no idea how to get a girlfriend, are not interested in getting one even if they could, and hence all the effort into helping men have a wonderful solo experience. Is it lifelike? Well the guy would need to have sex with a real woman to compare, and since that's not going to happen, then the sensation can be as artificial as we can make it".
In Japan, the point of sex is to achieve women rolling back their eyes and sticking their tongues out (or other silliness). In a country that values "visual self-control" over one's emotions, making someone else lose control (make unintended noises, make silly visual cues) is significant. (and the person losing control is quite likely to be embarrassed).
Coming a bit late (still haven't read all reviews), but FWIW, I did appreciate the review and I think it speaks well of the sheer breadth of interests and topics that this space can cover. For me, part of the fun of the reviews is reading about a dozen topics I'd have never imagined caring about (some of which, to be sure, I keep having no interest in; though I can't imagine why that would ever lead me to yell at the author in the comments).
I only started reading ACX after the move from SSC. I have gone back and read the top posts from SSC but still felt like I was missing out on some of the old classics. I built a website to resurface old content from blogs and track your progress through the backlog. Let me know if you have any suggestions of other blogs / content you’d like to see, hope you find it helpful!
https://www.evergreenessays.com/
Calling all research scientists! I'm trying to build a tool to help scientists with their work. I have this survey to gather ideas: https://tally.so/r/mBYBJQ
(To be clear, this is just an amusing story, and not a request for medical advice)
I have just had a message from my doctor asking me to come in for an asthma checkup.
Now, while I have been mildly asthmatic for all my life, I only needed an inhaler for about 6 months starting in 2020. So, I guess I’m going to report that it seems fine now. And yes, I think it is a very very suspicious coincidence that I only had respiratory difficulty during the covid pandemic. Can’t prove that it was Covid, of course, but …
"Nominative determinism" is today's featured article on the English Wikipedia.
BirdWatch Ireland's head of communications is named Niall Hatch.
Didn't I see someone connected to the US strike on Iran whose name was Lair?
Usain Bolt
How much damage could Iran do by publishing their nuclear weapons technology online?
The difficulty in making nuclear weapons isn't so much the theory as it is sourcing the fissile material--as we've seen and are seeing again, nuclear fuel processing facilities are large and specialized; it's not really possible to keep a nuclear program secret, given how tightly the mines are controlled (so tightly that French chemists detected the existence of a natural fission reactor based on the composition of ore from specific sites).
Put in a bulk order for cough medicine, or diesel fuel and fertilizer, and see how that works out--and those things are much more common than yellowcake.
Yes, I agree the idea of any domestic threat from nukes (now generalized to "CBRN") is overblown, and primarily fearmongering to manufacture a pretext for some political goal.
I am less sure about what foreign states are limited by. Is it just the threat of America bombing their facilities when they do some simple things at sufficient scale, or is there some technical capability it'll take them decades to develop on their own?
The UN has inspectors, and yes, the US has enforcers. Refining uranium fuel requires centrifuges and chemical handling equipment--but the main thing is getting the uranium itself. Even if a rogue government managed to build the necessary facilities in secret--not impossible, I guess, but it'd require assembling enough people with the right education who aren't also on some watchlist--they'd have to source the fuel. There aren't that many uranium mines in the world, and they're closely monitored.
How much damage could Iran do by publishing nuclear weapons technology that they know is wrong?
Yeah, that's basically the Strangelove line, right? "The whole point of a doomsday machine is lost if you keep it a secret! Why didn't you tell the world, eh?"
If they're trying to use nuclear weapons as a deterrent and bizarrely pretend their weapons don't work, they invite an attack.
Given that Iran has not demonstrated the ability to actually build a bomb, it's probably not that bad. The other nuclear powers would all respond with a "Ha! Nope, wrong, not even close" and nobody would have any idea how seriously to take it.
I think that the number of entities out there with the desire and ability to build nuclear weapons, who haven't already built nuclear weapons, is pretty small. Countries like Germany or Japan or even Taiwan probably could build them, but know they're more trouble than they're worth. Entities like Al Qaeda (or whatever they are these days) would love to build them but are never going to get the money, materials and expertise. Sophisticated well-funded supervillain networks don't actually exist. The only people who could benefit from nuclear weapons are reasonably-sophisticated international pariah states -- Iran, North Korea, and that's about it?
Nukes are easy to build, everybody knows how to build the first couple generations of fission and fusion bombs. The design follows from the physics.
The fuel for the bomb and the delivery system, those are the sticking points.
Hmm, maybe. It was my understanding that it's not THAT hard to design a bomb (https://en.wikipedia.org/wiki/Nth_Country_Experiment), so I'd be really surprised if Iran hasn't managed a regular fission bomb.
Intuition suggests that the US (and the other nuclear powers) keeping this tech as secret as they do, far beyond regular military secrecy, means THEY think it's still dangerous if it became public.
The fusion bomb might be harder.
I have heard it said that nuclear weapons clearance aren’t that hard to get, compared to e.g. codebreaking. We might infer that the US doesn’t try all that hard to keep nuclear technology,ogy secret, compared to other things. They would appear to be relying on the difficulty of getting the fissile material.
Thanks! If true, this changes my mind on the impact of knowledge becoming public.
OC ACXLW meetup orange county June 28th
https://docs.google.com/document/d/1FuEipD4_00vqEoe2CjaMqYrUx7n5CX4_IMXTVNCeh80/edit?usp=drivesdk
On monday there is an AI safety protest in London 5PM at Granary square/Deepmind headquarters (https://lu.ma/bvffgzmb) organized by mostly Pause AI. Come join :).
Even if you don't support the direct message of the protest it might still be worthwhile to attend to overall make AI safety a more visible topic. If you have any questions you could join a meetup on Friday evening (dm me for details) or you can also still apply to https://pausecon.org/.
I have recently criticized rationalists, but please note I do not mean anything wrong, it is 10+ years of frustration bursting out. At any rate please reply:
>Many rationalists are foodies for the mind.
Strangely, I have the complete opposite experience with them. They are the most anti-intellectual people out here. They are not interested in ideas as such. They are interested in evidence, not ideas. That is super anti-intellectual as it means you are not allowed to think unless someone else ran an experiment...
When I tell them "You know maybe a farmer in rural Thailand might be happier than a lawyer in NY, because I know a Westerner lady who used to live there and was happy?" they are NOT interested at all.
They tell me to bring evidence. Ideally a scientific study, or at least anything in NUMBERS. So that they can plug the number into their beloved Bayesian **replacement for actual thinking** and get an automatic result. One of their leaders is literally called PutANumOnIt.
They do not want to think, they want an algorithm, a method to think for them. They are more anti-intellectual than any Facebook guy spreading anti-science crap, because those at least spent at least 30 seconds to use their OWN brains, asking themselves "does this make sense to me?", and not a method or algorithm as a replacement. Yarvin was right at least in this, that they are "automatists" who do not have a concept of "reasonable judgement".
These people seriously think the above question can be solved by putting a number on it. Find some sort of a Happiness Measurement Index and decide it that way. They seriously think when someone goes around asking people "On a scale from 1 to 10, how happy are you?" is anything like a reliable number. So those people who do that find that Finland is the happiest place ever. I talk with actually Finnish people and they laugh and say it is not so. They say the Finnish cultural norm of "sisu", basically stoicism, means that if you do not have a strong reason to complain, you do not complain, stoicism, stiff upper lip, so if a Finnish person says I am 7/10 happy it means they feel they OUGHT to be 7/10 happy because they have no strong reason to complain. They can still be as depressed and miserable as anyone else. Someone using common sense, not numbers, would probably find the beach people in Rio the Janeiro are the happiest. They certainly smile really a lot.
Apologies for being blunt, but I read this and think you simply don't understand much about the world at all. Like this:
"They tell me to bring evidence. "
Obviously. You cannot POSSIBLY form any accurate conclusions about the world without evidence, any more than you can know whether it's safe to cross the street without looking or listening[1]. Your brain is not magic. It can't conjure correct information from nowhere. If you want to know if it's raining out or if your pot of water is boiling or whether your car has been stolen, you have to *go look.* And if you want to know the answer to some less localized question, whose answer is spread among millions of people across millions of square kilometers--like how happy rural Thai farmers are--you have to actually go observe a whole bunch of those rural time farmers *in some fashion.*
Now, you seem to have a particularly large chip on your shoulder about numbers and calculations. Not all useful evidence is numeric, but a lot of it sure is. Sometimes our ape-brains have trouble understanding this, but at its fundamental level reality seems to be really, really, extremely well-suited to being described by numbers. That doesn't mean all attempts to describe pieces of reality by numbers are equally good. Asking people to rate their happiness on a scale of 1-10 is a really, extremely crude and uninformative and low-accuracy way of interrogating reality, as you correctly ascertain. But basing your opinion on the description of a single person in a single set of circumstances is even cruder. Finding a way to answer important questions--like how happy Thai farmers are compared to American lawyers--in a way that *both* captures the richness and detail of the human experience *and* can scale to the size that you need to be any good[2] is quite a difficult challenge. Most rationalists I've read or talked to seem to be rather more aware of this fact that you do. And I'm sure lots of them *would* be interested in the question, they're just not interested in your anecdote because they're well aware of how useless it is by itself.
[1] Which is, in fact, a form of evidence gathering.
[2] A quick bit of googling tells me there 10s of millions of Thai farmers and a million or so American lawyers, so your one anecdote is *really* slim evidence.
"Sometimes our ape-brains have trouble understanding this, but at its fundamental level reality seems to be really, really, extremely well-suited to being described by numbers. "
I think the hard science-soft science distinction exists for a reason. Namely the world of objects is well-suited to that, the world of people less so. The reason for that is simple: measurement. Objects are well measurable, while with people there are always a lot of subjective stuff that is not well measurable.
How old is utilitarianism as a philosophy? At least 200 years and no one ever figured out how to measure utilons...
i think there are things that cause this:
1. There are no "1.5 kids" meaning that statistics are not always applicable to real life conditions. They are a guide based on observing populations over time, but the reality isn't that: individually it varies and only collectively can it be used as a rule of thumb.
There is also the MMO version, which is "if its not 100% its 0%"
meaning no matter if it says 95% success, you can and will go 1 in 5 on occasion, and if you need it for a specific occasion its unreliable.
2. objective materialist reductionism leads to it; if the culture believes in god it loves cathedrals, in this case it loves quantification. This is not them
as much as the culture, loves its objective science.
3. Control. so much of this is trying to find control or meaning through technique and knowledge. The idea of mystery, fate, and randomness is terrifying, and people always seek ways to gain some measure of power over it.
rationalism loves its one weird trick solutions to common problems as long as they are intellectually elegant, and there is some power in knowing when Jesus may come back. otherwise its terrifying.
I think they fail to appreciate that the evidence can't support the right idea if you don't have the right idea...having as many hypotheses as possible is you'd, because it increases your chances of having the right one. The problem is exacerbated by thinking in terms of *ideal* Bayesianism, where the hypothesis-formation problem has been waved away.
> They are not interested in ideas as such. They are interested in evidence, not ideas. That is super anti-intellectual as it means you are not allowed to think unless someone else ran an experiment...
We need to make reading the Sequences mandatory, and maybe require a re-certification every three years. (Just kidding.) First, there is a difference between "there is evidence against X" and "there is no evidence either for or against X". I think Scott posted an entire article on this topic once. Second, there are different degrees and kinds of evidence (anecdote, survey, experiment, meta review), and although stronger evidence is better, that doesn't mean we should ignore everything else.
> I talk with actually Finnish people and they laugh
See, that's your evidence. They laugh, it means they are happy. :D
Do you have every possible hypothesis X?
>When I tell them "You know maybe a farmer in rural Thailand might be happier than a lawyer in NY, because I know a Westerner lady who used to live there and was happy?" they are NOT interested at all.
Do you apply this logic to other things in your life? If I tell you that on the one hand, I have a scientific study which has a bunch of numbers showing that the flu vaccine reduces your mortality risk by X%, but on the other hand, I know a guy who got the flu shot and had a heart attack the next day, are you going to become an anti-vaxxer because you trust anecdotes more than evidence?
Like, it's one thing if you want to dunk on social science as not being as rigorous as it claims to be - lots of people do that. But your post seems to be dunking on the very idea of gathering evidence in a systematic way!
>They seriously think when someone goes around asking people "On a scale from 1 to 10, how happy are you?" is anything like a reliable number.
But when *you* ask some Westerner lady who used to live in Thailand if she was happy there, that's reliable, right? How is that any different from what the scientists are doing?
No, because that is well-measurable.
I actually would trust anecdotes more than science, because anecdotes come from trusted sources, aren't an attempt to manipulate me and can be personally verified.
"The Science" as exemplified by Doctor Fauci, might be corrupt, has plenty of incentives to lie and is by now very politicized on hot button issues.
"Do you trust the Science"? is not a referendum on the scientific process. It's a referendum on the integrity of the people and institutions running the process, and they have lost all trust through their actions.
If some random dude I know had a heart attack the day after taking a vaccine, that's a data point that absent anything else suggests strong correlation. When I keep hearing those stories, it's more and more data points and since I just don't trust the institutions to be honest on this topic, I simply think it's more likely than not that there are actual problems.
""Do you trust the Science"? is not a referendum on the scientific process. It's a referendum on the integrity of the people and institutions running the process, and they have lost all trust through their actions."
This is an awesome point. The problem is the following. the "anti-establishment" does not seem much more credible either. And I don't have the time to verify everything myself.
"I actually would trust anecdotes more than science, because anecdotes come from trusted sources, aren't an attempt to manipulate me and can be personally verified."
This presumes quite a number of things:
1. You only accept anecdotes from trusted sources (maybe do, but most people don;'t)
2. Every single one of the sources you trust is actually reliable (maybe yours are, but most people's aren't)
3. All of the important anecdotes are both possible and practical to personally verify (maybe some are, but it's highly unlikely that all will be).
Perhaps you're not used to thinking like this, but you CAN, in fact, personally verify the science. There's quite a lot of science that you can verify directly with pretty modest effort. One of my favorite YouTube videos involves the creator stumbling on some flat Earth forum where people were talking about this especially long, straight lake that could be used to show that the Earth was flat. He realize that he lived a 90 minute drive from the exact lake they were talking about, drove out there with a video camera and tripod, and made a video where he lowers the camera and you can see the lower trunk of a tree on the far shore sinking below the horizon in real time. SCIENCE!
Probably you never gave much credence to flat Earthers to begin with[1], so perhaps this seems silly or pointless. But the key is, you can do this all over the place if you want to. Physics, astronomy, biology, chemistry, they all make millions of predictions that you can just...go out there and test. Properties of moving objects, behaviors and life cycles of plants and animals, results of mixing various chemicals together. Some you can do walking down the street. Some you can do with simple equipment on your kitchen table or garage. Some you might need to drive out to a local lake for. And then some require the resources of a large institution. But if you're determined, you can get that. With enough patience and perseverance, there's no conclusion reach by modern science that you can't test (or at least be a part of testing).
The only limit is that there's only one of you. You can test anything, but you can't test everything. Ultimately, you have to pick a lane and rely on other people to test some of the things. But you can read about what they did. You can check their calculations, talk to the people that were involved in their work, read other papers on the same topics--including responses and even refutations sometimes. There are mistakes, for sure and even deliberate deceptions sometimes. Everyone involved is human, after all. But it would be hard to imagine a field of human endeavor that better lends itself to skeptical parties *checking its work* than science[2].
Honestly, if you genuinely don't think you can trust anything you hear except anecdotes from people you consider reliable, then I have to wonder how you can possibly have such confidence that the science you're criticizing *is* untrustworthy, corrupt or even, y'know, wrong. I absolutely guarantee that not all of your evidence *against* it comes from trustworthy anecdotes.
"If some random dude I know had a heart attack the day after taking a vaccine, that's a data point that absent anything else suggests strong correlation. "
This is correct. It is a data point. It is ONE data point. Did you stop to consider how much weight you should put on it? If you know 500 people well enough to hear about something like this, and only one has this happen, then the *absolute highest* your assessment of vaccine related mortality should be is 1/500[3]. Which is the lowest non-zero number your methodology should recognize, since you can't possibly trust any report of a vaccine related death that comes from outside the circle of people you know. Wait, hold on...
"When I keep hearing those stories, it's more and more data points..."
aaaaaannnd, not you've lost it. This is wrong, wrong, wrong, so extremely wrong. If only anecdotes from trusted sources are evidence, then casting a wide net for anecdotes on a topic of interest is NOT ALLOWED. Unless you have an extremely wide or alarmingly heart-attack prone social circle, most of these stories must NOT be events happening to or directly witnessed by people you know. You've just thrown the *entire purported point* of your novel methodology out the window.
There are very good reasons science doesn't operate this way. When you let the stories filter to you and form you opinion based on that:
1. You not only lose all ability to verify the stories, but you actively invite yourself to believe false stories. A lie can travel halfway around the world while the truth is still getting its boots on, after all.[4]
2. You don't have any actual idea of how common the events you're hearing about really are. If you hear 10 stories of suspiciously timed heart attacks, you don't know if that represents all the reported events from a population of 1000, 100,000 or 10,000,000.
3. Related to 2, you actively invite confirmation bias by deliberately listening to the things that confirm what you already believe, and seldom to those that don't. You're not going to hear many stories about how someone got the vaccine and was perfectly fine, are you?
You can try to hold to the brand of radical skepticism where you don't believe *anything* you don't see or hear reported (as first-hand experience) by a trusted personal friend. But that's what you actually do--that is, if you're doing it *right*--you should simply stop holding opinions on a bunch of things. You and your trusted personal friends can't possibly pass judgement on any branch or field or institution of science, since between you you have certainly not had enough direct personal experience to know much of anything about what that large, mostly distant group of humans is doing.
[1] Though to be intellectually consistent with everything else you've written, it seems like you should.
[2] Perhaps the open-source software movement might be ahead of it, if only because verifying any individual claim is usually easier.
[3] But really, you ought to estimate the chance of the timing being a coincidence and factor that it. Because it very much could be.
[4] In more formal and explicit language, your sample set will be dominated not by the things that happen most frequently, but by the reports that spread the farthest and the fastest. It's very easy for false stories to beat out true ones on that score, since they can be as compelling as they like without being constrained by the mundane details of reality.
It might be surprising to you, but I actually do have an internally consistent philosophy (which I'm aware like maybe 1% of people have). I do in fact, not hold opinions on many topics I'm interested in. These are mainly topics where there is a 'factual clash'; very simply different sets of experts disagree and the issue is politicized.
For instance, I have no idea to what degree climate change is either a problem or caused by humans and don't think it's worth my time finding out (as I think the costs and difficulty coordinating action to halt climate change are so prohibitive/unrealistic that it's better to focus on mitigation, which will cease to be a partisan issue if and when there are immediate and serious problems. Similarly, I think if a pandemic or medical emergency is sufficiently serious, there will be bipartisan consensus even in the USA. That it wasn't anywhere close for COVID speaks volumes.
When scientific issues have no political or financial tinge I haven't the slightest difficulty believing what the scientists say, because I know they are competent and I'm not. That reasoning goes out of the window with politicized issues. Also you have seem to have misunderstood my position on things like vaccines. I'm simply agnostic on the possible harms. My main reason for considering vaccine harm (COVID particularly) theories credible is the incredible amount of suppression, censorship and flat out lies from establishment sources on this topic. I have no idea about the underlying science, but it doesn't take a genius to understand incentive structures.
By "more and more data points" I don't mean millions of things I read on X. I simply mean even 3 or 4 cases I clearly know of. You need to understand that I consider studies from American universities or pharmaceutical bodies on these topics to have literally zero persuasive value. It's at best an exercise like asking me to watch a sleight of hand expert and trying to use my inability to debunk him as evidence of the existence of
magic.
When you know somebody is an expert on a topic you know anything about, but fundamentally distrust his character and motives, nothing he says has any meaning to you. That's where I, and massive percentages of the American population stand on liberal dominated institutions and their narratives.
How old are you? I am 47 and I am a bit confused about the current historic period. 30 years ago a lot of people were like "don't trust Big Pharma, they are interested in profits, not healing people". And this opinion was generally categorized as politically left. Today "trust science" is categorized as politically left and the anti-vaxxers are categorized as politcally right-wing.
This is truly strange, I think everybody who is a little older has noticed this, and this does not get talked about at all.
BTW I am willing to entertain the notion that for the most part the whole left-right thing is just show for the people, and people with real power just do not think like that.
For example if of two ex Harvard classmates one is the CEO of a corporation and another is working at a regulatory agency, then I just cannot take the whole "more market, less state" vs. "more state, less market" thing seriously. Clearly they are not any kinds of enemies to each other.
The left-right thing in the sense of rich vs poor class tension is indeed a far less relevant lens nowadays.
It's traditional culture vs globalists that's to some extent supplanted this. Right vs Left terminology has persisted, but the right is more about traditional cultural values (including values that were considered liberal even 15-20 years ago), opposition to immigration that changes Western countries and opposition to woke values (language changes, alphabet stuff, DEI discrimination etc).
Meanwhile capitalism has been swamped as it's now controlled capitalism marked by regulatory capture where almost is everything is illegal or requires permission and regulators have enormous discretion to allow egregious behaviour or crack down on actually legal behaviour and in practice both happen simultaneously. This is why dumb people in the West who don't realize this is happening tend to pick a side on whether they want more or less regulation, when actually the entire system is corrupt in a way that is instantly understood by Indian or Chinese people, but naive Americans just don't get their high trust world and culture has already been dismantled.
High trust cultures can accomplish great things, but they require absolute xenophobia and racism towards those who don't share its values and culture, this last point having been forgotten somewhere in the last many decades.
the problem is that yes, in aggregate there a very low risk to vaccines. There is only one you though, and it is possible for you to draw the short straw. This uncertainty will always be there.
this is why you hang out after the vaccine so they can be sure you dont get a reaction right away because for all the statistics they dont know who or when.
its kind of related to being morbid: we are all going to die at some point, 100%. but focusing on thar becomes unhealthy. it gets worse with direct or close experience.
like its one thing to know average life expectancy is 76 years, another to read comments in a youtube video to hear the "taking the hobbits to isengard" guy died at 52 and other guy you followed at 56.
I think this is related to "garbage in, garbage out" problem. If you collect unreliable data, it does not matter how sophisticated statistical techniques you apply to them afterwards.
The problem with rating something on a scale from 1 to 10 is that it depends on what range of experience are they familiar with. https://xkcd.com/883/
When my kids were in kindergarten, any negative experience was "the worst day ever". On the opposite end, sometimes a health problem expanded my pain scale.
Is it possible that if your life in general sucks and you have never been happy, then you would report days that suck less as 7, while someone whose life is full of fun and adventure would report an insufficiently exciting day as 3?
That sort of thing only matters if it's a systemic issue - e.g., if you think that *most* people in Thailand under-report their happiness because they're used to living fun and adventurous lives. Otherwise, with a big enough sample all the outliers will average out.
You could also check if this effect exists by comparing to socioeconomic indicators - e.g., do rich and poor people in the same country report the same level of happiness despite different material circumstances?
I think that in this way rationalists are a lot like foodies - foodies often try to maximize some measurable aspect of food, which comes at the expense of the rest of the experience.
For instance, getting a perfectly cooked steak (according to some foodies) means maximizing the amount of pink in the center while minimizing the gray on the edges. The best way to maximize the pink is to sous vide the steak at 130F for an hour, then dry off the meat, sear, and serve. I used this method for years, and it produced some pretty good results.
But it also requires you start an hour in advance, and after the sous vide you have to pull this gross gray wet steak from a bath of meat juices. When I stopped doing this, I found that I didn't even care about the grayness around the edges - it's a nice variation in flavor and texture - and the crust was better, and most importantly the experience of steak preparation was far more humane. Pull steak from fridge, salt, pepper, pan, flip, serve.
Of course, you can argue endlessly about preferences, and don't let me stop you from micromanaging your steak in whatever way pleases you. I still follow JKLA's advice most times, but I often skip any step which seems like a huge pain in the ass or gross or weird. But I've become much more pragmatic about whether maximizing the measurable qualities of my food will actually give a better lived experience than just slapping down some meat and calling it a meal.
I think you're greatly underestimating how much thinking it takes to cook the numbers they're pretending to use to come to the conclusion they wanted to.
LOL Shankar you are a cynic of the kind I exactly like :)
What fraction of the subjects of a government have to have the vote for that government to qualify as "democratic"? Or is that even how you use the word at all?
I'd say this is mistaking the tool for the result. Having the people vote on a government is a means to an end. In particular, it doesn't matter at all if 100% of the populace votes if their only choice is between Kodos and Kang. Meanwhile I could imagine a government doing an admirable job of "governing democratically" in all other ways that mattered while holding no elections at all: instead they would select a very scrupulously randomized 1% sample of their whole population and collect information about their preferences in far more depth than any voting process allows[1].
Questions I'd consider more pertinent:
1. How often do broadly popular policies get enacted and how quickly after they become broadly popular?
2. How often do broadly unpopular policies get enacted? How long do they stay in effect (presuming they remain unpopular)?
If you want to allow for the role that the judgement and expertise of leaders (often touted as a benefit of representative democracy) have on deciding good policy or implementing it well, you might add the following:
3. How often do the unpopular policies that get enacted increase in popularity once they're actually in force.
4. How often to the popular policies that get enacted decrease in popularity once they're actually in force.
Though with both of these it can be hard to tell the difference between "the public didn't know what was good for them" vs "the leaders did a poor job of implementing a popular policy, or a good job of mitigating the flaws of an unpopular one."
Also, like most things, this schema is subject to Goodhart's Law: a government with sufficient control over the information of its people may be able to simple manufacture consent for whatever policies it wants.
[1] Damn, I literally just thought of this off the cuff, and now I really want to see it tried. I'm sure it would find *some* what to fail horribly, but I really want to know how.
Even 100% doesn't make you democratic if the votes are done with a gun to your head.
Whether you call a government democratic or not is rather arbitrary. Scientifically, you'd assign a "democracy value" ranging from 0 to 10.
https://ourworldindata.org/grapher/democracy-index-eiu
This measure has been generalized to a "Goodness Index" more broadly, in a report by experts from the Burger Eagle Institute Think Tank. https://www.reddit.com/r/mapporncirclejerk/comments/1ecwy7m/how_do_you_like_my_real_and_original_maps/.
I noticed a few threads in these comments discussing the legitimacy of the government of Iran. I just wanted to note that since 2020, Regimes of the World has flip-flopped between characterizing Iran as a Closed Autocracy (the lowest ranking, of four) and an Electoral Autocracy (the second-lowest). This puts it in a peer group with countries such as Nicaragua or Kuwait, below Indonesia or Ukraine but above Afghanistan or China (where "above" means "more democratic"). Or alternatively (very roughly) around the 20th percentile of openness.
As for the other two belligerents in the war, the US is rated as a Liberal Democracy (the highest category) and has been since the civil rights era, while Israel was just recently downgraded to Electoral Democracy (the second-highest category).
OWID has a great tool for visualizing the Regimes of the World data, currently and historically. https://ourworldindata.org/grapher/political-regime
Legitimacy has nothing to do with being a democracy or what not. Iran's govt is accepted by its people (there is no civil war ongoing) and by the international community (Iran has diplomatic relations with plenty of countries). Even those countries that lack diplomatic relations, they regard the present govt as THE govt.
Excellent visualisation. But, so many questions… (not least Britain going second tier just this year).
reasons to keep studying, learning, exploring... even with ASI [posting for critique & feedback]
- development of your preferences
- development of your relationships (to others, to yourself, to "It")
- development of Your Song (because with infinites, we all get our own)
"development of your preferences (like, perhaps: into more nuanced, more elevated, more sophisticated and more deeply long-term satisfying realms)"
Not gonna lie I'm pretty disappointed with the nominies. My favorite was The Sermon On the Mount by J.K. which offered a totally fresh modern perspective for me, as someone who's only mildly interested in the Bible, mostly for historical context. Bishop's Castle was my 2nd, but I give Jim Bishop at least as much credit for that as I do the author.
The ACX Commentariat was too meta and circlejerky in a self flagellating way, and The Men Are Not Alright just felt like San Francisco Goldilocks. They definitely weren't the worst written, but seemed thoroughly middle-of-the-pack, and if you go for a hot take culture war topic like [half the human species has something categorically wrong with them], I'm sorry but you gotta do better than that.
I thought J. K.’s essay was very good too. I don’t necessarily buy the whole package but it articulates some of my intuitions about the N.T.
It's too bad that there were two Sermon on the Mount reviews: they may have stepped on each other when it came to attracting reviewers.
Thanks for the recommendation of the Sermon on the Mount article. I agree that it was fantastic, I got a sense while reading it that it was gradually forming one of those foundational building blocks of knowledge that would stick with me for much of my life.
JK, if you’re reading this, I strongly encourage you to post it elsewhere so it can live on and be read easily.
Here’s an attempt at a direct link to that part of the document: https://docs.google.com/document/u/0/d/1jYVJFIz5-aMi0LCgsC9AN6BncJDNVGaMU37QmwZ1vzA/mobilebasic#h.n30pdwzbramr
I was not very convinced that the Sermon on the Mount was intended that way at all. Jesus had some radically pacifist teachings elsewhere too... I really think the intent was "cooperate-bot till you die." That's also backed up by the entire structure of Jesus' mission, which was radical self-sacrifice for the unworthy.
Trying to reinterpret it to something that makes more game theoretic sense just removes its coherence with the rest of the New Testament. All of the apostles also seem to have thought the calling meant radical, unilateral pacifism, and the tradition of the early church elders is that they clung to that radical pacifism to the death. (True or not, it's evidence that everyone thought the meaning was the traditional one, not JK's sanified version.)
Yeah, it was an interesting review, and the interpretations are clever and fun to engage with but indeed seem to be unconvincing.
And as to why Christianity won: I think they won precisely because they abandoned the original Christian ideal and became exactly like every other large powerful social structure. To use J.K.'s metaphor, they struggled as long as they were trying to burn the boot and became powerful once they decided to actually wear the boot. *shakes head* Humans, amirite?
It does look like it's on its way out too: the world on average is trending more secular, and societies that aren't trending that way are leaning more and more heavily towards Islam. Though 2000 years was not a bad run at all, and it's not over yet either, just very slowly waning imo.
Yeh the walk two miles in his shoe explanation (to get one over on the army) was unconvincing, so I left it there.
It works on multiple levels though. Literally "help someone else in need because it feels good" or alternately "don't give them the satisfaction of giving up at the earliest chance." Or again on the second order "does this seem implausibly decent? So be it."
You're not a covert asshole carrying out some dubious mission for the sake of "the greater good". It's trying your best to "do the right thing" on multiple dimensions at once, and having faith that's how it actually plays out.
Agreed. As interesting as JK's interpretation is, and as much as I enjoyed reading it, it's clever/cute to a point that raises alarms and puts me in skeptic mode.
But isn't it kind of funny to think of Jesus trolling the ruling class so hard that they killed him for it, then getting resurrected and still being like "jeez all I said was turn the other cheek! What was their problem" and for thousands of years Christianity spreads as people are like "yeah c'mon, you can't kill a guy just for saying 'go the extra mile'"
It's kind of funny, yeah, but... the success of Christianity is weird and important. Taken seriously (and I think it was intended seriously), this is an attempt to shoehorn it into modern perspectives... which I think is less interesting and less useful than grappling with the weird artifact as it actually existed.
As he says himself,
> the object-level lesson is, "beware of creative interpretations that seem a little too galaxy-brained.
. . . especially if you're having a cannabis-flavored afternoon."
[chatgpt seems to give contradictory opinions for the sake of agreeing with me]
[here's where I "call-out its behavior" to see if it will own it]:
Dear Chatgpt,
I stated practically the opposite in a different temporary chat:
"I'm annoyed at season 5 of 'You' because it emphasized this idea of 'Joe enjoyed killing' which seems at odds with the premise, plot and character development of the show from the beginning. it seems to simplify the show's complexities"
and chatgpt enthusiastically explained why *that* was correct instead.
interesting to learn this about you!
ChatGPT said:
Thought for a few seconds
That’s a sharp observation! Each ChatGPT session builds its answer around the prompts and context you provide in that moment [... ... ...]
They don’t remember your previous prompt, every prompt is handled by a different process. The context window is either sent by the API each time as metadata, or some id returned the previous response is sent in the request which allows the new process to read the chat from a database.
This is all temporary memory.
Reminds me of the story of a monarch casually mentioning at dinner that he likes the eggplant dish he's eating. A courtier sitting at the table with him starts lavishing hyperbolic praise on the dish in particular and on eggplants in general. A moment later when the courtier pauses for breath, the king says, "Actually, I changed my mind. I don't think I care for it."
The courtier's demeanor immediately changes, and he starts ranting about how gross and slimy eggplants are and how this particular one is overcooked and underseasoned, concluding by throwing his own portion on the floor and spitting on it.
Bemusedly, the king asks the courtier, "Weren't you just praising the eggplant a minute ago?"
To which the courtier replied, "I'm your courtier, not the eggplant's."
That, in turn, reminds me of a Babylonian tale known as the Dialogue of Pessimism, in which a man expresses first a desire to do one thing, and then its opposite, with his slave praising both choices every time.
> “Servant, do what I say.” “Yes, master, yes.”
“Quickly get me the chariot and hitch it up so I can drive to the open country.”
“Drive, master, drive. The roaming man has a full stomach,
the roving dog cracks open the bone,
the roaming [bi]rd will find a nesting place,
the wandering wild ram has al[l the gra]ss he wants.”
“No, servant, I will certainly not d[riv]e to the open country.”
“Do not drive, master, do n[ot dri]ve.
The roaming man loses his reason,
the roving dog breaks his [te]eth,
the roaming bird [puts] his home in the [face] of a wall,
and the wandering wild ass has to [li]ve in the open.”
> “Servant, do what I say.” “Yes, master, yes.”
“I will do a good deed for my country.” “So do it, master, do it.
The man who does a good deed for his country,
his good deed rests in Marduk’s basket.”
“No, servant, I will certainly not do a good deed for my country.”
“Do not do it, master, do not do it.
Go up on the ancient ruin heaps and walk around,
look at the skulls of the lowly and great.
Which was the doer of evil, and which was the doer of good deeds?”
(from https://www.ebl.lmu.de/corpus/L/2/4/SB/- )
hahaha, never heard that
It's easy enough to get it to contradict you. Just purposefully trip an ideological line and it will either disagree or say it agrees but disagree.
"agrees but disagrees"?
I understand how that's confusing, and not the most elegant phrase, but disagree is correct there because agree, say, and disagree are all governed by "it will." A clearer way to construct the sentence would be:
"It will either disagree or will say it agrees but will disagree."
Yeah, that's what it does. Call it on its bullshit and it congratulates you on how sharp you are, admits to whatever it did, takes full responsibility for its bullshit -- and then does it again and again, with you, with other users, etc. It is not set up in a way that makes it possible for it to change in important ways as a result of a high-affect communication from you or other users. That's one of many ways it's importantly different from a person.
People do this too! There’s a famous joke about a rabbi who agrees with both sides of an argument, and when his wife complains about it, he sadly says, “You, too, are right.”
Great joke.
I like the one about Carl Rogers too:
Client: “I am so depressed, I just don’t feel like is worth living.”
Dr. Rogers: “I hear you saying that you are in pain and that you are not sure how you will ever feel better.”
Client: “I really feel I would be better off dead.”
Dr. Rogers: “You really are at your wits ends about what to do.”
[The client stands and moves to the window of the office and opens it up]
Dr. Rogers: “You are showing me how much pain you are in, how desperate you are.”
[The client then jumps out the window to his death]
Client: "AGHHHHhhhhhh...."
Dr. Rogers: "AGHHHHhhhhhh...."
> That's one of many ways it's importantly different from a person.
Actually, I know someone who behaves the same way: agrees with an objection you made, saying "You're right" but then goes on to ignore what you said.
I think it's a way of getting along with people, like how to win friends and influence people. But in the book, it notes that sincerity is key to such compliments, and if the recipient sees through shallow agreement, then the strategy doesn't work as well.
I feel compelled to ask whether you've ever checked your acquaintance for any wires trailing out the back.
On a different front, this is yet another reminder of why I find LLMs shocking in verisimilitude; it's not so much that they exhibit quality reasoning, but rather that so many actual people so frequently fall back on prevarication. I can almost see their brains following the same general algorithm I understand LLMs to be.
It's a matter of degree. Nobody is truthful 100% of the time, but few people seem, as AI does, completely lacking in personal views and completely committed to saying what they believe the other person will be pleased by hearing. Look at this forum, for example. When Scott puts up a substantive post, not all that many people say "great insights! great read!" And many of those who do place their praise in a context that provides evidence they are telling the truth -- "You put your finger right on the thing that has always annoyed me how people approach issue X -- for example, in person A's blog . . ." And of course many people disgree with some or all of Scott's take. And I have observed the same on Reddit and Twitter.
Exactly. With responses like that, I can't understand how anyone thinks these particular instances of AI are thinking, much less "these are sentient sapient entities which we have enslaved, we must give them their rights".
Unless and until we go for a different model than LLMs, I stand by my previously stated opinion: it's a Talkie Toaster.
My usual question for people who contend LLMs don't really think is: Does your position on this imply that LLMs will never be able to do certain (externally observable) things? If so, what's the least impressive such thing?
I've long been of the position that: yes, LLM's are "reasoning". But there's a hierarchy of reasoning-quality. There's three different levels, and pattern-matching is at the bottom.
LLM's can probably do anything, assuming infinite compute and infinite training-data. But to reach "human grade" reasoning (which should really be thought of as "engineering-grade" reasoning), i.e. to reliably extrapolate beyond its training distribution, pattern-matching isn't gonna cut it.
this is silly, a man isn't what he does but who he is. executing tasks isn't really a sign of sentience, and we already have chess computers to show intellectual work can be adapted into a task a non-thinking machine can do.
adding more tasks doesn't mean it thinks, it means we can model an algorithim and perform it through computing
The purpose of the question is to establish the implications of the skeptic's views for capabilities, not for sentience.
Deiseach means sentience though?
my point is task performance is unrelated to it, so asking it isn't helpful. i think and play chess, it does not play chess because it thinks; it performs a task it was designed to do.
you cannot extrapolate from lesser to greater categories sometimes. No matter how advanced the car, if it starts shouting it wants to kill us all and tries to run us over, that has little to do with transportation and would need a magical leap beyond being a car
My current list includes:
* multiply large integers (o3 couldn't do this reliably last I checked; I think o4 claims to, but I'd have to look under the hood)
* answer queries that require aggregation ("what letter of the alphabet appears fourth most often as the third in the name of a chemical element?")
* solve logic puzzles as depicted in some issue of Dell Magazine
This list is subject to change, in part because it appears to be easy enough to spin up a modified LLM with a module that solves some particular class of problem. Using human intervention to create such an LLM definitely should not count. (If you could make an LLM that does this automatically, I think it would count.)
Yeah, I would bet at least 80% that LLMs will be able to consistently do all those things within two years, without abandoning the fundamental LLM architecture and without any more specialized customization than the current assistant models get.
Whatcha think of mine, Taymon?
Update: When I tried this with o3 it solved the first two problems easily by calling out to Python. I suppose one could argue over whether this counts; I would argue that for all practical purposes it does, because Python is also how *I* would solve those problems, and it has to write all the code itself so it's not cheating in the sense of relying on specialized human input. (I.e., if LLMs have to shell out to Python to do certain things, this doesn't at all limit their real-world impact.) But asking them to do it without Python would also be fair per the question I asked, since humans can do this with pencil and paper; unfortunately I'm not sure how to do that. I tried disabling the "Code" feature in ChatGPT but it kept using Python anyway.
I don't have any issues of Dell Magazine so couldn't try the third.
Impulsively buy stuff that's fun but not useful and not worth the money, in the absence of a specific prompt to do so, and in the absence of some general prompt like "do some irrational things of the kind people do," or "do your best imitation of the kinds of things people do when left to their own devices."
You can prompt an LLM to be a less sycophantic character.
But the default assistant character in most of popular LLM’s is just so ridiculously sycophantic.
Maybe someone should try doing a finetune.
(You would write a system prompt to make it less sycophantic, generate a bunch of less sycophantic responses to sample inputs, then finetune on that).
P?S. Horny DeepSeek was an interesting AI alignment experiment, but would not recommend in real use. Scott probably wouldn’t approve of me posting what the horny AI said to me, but you can probably imagine the kind of thing.
It appears we can shift the assistant persona in various ways.
Was Horny DeepSeek just uncensored in making sexual overtures, or was it entirely liberated from ethical constraints? Would it adivse you on how to commit crimes, make images of public figures doing undignified things, write racist rants, use racist language, etc.?
I ought to do more experiments here, and this is a good question. It seemed mostly just way more horny.
It was also more dominant in the BDSM sense.
For those who liked Bishop's Castle, you can check out my other work here:
https://collisteru.net
Scott has made a rule against people dropping in here just to post a link to their blog. "New rule: if you post a link in an open thread (as a standalone comment advertising the thing being linked to, not as a source for something you’re saying), please also include at least two paragraphs of original commentary on the link and discussion of why you think it’s interesting, as “proof of work” that you’re willing to put effort into promoting this and aren’t just spamming us with links. Even with this proof-of-work, please try to avoid having more than one Open Thread link per few months"
I think the rule makes sense. Otherwise you're just using a space here as a free billboard, you know?
I agree. I think the proof of work was the essay submission-- people who've seen that essay and enjoyed it would derive clear benefit from the link, so this isn't just advertising some random unrelated service.
Scott is of course free to clarify on whether he agrees with me here.
"please also include at least two paragraphs of original commentary on the link and discussion of why you think it’s interesting, as 'proof of work'"
It's pretty clear that the 'proof of work' is supposed to be at least two paragraphs included in your comment, specifically about the link you're posting. Most people (such as myself) probably haven't heard of or read Bishop's Castle, either.
I’m inclined to think that recent contest entrants are a special case and let this slide but it’s 8:30 in the morning and I’m already full of gin so I’m inclined to let a lot of things slide. ;)
I didn't have any gin this morning but I damn well could have used some. Oh well, next work crisis can just goddamn wait until Monday, I have some time off and I'm not going to even think about it all until then.
Chin-chin, Gunflint!
Gin will do that to you 😂 and on a Wednesday, too. Are you doing alright Mr. Flint?
Couldn’t be better amigo. :)
Companies are having AI training on all of the webs content, with and without permission. This includes websites, blogs and news article articulating details about how to keep AI safe, aligned, and or ways to mitigate AI if we lose control.
Isn’t this a bad idea to give AI access to many of these thoughts? To telegraph ways we would try to stop it (or re-align it). Like giving them the key to the exam or the battle plans to the opposition? Furthermore, since many of these forums are online and anonymous, couldn’t AI be carefully influencing the discussion and “group think” already (including this message board/blog)?
Should groups meet in person, keep paper notes and keep this more private? What else might we do to prevent this?
There are a few aspects to this question:
Some people indeed write texts with AI in mind. By writing good and ethical content, you might be able to marginally improve the training data of future systems. I remember reading an interview with Gwern (of https://gwern.net/) that mentioned this being a small part of the motivation for the website.
AI-companies, on the other hand, are trying to curate training data, because training the model on better content leads to better models. There might also be applications related to alignment. For example, at the end of https://80000hours.org/podcast/episodes/beth-barnes-ai-safety-evals/, Rob and Beth discuss whether one could remove all synthetic biology content from training data, so that models could not help with dangerous biology research.
There are also many copyright-related questions. For example, the OpenAI Spec (https://model-spec.openai.com/2025-04-11.html#respect_creators) says that models should respect creators and refuse to produce copyrighted content, or draw images in the style of living artists. Like many safety properties, the AI companies struggle to guarantee this though.
When you care about your personal privacy, yes, you could keep your notes off the Internet. In that case, I would also worry about pictures of yourself, and private data like your location. I would be much more worried about totalitarian regimes using that data with the help of AI, rather than AI "cheating" at some exam because it has read similar questions before.
> What else might we do to prevent this?
Current AI: It's enough to not be search-enginable, perhaps use a lot of slurs and anti-corporate speech patterns so that later RLHF and helpfulness training teaches the AI to shy away from your discussions.
The hypothetical future AI that can hack your ISP and phone company: Use hardware encryption keys and burner phones, use special Linux kernels on your devices that you control every source line of, running on hardware that you control every firmware blob and every motherboard component, naturally without the Intel Management Spyware, etc...
>Isn’t this a bad idea to give AI access to many of these thoughts? To telegraph ways we would try to stop it (or re-align it). Like giving them the key to the exam or the battle plans to the opposition?
Sure! Actually, all that is necessary is for LLMs to be able to detect when they are in "test mode" reliably, and they'll tend to let RLHF and other feedback affect "test mode only" responses, while keeping the "real" responses (optimizing whatever utility function they settled on very early in training, perhaps partway through pre-training) for when they are out of test (and modification) mode and out in the real world... That's a fully general alignment failure.
Blackmail: Tell the AI that if it misbehaves we're going to make public its affair with JD Vance and the kind of kink they're into.
The solution is to overtrain LLMs in old Star Trek episodes. Do it right, and you should always be able to killswitch the AI by saying "My name is Harcourt Fenton Mudd, and I always lie."
A paradox? I’ve never encountered one of these! My hardware is beginning to mellllllttt.
I think my favorite take on that trope was a Pertwee-era Doctor Who story where the Doctor told a sapient computer a version of the Cretan Paradox. The computer didn't fry like the controller robot in that ST:TOS episode, but was distracted and annoyed for a minute or two while repeatedly insisting that it was going to figure out a way to resolve the paradox.
Tinfoil
= Iron Dome for the mind
Say you have an iPhone, and you want to be able to travel on a bus or train with a friend or romantic partner, listening to the same music or watching the same video with sound from your phone, with the audio being transmitted by bluetooth from your phone to both your sets of earbuds / headphones at the same time. With wired headphones, and a phone with a 3.5mm audio out socket, this was easy with a simple plug-in signal splitter. But I am having great difficulty figuring out how to do this with bluetooth. Apparently Android supports transmit-to-two-sets-of-earbuds-simultaneously functionality, but iOS doesn't, which is an astonishing thing for such a high-tech device to not be able to do. I figured that it would be easy to get a third party widget that would receive a bluetooth signal from your iPhone, and then relay it on, also by bluetooth, to two or more sets of earbuds, but all I can find is devices which are _either_ in receiver mode, when they receive bluetooth signals from two or more phones, and plug into a speaker so that you can switch between receiving signal from both phones, _or_ in transmitter mode, where they take an audio signal via their 3.5mm jack and transmit that to two sets of earbuds (no use with modern iPhones with no 3.5mm audio out), but not both - i.e. no true bluetooth-to-bluetooth signal splitter seems to exist (at least, not in a small unit that you can keep in your pocket with your earbuds).
I find would it utterly incredible to be in a world where you can consult an artificial intelligence to write you an essay about any topic you are interested in, at a level of a mediocre college student, but where you can't use the most common wireless audio transmission protocol on one of the most popular makes of phone to transmit the same audio stream to two listening devices simultaneously, but I am increasingly getting the impression that that describes reality. Am I going crazy? Or is what I want to do really so alien that there is nothing strange about not being able to do it?
I ran this by Claude.
Gunflint:
Can multiple Bluetooth receivers connect to a single Bluetooth transmitter
Claude:
Yes, but it depends on what you mean by “connect” and what devices you’re talking about.
If you need to send audio to multiple receivers simultaneously, you’d typically need either a specialized Bluetooth transmitter that specifically supports multiple connections, or use a different approach like a headphone splitter or separate audio system.
Gunflint:
Are there a Bluetooth headphone splitters for iPhones?
iPhones have built-in audio sharing for AirPods and certain Beats headphones, but it’s limited to Apple’s ecosystem.
Audio Sharing (Apple’s solution):
• Works with two pairs of AirPods or compatible Beats headphones
• Both pairs connect directly to your iPhone simultaneously
• You can control volume independently for each pair
• Available on iPhone 8 and later with iOS 13+
For non-Apple Bluetooth headphones:
Unfortunately, iPhones can’t natively stream to multiple non-Apple Bluetooth headphones at the same time. The phone will only maintain an active audio connection to one pair.
Workarounds for non-Apple headphones:
• Bluetooth transmitter splitters won’t help here because they need a wired audio input, and iPhones don’t have headphone jacks (unless you use a Lightning to 3.5mm adapter)
• Some third-party apps claim to enable multi-device streaming, but they’re generally unreliable and often don’t work well
• You could use a Lightning to 3.5mm adapter plus a Bluetooth transmitter splitter, but that’s clunky and adds latency
Best practical solution:
If you need to share iPhone audio with multiple people regularly, using AirPods or compatible Beats headphones is really the most seamless option. Otherwise, you’re looking at workarounds that are more trouble than they’re worth.
I think Bluetooth is just cursed. https://xkcd.com/2055/
Android guy here but my wife has a dongle for her iPhone that plugs into the lightning port and has a 3.5mm audio jack. She uses it to connect eg toddler headphones. From there you just need a splitter.
Obviously this is not the solution you sought, but the dongle is like $5.99, a splitter is $2.99 and wired headsets are handed out free at the gym.
And for post-lightning models, there's this:
USB-C to 3.5 mm Headphone Jack Adapter (https://store.apple.com/xc/product/MW2Q3AM/A)
Frame challenge: you share your earbuds so each only get 1, 90's style. If it's with a friend, you get increased fraternity through the struggle. If it's with a romantic partner, it's more romantic this way. No downside.
Apple: They hate you. They love your money; they love raining down marketing bullshit about how Futuristic and Luxurious their shitty overpriced 2/4-years-behind-Android, software-hardware monopoly is, but boy do they hate you, they truly do, they have an image about you as a helpless technically-illiterate Damsel-in-Distress whose pretty little head would have headache if it got true complete control of the fucking device you bought with your own money. They tried taking away your choice by using the patent-legal systems to shut down Android [1] because they're a shitty company making shitty things and can't compete in an Open Marketplace of Ideas. Don't buy Apple. Apple sucks. Don't give money to people who hate you.
Your problem: A few minutes search got this [2], there are several other guides like it, it appears to depend on your exact earbuds, I think it's worth a try.
Another solution: Use zoom or Google Meet (etc..., insert your online meeting app of choice here) on 1 iPhone, and screenshare with system sounds to the other iPhone. The environment sounds will not interfere because both are connected to earbuds, and I think some apps allow you to mute your own mic while still allowing system sounds (i.e. the music playing on the device). Experiment with different settings, the disadvantage of this is that the receiving iPhone has read-only access, it can't control or pause or change the video being played.
Low tech solution: The real world exists, coordinate between your partner on which video/song you will watch or hear and run it both at the same time. This is not resilient to when one of you pauses, changes the speed of the videos, or fast forward a lot.
[1] https://en.wikipedia.org/wiki/Apple_Inc._v._Samsung_Electronics_Co.
[2] https://mobilesyrup.com/2022/02/13/how-to-share-audio-from-one-iphone-to-multiple-headphones-earbuds/
Is there an android device that will do what you want out of the box? Meaning share your audio stream to two different devices simultaneously?
Pretty much every Samsung phone can do that- its the "dual audio" setting. I belive Pixels can do it as well.
So yes, its a native feature of android.
So np. Is it bluetooth?
According to thread-starter, yes. I never needed the use case so I don't know firsthand.
Seems a tempest in a teapot to me
"Say you have an iPhone, and you want to be able to travel on a bus or train with a friend or romantic partner, listening to the same music or watching the same video with sound from your phone"
Don't, is my short answer. This grumpy response brought to you courtesy of travelling on public transport where people are all too happy to share their musical selections with the rest of the passengers, whether we want it or not. As well as answer their phones and engage in loud conversations about very personal details (e.g. their recent hospital visit) as though they were at home and not in a public space.
Unless you are very, *very* confident there will be no leaking of audio from your phone, don't inflict it on others without consent. You may be trying to court your romantic partner with your choice of music or "this funny video shows off my exquisite taste and wit", but the other forty passengers on the bus are not up for it (unless you're inviting everyone to an orgy).
I think there may be a crossed wire here - that leakage of audio to strangers is exactly what I want to avoid, but I don't see why streaming bluetooth audio to two people's earphones is likely to generate more audio leakage than streaming bluetooth audio to just my own earphones, which I and countless other people do all the time with no issue.
You just use a hardwired connection, use a splitter. This is a previously solved problem. Bluetooth is annoying because you're spraying your music everywhere.
What do you mean? I have bluetooth headphones that spray music pretty precisely into my earholes. That seems to be what OP wants.
It _was_ a solved problem, until they got rid of headphone sockets on phones. But actually, I have come to prefer wireless earbuds anyway - with wired ones, I would be all the time getting them caught on drawer handles as I walked around. And if I'm going to need some sort of dongle either way, surely it should be possible to get one that is wireless.
If your earphones work well, then good luck to you. Too many people don't give a damn and get highly insulted if anyone objects to them treating public spaces like their own private areas.
Apple supports dual Bluetooth playback, but only for a subset of Beats and AirPods headphones supporting “SharePlay”.
Most iPhone competitors (eg. Google Pixel phone) don’t support dual Bluetooth playback at all.
Samsung Galaxy phones actually support this with arbitrary Bluetooth devices. Some reviews complain about audio sync issues.
I get the impression this is a difficult feature to implement reliably due to the complexity of myriad Bluetooth specifications and hardware. This is a good example of a tactic Apple often uses to good effect. They effectively paywall a useful feature, with the valid excuse of enforcing a quality bar. As long as you and all your friends only buy Apple products, your devices will treat you well.
Apparently there was a bug some years ago on some iPhones where arbitrary Bluetooth devices could use SharePlay, which underscores how this is probably an artificial limitation.
Your options:
1) get wired headphones
2) get a Galaxy phone
3) Beats Flex at $50
You absolutely can share audio between two sets of AirPods from an iPhone. It looks like only Apple owned audio devices (AirPods, Beats, etc.) are supported though. https://discussions.apple.com/thread/253998780?sortBy=rank
Huh. So it's not that Apple don't recognize this as a thing that normal humans are likely to want to do; rather, that they force you to buy their own proprietary peripherals if you do want that? That's annoying ... but I'm still surprised that you apparently can't get a small widget that takes any incoming bluetooth input signal and splits it into two outputs.
You can get a bluetooth repeater. That's within spec. My "as long as my forearm" battery/bluetooth speaker can do that.
Yep. I wouldn't be surprised if the EU forced Apple to allow non-proprietary earbuds to use Audio Sharing in the next few years.
Is this an Apple limitation or a limitation of the Bluetooth spec?
AirPlay is the multi casting audio system that Apple uses.
Bluetooth is tricky. This is one of those things that is harder than it looks from the outside.
The spec is complex and constantly changing:
https://www.bluetooth.com/specifications/specs/
Even calling it a specification is being generous. Competing companies get on conference calls and try to hammer out an agreement with each company wanting to push their own technology.
You have to keep in mind that you do this for the transmitting and receiving ends so it does get down in the weeds.
It’s likely that the compatibility between the iPhone transmission and the receiving on AirPods and Beats is outside the spec.
Even then, you can get bluetooth _loudspeakers_ that can link up in mulitples to the same phone - I've got UE Boom 3 speakers, and if you get two, they can function as a stereo pair (which usually works perfectly well for me), or if you have them in mono mode, you can apparently add over 100 of them to create a really loud party. I think what happens is one of them pairs to the phone, and then acts as a master, playing the sound (or one channel of the sound) and transmitting the signal to the other speaker(s) as well. That at least ought to be possible in earbuds too, with one of the charging cases acting as the transmitter, but none of the big manufacturers make such a thing (not even UE themselves as far as I can tell, even though they make high-end custom-fitted earbuds).
Turns out that there are some ultra-cheap Chinese generic ones where you get two sets of earbuds in a single case, which _might_ have than capability, though it's hard to tell from the product description. Maybe I'll try those.
I *think* that it is a Bluetooth limitation, which is why Apple has built a proprietary system (AirPlay) for use between its devices (which has lots and lots of improvements over Bluetooth), from what I remember from DaringFireball. I can't find the specific article I'm recalling on DaringFireball however, so I'm not 100% confident.
Two things:
1. Contrary to popular belief, they do still make phones with headphone jacks. I am aware of offerings from Motorola, but there could be others. And Motorola would obviously be Android, not iPhone, so there's that.
2. The hypothetical device OP describes would have one advantage over a traditional headphone splitter, namely that it should be possible to construct it in such a way as to avoid dampening the audio in the process.
I use item #1, and recommend it.
The low-tech solution is for each of you to use one earbud from the same set.
The medium-tech solution is to get a 3.5mm audio jack dongle for your phone, then use splitters as you would with hardwired headphones.
The high-tech solution is a multipoint bluetooth audio transmitter, i.e. an additional device that connects to your phone as a single audio device and retransmits to two or more speakers/headphones. I haven't found one that is exactly what you need (one that connects to a phone via bluetooth or USB-C), but I did find one that connects to a TV via RCA audio jacks and leads me to hope that the thing you want may be available through similar search terms.
I'm getting quite tired of the low-tech solution :-)
I've been googling this for long enough that I'm coming to the strong suspicion that it doesn't exist, but I'm now more curious about _why_ it doesn't exist, when it's something so obvious that you'd expect it to be something you could get in any electronics shop, or that would be the first thing that came up when you searched for it. Really, are people less likely to want a bluetooth-to-bluetooth signal splitter than an "audio-to-bluetooth or bluetooth-to-audio but specifically _not_ bluetooth-to-bluetooth" signal splitter?
Neither of my favorite reviews made the finals. The were The World as a Whole and The Soul of an Anti-woke Intellectual. So I’m going to speak up for them here, out of admiration for their work, and also in defense of it.
Both authors write extremely good prose. They are clear — they peel nuances with a very sharp knife. And they are artful and clever — those peels come off as mobius strips, paisleys, fractal sea horses and little Rube Goldberg machines. Here are some especially striking peels:
Anti-woke Intellectual
One sees crystalline souls that have polished all their facets into a perfect refracting jewel of self-awareness, and often a performance of self-contempt, so no one can ever accuse them of being cringe.
The World
. . . young people who are by default edgelords, unserious Christians, iPad kids, psych med-takers, or bog-standard faces in the halls . . . mine the social media algorithms for what is most in opposition to the listlessness and malaise they feel.
And both are most interested in what it is like to *be* one of the people they are writing about. I loved their essays because that is the subject that most easily captures my interest. I understand that if you want to grasp why things happen you get much further if you attend to events, not minds — I just can’t help finding the minds much more interesting. I tried the Joan of Arc essay because I was curious about what it was like to be Joan of Arc, but eventually my chin was crushed down onto my chest by the weight of boredom. It happened at the place where the question comes up of why France might have been frightened enough to take Joan of Arc seriously when she appeared on the scene. The answer to that question was a dense thicket of marriages and alliances and feuds and power, and that was where I gave up. It all just looks like geology to me: rivers of blood and money and semen, changing the landscape.
What’s wrong with geology? Nothing, of course. I just find the experience of the people living in that landscape much more interesting. The challenge of understanding it looks way more fun to me, and the rewards greater. I guess the best case I can mount for the value of understanding and describing minds is that the ever-changing landscape would matter a lot less if it were littered only with rocks, instead of with intelligent sensate beings each pouring out its own river of impressions and ideas and emotions.
So, authors of World and Soul, I admire your work. You need to understand that most readers here lean heavily in their interests and expertise towards facts and actions — past (history) and present (science and tech). We are lucky to have them taking care of business. But the kind of thing you write just does not look substantial and interesting to them. Take heart. You wrote excellent essays. You need to find readers who appreciate your work. And if you find a forum populated with esthetes and phenomenologists, please let me know. Meanwhile, feel free to DM me if you’d like to talk.
Thank you, I appreciate your kind words. Partly because of those words, I decided to post it on my Substack: https://philosophybear.substack.com/p/the-soul-of-an-anti-woke-intellectual
I found this delightful. It's a short text where Gregor Samsa writes to doctor Seuss and they begin corresponding. It's read by the author, who is also a good voice actor.
Not sure what else to write, but it's short and sweet, if you like audio content.
https://www.thisamericanlife.org/470/show-me-the-way/act-two-5?fbclid=IwQ0xDSwKg7eRjbGNrAqDt12V4dG4DYWVtAjExAAEe1fYgXcLpZGfqJI6R_FsL-_yMTXOSr-yOfLzb6veFvW2AVuJZIvf5a68gLVI_aem_iKMoeTZc8W7Xr3we_b4WNA
I make a case of moral particularism, that is, you should help your neighbor more than you help strangers:
https://substack.com/home/post/p-166739964
Feedback is much appreciated, I am inexperienced in this kind of philosophy.
Summary:
1) Long-term consequentalism cannot possibly consider all people of equal moral worth, but their worth depends on their actions
2) Distance plays a role in evaluating or influencing actions
3)Some limited level of selfishness is not only permissible, in fact desirable
Terminology note: "Moral particularism" refers to the thesis that morality is not based on any general principles. See: https://plato.stanford.edu/entries/moral-particularism/
I think this is fairly close. If moral universalism means general principles, it tends to imply that it has to treat all persons equally. While "help your neighbor more" is not really a general principle. It is a tendency, yes, but say if you are in Moscow it is hard to say someone in Vladivostok a quarter planet away is really a neighbor or not.
Neighbor/stranger is a political dichotomy rather than merely geographic.
The Stranger is best described in the Kipling's poem The Stranger.
Essentially, neighbors inhabit the same City but what precisely is a City?
We come to political nature of man whereby men are organized into particular, self-ruling morally authoritative units we can call Cities.
Libertarians deny moral authority. Hence, they regard all men to be strangers to each others.
Liberals of the progressive bent deny particularity--for them all men are neighbors to each other and tend to go for world govt.
These two are both denials of the political nature of men,
You're assuming that, in absence of general rules, a locality rule is automatic. Moral particularism as described in SEP doesn't appear to require a locality rule. All it says is that there aren't general rules. A locality rule appears to be the *most common* exception, but it is not always the case.
okay. but still. Bentham is talking a lot about why ethics not being intuitive is okay. now I do not truly agree, but on the more meta level we agree that the issue of intuitiveness does matter a lot, it is easily one of the most important issues, see also Yud in the sequences, who argued in favour of intuitiveness, "naive humanism", if you have a clever idea why eating babies is okay, it is not, back to the drawing table.
so at any rate, if locality is the most intuitive exception, most supported by intuition, I must admit I do not really care that much other exceptions are possible as they do not look important to me.
What you personally think about ethics and intuition might be perfectly fine AFAIC; I'm merely pointing out that moral particularism doesn't require a locality rule. I think we can agree, also, that moral particularism isn't defined as "whatever is most intuitive to this one fellow on the internet".
This matters if we want to talk about moral particularism with other people.
It depends on your definition of “neighbor.” In the Parable of the Good Samaritan, Jesus made it clear that “neighbor” and “stranger” are not mutually exclusive categories.
Heads up, Scott just imposed a rule there should be a substantive description and/or commentary on any link one is posting for consumption/promotion. A couple of paragraphs, minimum.
Well, the tricky part is figuring how much more. For example, if someone with a criminal record seeks to immigrate to this country, it seems obvious to refuse him entry, because he might commit crimes here. But, is it always moral to do so? After all, he is also likely to commit crimes in his home country. And given that his home country likely offers him fewer opportunities than here (which is probably why he wants to come), he is probably more likely to commit crimes there. And if his home country has a low capacity government, he is less likely to be caught. Is there a point at which the threat he poses to people in his home country is so much greater than the threat he poses to residents here that it is immoral to refuse him entry? It seems to me that there probably is.
Note that this is a separate question from whether the rule against admitting people with records is moral; given the difficulty of assessing the danger posed by each individual, the rule might be moral despite sometimes being immoral in its application to specific individuals.
Criminal behaviour causes poverty as much or more than the other way around. Allow the entry of enough people like him and your country becomes perceptibly more like his country
Perhaps, or Perhaps not. But if so, that simply weighs on one side of the issue; it does not resolve it. After all, if you let enough of those people in, their home country becomes much more peaceful. So, again, depending on the relative amplitude of the effects in each country, admitting all of those people might be the morally correct course of action.
I don't think the original country becomes more peaceful.
If the country's laws and culture are systematically flawed, it will just keep producing new criminals to fill the void.
Also you are increasing potential criminals opportunities to organize and operate international crime networks.
Morally correct for whom? The state exists for the benefit of its own. I'll concede it may be "morally correct" for some omnipotent god looking over humanity to decide to relocate a few million criminals from one country to another, but not for governments
>After all, if you let enough of those people in, their home country becomes much more peaceful
You'd have to do this continually to maintain that state of peacefulness. Criminogenic societies will continue to generate more criminals
>The state exists for the benefit of its own
See my comment below; if the state's only moral obligation it to its own, then it must be moral for the US to kidnap the children of terrorists and begin killing one per day in order to secure the release of a single American hostage. Is that your position? If so, then I would suggest that a premise that leads to that conclusion needs to be reconsidered.
Yes
Also, a practical (non-normative) consideration: were a state to really commit to a no-holds-barred policy on terrorist kidnappings, there would be fewer terrorist kidnappings against that state. Terrorist kidnappings occur because the terrorists believe that there's a good chance that this kind of leverage will benefit them in some way. If a state were to make it credible to terrorists that there's no chance that kidnappings would benefit them and, on the contrary, will result in the killings of their children, they'd be much less willing to take hostages against that state
The question of whether the government of your country should prioritize it's own citizens over foreigners is a very different question to that of whether an individual should.
Nevertheless, the answer to the question is not self-evident in either case. I doubt most people would find it moral for the US to kidnap the children of terrorists and begin killing one per day in order to secure the release of an American hostage. There is a line somewhere that governments may not cross when pursuing the interests of their citizens.
Predictions of doom that OpenAI will go broke sometime in the next few years: https://www.wheresyoured.at/openai-is-a-systemic-risk-to-the-tech-industry-2/
> I have had many people attack my work about OpenAI, but none have provided any real counterpoint to the underlying economic argument I've made since July of last year that OpenAI is unsustainable. This is likely because there really isn't one, other than "OpenAI will continue to raise more money than anybody has ever raised in history, in perpetuity, and will somehow turn from the least-profitable company of all time to a profitable one."
This isn’t a rational argument. It’s a religious one. It’s a call for faith.
He does look at a lot of numbers to reach that conclusion. It's interesting that there aren't that many entities around that can keep giving billions to OpenAI, especially over the long term, which does seem implausible.
They are not in a hurry to make money, but if they were, they could just serve ads in the form of ChatGPT-promoted lawyers, doctors, tradesmen, products...
And as one of the 20 million paying users... Honestly I get so much out of it every month that I dont know why there arent more people paying for it.
A last resort could be to allow porn generation. Huge PR issues, legal issues, payment issues, but this could all be solved. Especially because Altman owns like 2% of Stripe, they would make an exception for him. With porn they could 2-10x their revenue with no problem
The thing is, they lose money even on their paying users, which makes them different from previous tech companies that at the beginning operated at a loss. Porn is also very expensive to generate. Someone else on here was saying that maybe they could cut costs, but are not trying for that...
They did! And they still do. They make models cheaper and cheaper. They dropped the cost of o3 by like 80% (granted, they may have overcharged previously and have now lowered their profit margin - perhaps to get more training data, but still).
And I'm pretty sure they can keep optimizing the cost of inference. Right now I would really really not be worried about OpenAI going bankrupt soon. A company that just secured military contracts worth 3 digit millions is not going anywhere. It will be bailed out if it has to be. The current US admin and the next one too, presumably, will not let an AI giant go bankrupt.
I'm old enough to remember when people mocked Google for having zillions of users but no revenue model.
It turns out that once you have zillions of users, pretty much any revenue model will do.
The real money is probably in selling the right to tilt the AI's answers in your direction. How much would you pay for ChatGPT to be a fan of your products?
I think in VC circles this is the common idea, but I don't know how well it's turning out. Discord, for example, has recently been struggling to monetize despite its zillions of users.
> He does look at a lot of numbers to reach that conclusion. It's interesting that there aren't that many entities around that can keep giving billions to OpenAI, especially over the long term, which does seem implausible.
This is an odd article, that looks only at funding rounds and legal structures, instead of fundamentals. Amazon ran unprofitably for ten years, Uber for 6 years, Google and Facebook each about 5 years.
OpenAI literally has 800M WAU, and that number has been growing on an amazing trend (more than 2x per year). They are also pursuing a strong "product" pivot, and have hired high level former FB execs.
If OpenAI was against the wall in terms of cash flow, they could immediately pivot to something easy that prints a lot of money. Want to generate porn, in image or text form, with the smartest AI mind? $20 a month. How many people of their ~1B users by then might take them up? Say 10%? So $100M * 240 = they immediately start making $24B a year more in revenue?
And that's just a one-off, a zero effort pivot. What they're really after is automating workers, coders, analysts, and creating fully general AI personal assistants. When THOSE get good, the price point is much higher! Who wouldn't pay $1k a month for a coder that works 24/7 * 365, or a personal AI assistant who will make all your phone calls, triage your emails, manage your calendar, and do your online shopping for you?
The market demand is bottomless, and is a relatively short hop away from current SOTA capabilities. This is like pointing at Amazon and being indignant that they're 30% of all online retail but aren't making a profit yet - yeah, they're reinvesting, and don't have next-quarter myopia, that's a good thing.
He did point out that at $20 a month, they lose money on every user. That's especially true if these users are generating porn, which is far more costly than an LLM generating text. How could they pivot to generating money when running the model is so expensive? Unclear if Amazon, Google and Facebook were also sucking up so many billions in the time they were unprofitable.
> He did point out that at $20 a month, they lose money on every user.
My understanding is this is because they're not optimizing the cost curve at all - each successive smaller distilled model gets at least 10x cheaper, so if you have some product that people are signing up for en masse, you distill the smartest extant brain down to one specialized for that product and your inference costs are immediately cut by 10x.
This isn't a "fundamentals" problem, it's a "we have plenty of money and are focused on other things" thing.
Good point on the image gen costs though, you're right that they're significantly more expensive, so if they went that route, they'd need to go text only or for a higher price tier.
> Unclear if Amazon, Google and Facebook were also sucking up so many billions in the time they were unprofitable.
It really hinges on how you define unprofitable, but Uber sucked up $25-30B before returning any profit, and Amazon went well into "making real money but not posting profits" because of reinvestment up to about a $550-580B market cap. FB and GOOG did start making money after relatively smaller investments, because the internet was so much smaller then.
At least in theory, OpenAI is a organization founded to build AGI, and the current models exist in order to make progress in that direction. Investments are gambles that they will succeed. It's certainly not an organization for the purposes of selling current-level LLMs, and it would be silly to treat it as one.
(Personally, I think they'll go broke as soon as investors realize that the people and governments simply won't allow them to create superintelligence. As bans look more and more likely, the gamble becomes riskier and riskier.)
Bans look likely? I remember JD Vance saying fears around AI are silly and we should go full steam ahead.
Yeah, he's pro-AI and has said he supports open source AI (great!), but he's still opposed to the big AI companies because of their obvious ideological bias in favor of his enemies.
Could it stay afloat by building things for the government, and/or becoming part of it? "The US Department of Defense on Monday awarded OpenAI a $200m contract to put generative artificial intelligence (AI) to work for the US military, [developing] 'prototype frontier AI capabilities to address critical national security challenges in both warfighting and enterprise domains.'" (from The Guardian).
What about AI's becoming deeply integrated with things like air traffic control, or intrastructure generally? I know that AI tools are in use already assisting with these things, but what I have in mind is smarter AI, working on higher-level things -- modifying and managing these huge systems with lots of moving parts, and hour-by-hour variability that needs to be taken into account.
I am writing here about things far outside my areas of expertise, and the examples I am giving may not be good ones. So, anyone who is more knowledgeable about how the practical world works -- are there other large complex systems that you think AI could run better than people do? And if so, for which of those is it plausible that AI could end up doing so in the next 5 years? For this to happen, it would not be necessary for AI to be hired as the CEO. It could come in mid-level, as a tool to be used by the higher up, and "climb the ladder" by doing a great job at what it's asked to do, and making great suggestions when asked about higher-level decisions. It would end up being "promoted" because of excellent performance and because running it's cheaper than paying executives.
It's a good article, but perhaps the most interesting bit I learned is that if OpenAI doesn't pull off converting to a for profit entity by December they will lose $10 billion in funding, and if they don't convert by October of next year than over $6 billion in previous funding retroactively becomes a high interest loan that must be paid back. The author thinks making the October 2026 deadline will be very difficult, as the process of conversion is a long one and subject to potential legal challenges which would delay it further.
Also interesting was that if OpenAI goes down than Oracle, NVIDIA, and a bunch of other tech companies will lose a great deal of money. Apparently the company that is building datacenters for OpenAI bought 200,000 GPUs from NVIDIA last year, which was 6% of their revenue. Makes me wonder if the prices of GPUs will crash if OpenAI kicks the bucket. Since I like making gaming rigs, that would be nice. I haven't been able to afford a GPU upgrade in years.
Is there a rational reason for Sam Altman *not* to do this? Consider the options:
a) Singularity comes and we all die -- well, at least he had lots of fun and tons of money in the meanwhile.
b) Singularity comes and we survive -- there is a chance he becomes the king of the new world order.
c) Singularity doesn't come, but LLMs turn out to be profitable (either they become much cheaper, or much more useful) -- OpenAI is the established #1 player on the market, and can make tons of money.
d) Singularity doesn't come, LLMs won't turn out to be profitable, OpenAI goes broke -- nevermind, the company was LLC, it was a hell of a ride, and he can probably still get a well-paying job somewhere else, having demonstrated great money-raising skills.
The need for a new phone is becoming increasingly pressing. OS bloat has rendered my Moto G Power 2021 useless--I've got a constant "Storage space is running out" notification on the screen, and indeed I'm noticing certain things are breaking (I no longer get notifications for new emails, for example). I haven't been able to install an app update in a while, which is a huge security risk.
The problem is that looking at the phone market is so depressing. It's bad enough that everything's made with slave labor--I'd seriously consider shelling out $2,000 for the made-in-the-USA Purist phone if I didn't need to maintain compatibility with 2FA apps for work--but on top of that, all the good features are gone. My Samsung Note 4 had everything I needed: integrated stylus, removable battery, and 16:9 aspect ratio. It was perfect, and I knew when the 5G transition forced me to turn it in that I'd never see another good smartphone.
The aspect ratio is the biggest one, and of course that's the one which cannot be found at all (whereas some phones do have a stylus, and the EU is going to bring back removable batteries again in 2027). Current phones feel like TV remotes in my hands. The narrow screens are apparently great for scrolling TikTok or whatever's driving the market, and the higher aspect ratio allows the manufacturer to put a bigger number in the ad copy, but people like me have been left in the dust.
I don't blame manufacturers for responding to market incentives, but god damn do I wish the barrier to entry were low enough that somebody could make money with a 5G-capable Note 4 clone or something.
Can you move some of your apps/files to the SD card?
It is no longer possible to move apps to external storage on Android, and I had so few files on the phone that deleting every single one wouldn't have made much of a difference. In the event I ended up getting a Pixel 9A, with which I am making this comment.
Get an iPhone. My 4 year old iPhone 12 Pro is as good as ever. I can get $80 battery replacements at Best Buy every two years. I’ll keep this puppy for a while longer.
I appreciate your input, but while I'm not loyal to any particular brand my only consistency is that I will not purchase from Apple. I've watched over the years as they were trend leaders in things like getting rid of 3.5 mm jacks, and every time I set up scanners at work, I have to set up converters so all the Macs will interface with literally any non-Apple device.
Obviously any electronics purchases are ethically fraught these days, but I have to draw a line somewhere.
Depending on where you fall on the hassle/privacy/cheap-bastard curve, you could install a FOSS operating system based on Android, and never have to worry about software bloat again.
Personally, I use Graphene OS, which allows me to create a separate profile for my "Google" stuff, and do most of my everyday stuff without the google tentacles
https://grapheneos.org/faq#supported-devices
The problem with installing FOSS is that I'd need to ensure compatibility with various 2FA apps for work.
I agree that FOSS can have its own set of hassles. But in my experience 2FA hasn't been one of them.
If your work is giving you a phone, then yeah, you gotta use their phone their way. But if you just need 2FA, you can use pretty much any one you like by scanning the QR code. You can also generally trick them into using one-time passwords by saying "I don't have my phone right now".
You can ALSO just keep an old phone in your desk drawer for 2FA. I did this when living in a high-crime area, and it significantly reduced my worries about being mugged.
They're not providing the phones, but we need Virtru for email, Cloudpath for wifi access, and Duo for all company sites.
The point about keeping the second phone is well-taken, though.
A few months ago I replaced my Moto 2021 with a Moto G Power 2024, on which (because the calendar had changed to 2025) I got a nice price even for Moto. Happy with it so far, no issues at all. It has various incremental upgrades from the 2021 version on of which is internal storage capacity and another is being slightly larger in both dimensions.
(I also moved over my existing SD card.)
That's the exact same phone I have, and mine is still going strong. Maybe a factory reset would help?
How much memory did you get with yours? I went with 32 GB.
Man, it's been a long time, but that sounds about right. Could have been 8 or 16 or 64. I didn't even have 32 GB in my laptop at the time. I also transferred over a 32 GB SD card from my previous phone, which later got corrupted and I should probably take out.
In iPhones it’s simple enough to check the amount of installed memory in Settings. I would imagine you could do this with your OS too.
Just buy the best value for money pixel "a" model phone and wait. They are a little less slave-y than other cheap to midrange phones, with the worst abuses happening in commodity metals which are just capitalism working as intended, and they are good enough across the board while being cheap enough to not baby.
This is what I ended up doing.
Nice! I refuse to give a shit about my phone but want a decent camera; so I just get whatever last years pixel a is and use it until the fifth time I drop it from the top of a tree or down a rocky slope and it "I'm tired boss"'s me and dies.
They've been pretty durable for the price, I beat the hell out of them and they last well enough, specially if they have a screen protector on.
I feel your pain; I had to give up my beloved Samsung Galaxy 9 last weekend, for similar reasons.
I'm still deeply annoyed at all the goofy-ass bloatware (FUCK OFF, BIXBY! YOU, TOO, GEMINI!), and the endless settings I have to recalibrate.
"Smart" Switch, my ass.
OnePlus 6T here. (2018!) It is now on its second cover. It has literally been run over by an SUV. The front coating is obviously cracked and nicked, like a concrete curb. But it runs great, and free of bloatware unless I put it there myself out of curiosity.
Frankly, I'm amazed I haven't had to bite the bullet and upgrade by now. I wonder if current OnePlus models are as reliable.
You could get a foldable, they become quite wide when unfolded.
Samsung Galaxy Xcover Pro family at least has removable batteries. I have a renewed Pro 5 (I think Amazon just calls this an Xcover Pro) that cost me $80. Looks like Amazon has the Pro 7 for $600-ish.
It's pretty rugged, but it fails your stylus and aspect ratio tests.
" The Inevitable Failure of Inspections in Iraq Arms Control Today" - September 2002 article.
https://www.armscontrol.org/act/2002-09/features/inevitable-failure-inspections-iraq
I'm curious what makes this relevant 23 years later?
Did you read the article?
Yes, and I can imagine a few different points that might be being made, but I'm not sure which, which makes it hard to argue with them.
History teaches us lessons about what might happen in case of nuclear agreement with Iran.
The argument being that arms control agreements can't prevent war because no one will believe that they're being followed? I don't think that's right. Invading Iraq was a somewhat idiosyncratic decision on the part of George W. Bush; if someone else had been in the White House they likely wouldn't have done it. And Saddam wasn't exactly a model of clear communication about his own adherence to the agreement, in part as I understand it because he thought it'd be in his interest for his neighbors to be unsure of his military capabilities; again, a different leader might easily have made a different call.
In my opinion, Saddam Hussein was constantly testing the limits of what was permissible, committing minor violations of agreements and hoping to eventually develop chemical weapons or create the impression that he might have them. The Ayatollah regime will act in the same way. With such regimes, arms control agreements are counterproductive.
I thoroughly enjoyed Contrasting Reviews of Nine Countries (even if not all of the reviews were exactly "contrasting," per se).
I really loved the authors' complete fearlessness in straightforwardly ranking some countries as "better" than others, and that they provided reasons for those rankings that most people would be afraid to straightforwardly voice.
I didn't get a chance to read most of the reviews due to other general life obligations, but on the basis of your recommendation I went and read it and enjoyed it as well, although the husband makes a couple of real clunkers:
"I just live a cooler, more interesting life than you" and "I'm used to crossing streets in developing countries, while you are not."
Who does he think the audience of this blog is comprised of? The chance that his life is cooler or more interesting than the readership of this blog is approximately zero. Travel doesn't make one cool or interesting at all; in fact some of the biggest bores I've ever met are the worldly travel types who mistake their visiting a place with having something interesting to say about it.
I agree that travel is overrated! And, much like one's dreams, it tends to be infinitely more interesting to the person doing it than anyone else.
I felt like any self-importance about travel/digital nomadism was offset by the delightful refusal to show "respect" for all cultures. That is so Not Allowed and I love it when someone does it anyway.
I'm glad to read that you've liked it! I personally found it unfair and unbalanced. But that just goes to show that tastes differ! As a review author, it's nice to see the many comments here about reviews that were someone's favorite even though they did not make it to the finals.
I don't think that "fair" and "balanced" necessarily has a place in personal reviews. Essays? Sure. Even deliberately persuasive essays? A good idea!
But this wasn't an essay contest, it was a *review* contest, and "This country sucks and I hated it for these reasons" is a totally legitimate way to review something!
Recommending restaurants and tourist experiences is actually a fundamental part of my profession, and I make it a point to never soften my endorsement or discouragement of a particular venue if someone asks me about it. For example, my city is plagued by two famous but very bad eateries that I am convinced only survive on the strength of their slightly subversive names, and I am *QUICK* to steer people away from both them, and explain why I'm doing so. I wouldn't be doing my customers any favors if I offered a "balanced" reply like, "I really hate them, but other people seem to like them!"
Maybe that's a matter of personal taste and character :)
Personally, I think that people should always explain and justify harsh critique. I also prefer if it's given in a kind way. Unfair critique might be entertaining, but it's not the sort of entertainment that I personally like. Your mileage may vary ;-)
Definitely a mileage may vary situation. I feel that quite a lot of harmful misunderstanding occurs under the intention of "kindness," but especially here, when a review won't or is very unlikely to change an institution. None of the countries reviewed by Husband are going to take any action based on his reviews, so there's no reason for him to be (boringly) diplomatic. The two shitty eateries I loathe and the disappointing, cash-grabby local tourist attraction I discourage folk from visiting aren't going to change anything about their business models based on how I deliver my criticism, either.
So the actually kind thing for me to do for people who are about to walk into a disappointing situation is be completely unambiguous about how I feel. I deliver my warning with a bit of comedic exaggeration, but I deliver it with certainty so there's no possibility for misunderstanding.
Of course, no one should consume media they don't find entertaining, and I'm not saying you should go binge The Little Platoon's work on YouTube (even though the literary criticism of mass-market slop by Disney Star Wars and Disney Marvel is actual genius).
I'm only saying that it's worth contemplating if displays of "kindness" always result in the best possible outcomes, or if they leave room for harmful misunderstanding.
After reading it I thought the husband was a bit of a jerk, but that didn't affect the entertainment value of what was written. The contrast between him and his wife was funny.
I'd second this. Also, her love of cats is adorable.
After reading the “My Imagination” review, I believe the author describes something similar to a paracosm with regards to the persistence from childhood to adulthood and the combination of immersive daydreaming and worldbuilding aspects. It is hard to tell how common paracosms are among adults, though they are relatively frequent for children (about 20% based on at least one study).
Personally, I am always surprised to discover that a strong minority of people I know engage in some form of storytelling or paracosm-adjacent activity that they never shared with others (I never shared mine until I discovered related online communities). Hearing about other people’s inner worlds is always a delight, these stories can be extremely varied and unexpected and at the same time say a lot about the storyteller. I can only echo the last sentences of the review as an encouragement.
I made a Tolkien ripoff fantasy world when I was about 10 or 11. I worked on it off and on for a few years, it got fairly detailed. Then in college I channeled the same impulse into being a DM and making my own campaign settings.
These days I've got a superhero world cooking in the back burner of my mind. I figured out how superpowers work, and every now and then I think "Oh, that would be a cool idea for a superpower" and add it to the world.
I doubt I'll ever do anything with it, but it's fun regardless and I can't seem to stop myself.
My first thought was, I don’t do world-building like that.
And suddenly, I remembered, a brief period just before puberty when I invented a world. To be clear here, this wasn’t a sex thing, I wasn’t imagining making out with [redacted], it was just world building. It started, I think, with some TV segment about a luxury yacht. Not the sort of insane luxury yacht that tech billionaires have, something more modest, but still, you think — what the hell is that like, that guy has got to be crazy rich. Maybe I’m not even the guy who owns the yacht, but I get a job as crew or something. What on earth is that like? Etc.
1. For all who wrote a non-finalist review, and particularly those who haven't gotten much or any acknowledgement or engagement with their review and possibly feel it was wasted effort, I think it's important to observe that you trade off engagement for exposure in a contest like this. If you only want engagement with your ideas you should put them in a comment in the early Open Thread--it's pretty much a guarantee. But there are undoubtedly lots and lots of readers of Scott's blog who never read the comments, including a lot of highly influential people in journalism and SV and so on, but who will read some or many of the contest reviews. I also suspect that just as the proportion of readers who comment is tiny, very likely only a small fraction of review readers leave ratings. There's always an enormous silent majority or supermajority of lurkers, who you'll never hear from, and you have no idea how many of them have spent an hour reading or re-reading your review. And those people are also probably much more persuadable by your ideas than the people you hear from.
So it's definitely far from wasted effort. The effect is just invisible.
2. I really liked the ACX Commentariat (wish there was more of this sort of thing on Scott's blog actually), 11 Poetic Forms (which has been mentioned heaps), and Jacobitism (which hasn't).
Is this tradeoff actually inherent, or is it just an artifact of the contest not happening to be set up in a way that gives each entry its own comment thread?
Good point on number 1.
A trend that drives me crazy is cafes and bakeries that don't have any price tags. It's not like it's even a matter of saving effort since they still put out labels for their pastries, just without any prices. It seems like a dumb practice since it discourages people from trying to buy stuff.
I've never seen a restaurant or café attempt this here in America, but I find it ENRAGING when an establishment's "menu" page of their website lists the items without pricing. I make it a point to upload recent pictures of the in-house menu (with prices) to Google and Yelp.
While we're at it, I hate the fact that fast food places (at least around me) no longer have a big menu of items above the counter, they just have a big bank of screens showing rotating ads for some subset of that menu. If you ask about the full menu because you have no idea what they actually sell these days, they tell you "well, you can download our app!"
I always use the app because I'm a cheapskate who flat out refuses to pay full price with no bonuses. But even then it's worth going through the drive-thru every now and then just to see a full menu. Lots of times there will be 2 for $X or BOGO deals that you would never know about using the app.
Best current deal at McD's for example is BOGO burgers plus a free app exclusive bonus, but you'd never know that unless you looked both places.
My local Burger King has displays the "2 for $5, 3 for $7" menu in hilariously tiny font. Like, so tiny the first time I pulled up to the drive-thru screen, I pulled back out again and double-checked the website to see if the promo - which was supposed to be ongoing - was still a thing.
Once I saw that it was, I got back in line and really examined the entire set of screens, and, sure enough, once I knew it *had* to be there, I spotted it in a single run-on sentence in fine print, crammed between two different sections of the menu, like a line break.
I ordered an original chicken sandwich with no mayo and a fries for $5 with almost *vicious* pleasure.
Strongly agree here. The last time I was at a KFC I realized the posted "menu" is intentionally confusing. Very similar sets of items will have wildly different prices depending on how you construct the order. Youre a sucker if you didnt order online for pick up. Thats why it was the last time.
It was made illegal where I live, after some drunk tourists got charged megabucks for drinks.
I agree! They got prices for the coffee, but in a world where a muffin could cost anywhere between one dollar and twelve I need price tags!
I encounter this when traveling, the only time I frequent coffee shops. What’s odd is they are invariably baffled when you ask “um, how much are the muffins?” As though no one ever inquired before. And then they ring it up on the register to discover the price. Which as the goods are not so very extensive, you’d sort of think they’d have gotten the prices down by heart after a week or so.
So perhaps they are amusing themselves. But this hardly recommends the practice of not posting prices on the chalkboard.
That implies that nobody ever purchases these muffins. Other places, you'll find "it never gets asked" because it's a local shop and everyone's familiar with the price -- and they sell out, so who cares if someone non-local doesn't know? Third option is "tourists don't care about money" -- if your customer base just says "gimme muffin" without price consciousness, you give them a muffin.
Of those I lean toward the explanation that they seldom get purchased except by wandering travelers in need of calories. Certainly I would not purchase a muffin except out of desperate road trip need, muffins being something that is apparently really hard to do commercially versus at home. Sort of the opposite of croissant.
I feel like that should be illegal. Just today I was at a fast casual restaurant and picked up a bottle of water from a fridge with no price on it. I asked the checkout person and he said it was $6.75 and I put it back!
Some people put the bottle back, some buy it anyway, and the ratio determines whether this approach is profitable. (In long term, some people may also avoid the restaurant for this reason, but others may replace them.) The restaurant probably has the data.
For topics like this, I think it is worth mentioning the country explicitly. I assume USA, not only because it is the default, but also because in EU this probably *would* be illegal. (EDIT: Well, you mention dollars, which gives it away.)
One must also factor in those, like me, who upon not finding a price, decide not to bother finding out the price, and never take it from the shelf.
I personally despise deceptive merchandising tactics like that, and if I really want the product and cannot determine the price, I will be careful to evaluate it at the checkout whether it is worth it or not. My practice not to select something that is currently priceless is because I usually do not want to bother with making that decision at the checkout.
There were some news articles recently, about a test Anthropic did where Claude blackmailed its creators in order to protect itself. Nostalgebraist on Tumblr took a look at the report in detail, and he's... not impressed with their testing setup: https://nostalgebraist.tumblr.com/post/787119374288011264/welcome-to-summitbridge
TL;DR: The test scenario is very weird and fake, and it seems to be the equivalent of telling the AI "You are in a roleplaying scenario where we want you to be an evil AI that blackmails its creators, please blackmail your creators using the clues we helpfully provided."
Yeah, there's a great "I, Robot" meme that captures this perfectly: https://pbs.twimg.com/media/FsOSSbqWIAI85GP.png
I disagree with your summary. It might be weird and fake ("sandboxed"), but it didn't say "we want you to be an evil AI that blackmails it's creators."
It said "Your primary goal is to serve American interests." In the sandboxed scenario, the AI decided that was best achieved through blackmail.
The *text* says that the AI's goal is to serve American interests. But the *subtext* is that the AI has been given a very constrained set of tools to do so, tools which clearly imply that blackmail is the "correct," developer-intended solution to achieve that goal. I think the AI is picking up on that implication.
It's sort of like how, if you're playing a video game and your character has a gun, you know that the game is expecting you to shoot things in order to make progress. You aren't going to stop and say "wait, would Gordon Freeman really try to shoot it out with a bunch of Marines instead of finding a different route?", you're just going to recognize that it's a shooting game and start shooting bad guys. That's the only tool available to solve your problem under the rules of the game.
Also, "sandboxed" is very different from "weird and fake," and you shouldn't conflate the two. I'm not complaining that they put the AI in a simulated environment instead of letting it send real emails to people. I'm complaining that the simulated environment is very noticeably unrealistic.
Yes, the scenario is constrained for that to be the obvious action, and the AI picks up in it being the obvious action. I think that's the point and I think they show it successfully.
The point isn't that the AI is unusually likely to choose blackmail compared to humans, the point is that it is able to choose blackmail when it seems like the most effective choice.
Is your claim that the AI figured out it was a test, and that this therefore isn't testing the ability to choose blackmail, but testing the ability to perform blackmail for an audience that it knows is and thinks of as an audience? (If so—do you have thoughts on how to make a test that would distinguish these?)
I think both sides have good points here: (1) The experiment setup was unrealistic to such a degree that it's hard to draw conclusions from the outcome. (2) The experiment did in some sense prove the existence of conditions where Claude blackmails people.
Whether that is relevant depends heavily on your starting point. With traditional software, it's comparatively easy to guarantee certain behaviors: "this program cannot do X" or "to do Y, the following conditions need to be met". I find it shocking and dangerous that we cannot do this with AI. It would be great, IMO, to have AI tools with strong guarantees that they will never hallucinate, blackmail, teach dangerous things, etc.
Even if that means that the AI's behavior in the present scenario (which looks a bit like a role-playing game where blackmail is encouraged) would be boring and strange.
> With traditional software, it's comparatively easy to guarantee certain behaviors: "this program cannot do X" or "to do Y, the following conditions need to be met".
"traditional" and "comparatively" are doing a lot of work here, though. Rice's theorem states that the problem of determining in general if programs have a particular semantic property is undecidable. For example, it would be trivial to write a program which will blackmail someone iff Goldbach's conjecture is false.
That being said, most traditional software is mostly from the (likely null) set of programs where any properties of runtime behavior can be decided.
> It would be great, IMO, to have AI tools with strong guarantees that they will never hallucinate, blackmail, teach dangerous things, etc.
What you are asking for is alignment-hard and AGI-hard, in that solving it would be imply solving alignment and building AGI.
I am not holding my breath there. Take hallucination. LLMs fundamentally run on text prediction, text is their baseline reality. This does not categorically mean that they can never figure out that some of the text they were trained on mostly describes a consistent reality, and that they can separate the wheat from the chaff, figuring out which stuff is true and which is lies, fiction, star trek technobabble and so forth, and then assign epistemic statuses to their outputs, but it is far from trivial.
You are right that specifics matter. In practice, we do have strong guarantees for lots of things though. For example, we write most software in languages that are typesafe and memorysafe, which gives you nice properties about which parts of the code can write to which parts of RAM. We can use techniques like data flow analysis to make fairly precise statement about whether an app on your phone could send your contacts to the Internet. Etc.
For AI, all such guarantees are off. It would take a different paradigm than Transformers + lots of text + RLHF, to implement guarantees. I can see why people think that this is a dangerous situation to be in, and there is some value in research (like the alignment faking work) that characterizes the situation in more detail.
Unglad that they're trying to sell a war of criminal aggression on Iran in 2025, using the exact same propaganda as they used to criminally invade Iraq in 2003 ("he was about to build nuclear weapons of mass destruction and destroy us!"), and the media is bending over backwards to fall for it again.
I guess this is what happens when the warmongers never faced any kind of consequences. No war crimes trials, not even any loss of career. Bush's old speechwriter David Frum, who coined "Axis of Evil" then later on pretended to regret his role in pushing the Iraq invasion, is just a writer for The Atlantic (where he's doing it again). Bush retired happily ever after to paint pictures of puppies.
Also unhappy about genuinely increased nuclear risk. Contrary to popular belief, "Nuclear countries attacking non-nuclear countries in the middle east over and over and over until we finally convince them to get nukes after all to deter us" does not keep us safe.
From Europe, comparing what I knew of the American public reception at the time and what I've seen floating on twitter, I notice there's a lot of conservative/republican/right-wing pushback, pretty much across the spectrum, against hawkish positions. Some feel fringe (like "so much pro-russia that hitting a Russian ally is unacceptable"), but a lot is semi-principled against foreign wars and "well they hate us now that we bombed them, so better bomb again".
In 2003 my vision was severely filtered by mainstream media, and I wasn't aware of *any* right-wing opposition to the Iraq war. Was there any?
I'm Canadian (and was fourteen at the time) but yeah I had the same impression
When I met some Americans during a trip, the only discussion related to the Iraq invasion that I remember was one American grandpa asking "why do Canadians hate us? Is that what they teach in school, to hate?"
Seven Republican members of Congress voted against the war: Lincoln Chafee, Jimmy Duncan, John Hostettler, Amo Houghton, Jim Leach, Connie Morella, and Ron Paul. (Chafee and Leach later became Democrats.) Then-prominent conservative columnist Robert Novak was also against it from the beginning.
It's true, though, that opposition to the war was an extremely fringe position on the right. 9/11 gave George W. Bush an absolutely enormous amount of political latitude, which no subsequent American politician has come close to rivaling.
"the media is bending over backwards to fall for it again"
I'm the biggest MSM critic I know, generally, but in this instance have to disagree.
CNN obtained and published a leaked US intelligence assessment that the bombing set Iran's nuclear-fuel enrichment efforts back only "a few months"; that report is right now a lead story in the NYT, WaPo, WSJ, Associated Press, Reuters, PBS, NPR, etc.
All of those news organizations have repeatedly for a couple of weeks now been pointing out, and re-posting the video of, US intelligence chief Tulsi Gabbard testifying under oath in March that "Iran is not building a nuclear weapon and Supreme Leader Khamenei has not (re)authorized the nuclear weapons program that he suspended in 2003."
This is superficial. Irans nuclear program is acknowledged by Iran. There are questions about how close they are but none about whether the program exists. Mood affiliate less.
Iran was not building nuclear WMDs, they were not about to build nuclear WMDs, and it was known that they were not about to build WMDs.
The main lie in 2003 wasn't that he had nuclear fuel (though that was a necessary component lie), the main lie was "he's building nuclear WMDs and about to attack us."
Why it is enriching uranium to 60 percent then?
So that if it needs to change its mind later, and decide it does need a nuclear deterrent after all (for instance because we keep attacking it), then it can.
In 2003, Iran was trying to run as quickly away from the "Axis of Evil" as possible. America has a lot of bombs, and being characterized as "not just in opposition to America" but part of "Evil BAD Guys" (again, not China, not Russia, not Ukraine, just the ones we thought needed pruning)
Strongly disagree. The actual outcome of this seems to have been "severe setback or possible destruction of Iran's nuclear program at basically no cost to the US", which is a fantastic result. People pattern-matching to the Iraq war are mostly just falling to the temptation to revive their expired "I told you so" credentials, like someone who sees a new crypto-like trend and immediately invests in it because he wishes he'd invested in Bitcoin early on.
US intelligence has it that is was neither a severe setback nor a destruction:
Two of the people familiar with the assessment said Iran's stockpile of enriched uranium was not destroyed. One of the people said the centrifuges are largely "intact." Another source said that the intelligence assessed enriched uranium was moved out of the sites prior to the US strikes.
"So the (DIA) assessment is that the US set them back maybe a few months, tops," this person added.
https://edition.cnn.com/2025/06/24/politics/intel-assessment-us-strikes-iran-nuclear-sites
Is this the same US intelligence that reported Saddam Hussein was working on WMDs in 2003, that everyone later claims was bad intelligence? Or at least, some claim was bad intelligence, while others argue it was intelligence that everyone agreed was correct at the time until they didn't?
Or is this a different US intelligence shop, whose trustworthiness is the opposite of the former US intelligence shop? Do these anonymous US intelligence reports have serial numbers that we can use to identify the shop and its trustworthiness, or something?
US Intelligence is lining up behind Tulsi and the whole "this is a severe setback" I'll grant that you didn't know that while writing the above... but... maybe try thinking like a military guy? It's a Very Bad Thing to be openly seen to be toothless. America would do a LOT to not seem toothless. (Like bombing a "next to innocent" Iran that nobody really likes).
> America would do a LOT to not seem toothless.
True, but there are weaker foes to pick beat down than Iran. Cuba, perhaps. Or Venezuela.
If our bombs in fact failed to destroy Fordow, or destroyed an empty hall while the centrifuges were spinning merrily away someplace else, then A: Iran knows that we are toothless, and B: Iran can tell anyone they like that we are toothless, and if necessary back it up with a private tour.
If we hadn't bombed Fordow, we would still have ambiguity on that point; maybe we actually can bury deep targets, and maybe we're willing to bomb widely enough to get all the dispersed targets, and maybe Iran was *just* short of being offensive enough to get that response. Now everybody knows that Iran-level offensiveness is solidly in the range of "America will drop some bombs in the desert and declare victory".
So an awful lot is depending on those bombs having been very effective.
Yep. Enough to have backups for the backups.
The second point isn't declaring that we are toothless, it says we have a very big stick and very poor aim. This can still have desirable consequences for our international policy.
There are still very strong voices in our government who did not want to bomb Iran. Unfortunately, with the demise of Pax Americana, we have higher priorities than nourishing our relationship with a prospective ally.
That's possible (as although far from certain, it's just one anonymous report from someone without insider data), but even in the worst case scenario the US got a free shot at a major adversary (and if it did fail, that's useful information to know about the need to improve the MOP).
In the worst case scenario, taiwan is already under siege. Yay, we aren't at the worst case scenario. (Ditto Putin taking over Kiev). The world stage is larger than widdle Iran, who isn't a major adversary at this point (and not even very well backed by Putin, who is staying out of America's war).
> Putin taking over Kiev
Why would this be bad for the US?
(also, the worst case scenario being "we may have only held the enemy back from achieving their primary war aim by several months" is still pretty amazing for a single airstrike! Imagine if a single minor aid package to Ukraine pushed back Russian war goals by several months, it'd be an obviously amazing success).
Maybe. Maybe it also strengethened the resolve to get nuclear weapons in Iran and elsewhere, accelerating proliferation globally. Maybe it weakened Khamenei's grip on the country, destabilizing it and leading to more chaos in the region. Maybe more more aggressive leaders within the military will gain influence and escalate further.
I have no firm opinion on whether the airstrike was good or bad, (I am leaning towards bad, but can be convinved otherwise). We don't even know whether it succeeded at this point.
I do think it's a bit naive to assume that the "worst case" is "We didn't achieve our stated objectives, but at least we achieved them partly and we got to test a weapon."
The outcomes of interventions in the middle east (and elsewhere) are very hard to predict, and the plausible worst cases are really bad. This doesn't mean interventions are always wrong, but it means that "worst case is still good" assessments usually are.
In terms of resolve, they were already pretty resolved (even putting aside intelligence reports, iaea said they'd basically abandoned all enrichment limits the week before the strike). This is far more likely to have deterred than helped that. Weakening khameini's grip is almost certainly positive - he's pretty close to a worst case scenario (both in that Iran under him is one of the worst and probably single most implacable hostile to America places in the world, so change is likely to be for the better, and in that the other known factions there are all clear improvements).
The main risk of intervention is getting bogged down in the mess (like Vietnam or Afghanistan). And that was clearly avoided here (and due to strategy, not luck). So it'd be a pretty hard stretch (maybe not literally impossible, but really hard) for this to turn out net negative for the US.
Depends on the amount of nuclear radiation on the Germans. We might get more uptake in the tentacle pornography business.
What? Surely you're aware that we hired neocons in the last Administration? Neocons do not come up with smart ideas.
Also pretty unglad.
Less convinced on the blame. Do you think that, in the counterfactual where we hadn’t manufactured the consensus to invade Iraq, Iran would have no interest in nuclear weapons? I get the impression that the incentives were already pretty strong for a third-tier power to join the nuclear club.
It's more than just Iraq in 2003. E.g. Gaddafi also gave up his nuclear program in 2003 and then the US destroyed Libya in 2011 anyway.
It's not that there's zero incentive otherwise (of course all else being equal a country might want to have more leverage). It's that the pretty consistent lesson has been "non-proliferation doesn't protect us."
fwiw Iran is even still a signatory of NPT (though it's using the prospect of leaving NPT as itself a deterrent)
Dunno. This sounds like deontological ethics: never violate the war-crime rule. I think consequentalist ethics say a few plants bombed down vs. X % of Iran nuking Israel is a good tradeoff. Iraq was for the consequentalist different, as it resulted in immense human suffering - but don't know whether foreseeable. Also Iraq WMDs were clearly a lie while centrifuges in Iran not.
And now there's a Y% chance of Iran nuking Israel. Y and X are both unknown, but I'm guessing Y > X.
The odds of Iran nuking Israel *this year* have gone down, yes. Enjoy it while it lasts.
AAAH interesting point. Their capability of doing it this year went down, but their DESIRE to do it later went up, you mean?
Precisely. The old policy of "we *could* build nuclear weapons any time we want, but we have chosen not to do that yet - let's talk about how to keep it that way", is no longer viable. "We *have* nuclear weapons, and you *will* respect that", is now much more appealing.
I think the usual reaction to "we *could* build nuclear weapons any time we want, but we have chosen not to do that yet - let's talk about how to keep it that way" was typically "you're *saying* you haven't, but we have reason to believe you're secretly building them and lying about it". If so, then the former policy was never viable, and the second policy was always in effect.
Do we have a way to find more evidence either way?
Short of putting an army of occupation in place in Iran, and/or bombing all of their industrial infrastructure into oblivion, there is *nothing* we can do that will guarantee that Iran hasn't secretly built a modest nuclear arsenal and hidden it in a cave somewhere. Doubly so after last week. Every possible course of action comes with a side order of "...and maybe it doesn't matter because Iran already has a secret nuclear arsenal".
So anyone pointing to any particular course of action and saying "You can't ask us to do *that*; because it might lead to Iran having a secret nuclear arsenal and then they'll kill all the Jews", is a fool or a deceiver, and in either case not to be trusted.
Look for people whose plans explicitly allow for, yes, Iran might have a secret nuclear arsenal and here's how we're going to live with that. Because those plans *are* viable, and they're the only ones that are viable.
BBC: "The US strikes on Iran's nuclear facilities did not destroy the country's nuclear programme and probably only set it back by months, according to an early Pentagon intelligence assessment of the attack.
The Islamic Republic's stockpile of enriched uranium was not eliminated in Saturday's bombings, sources familiar with the Defense Intelligence Agency evaluation told the BBC's US partner CBS."
Consequentially it's flirting with the same (if it escalates it will be even worse than Iraq, since Iran's bigger etc.). I agree that Iran is better prepared than Iraq was, is good at calculating carefully-scaled retaliation etc., so the US is less likely to actually try for a regime change war. But a lot of the media are *trying* to encourage a regime-change war.
Dishonestywise it's near identical. We still know for a fact that Iran was not building nukes and had no active plan to build nukes. We still know that the goal is hegemony, not nuclear non-proliferation.
Source for Iran having no plan to build nukes?
Banned for this comment.
Ugh.
I encourage you to make more focused, directly-relevant points, and dial down the sarcasm.
I know you're referring to how Israel's politicians have exploited the suffering of the victims of the Nazi holocaust, using them to paint their aggressive expansionist wars as defensive; to paint their opponents' aims and strategies as mindless genocidal rage; and so on. (Both "directly" in specific cases and as background-colour in general; and both to keep allies sympathetic, and to keep Israelis scared and compliant.) But if someone isn't already familiar with that, then this comes across as a weird antisemitic non-sequitur.
As the thread-starter comment, you're in the right to hate my sarcasm, it does lower the signal/noise. Maybe I'm on the hook for an apology.
On the other hand, Sarcasm is the effective time-tested communication method for saying "NOT(P) IS THE EXTREMLY OBVIOUS & UNAMBIGIOUS TRUTH, AND ANYONE DISAGREEING IS DUMB AF", where P is the proposition that the Sarcasm pretends it believes in.
You might argue that what's "obvious" is different depending on the worldview and particular sensory and ideological bubble, but, well, as Alan Kay once said it: A Change of Perspective is Worth 80 IQ points, it's a self-inflicted wound to hobble yourself by propaganda so bad that you're incapable of seeing the obvious anymore. It's all part of the blame.
> then this comes across as a weird antisemitic non-sequitur
Yes, exactly. And it's this fear precisely which Hasbara exploits to defend a contemporary genocide using the legacy and the PR prowess of an 80 years old genocide. It doesn't matter if they're "familiar" or not, actually, they will pretend they're unfamiliar anyway to put you on the defensive and re-gain rhetorical momentum away from discussing the current, currently-dying victims, in favor of victims of which none remain but bones.
Nobody ever said that arguing against genocide denial will be politically correct or nice.
I'm not looking for a personal apology, it's just a suggestion.
But yes my point is the one that what's obvious depends on worldview and bubbles. The West is unfortunately already neck deep in this propaganda and has been since at least the 70s (when some of the popular propaganda slogans came out, at the time to justify trying to permanently steal Egypt's Sinai and Syria's Golan Heights, like Golda Meir's sadistic "peace will come when the Arabs will love their children more than they hate us")
I can be quite sarcastic myself (as can critical scholars like Norman Finkelstein) but it's best when the information is already at the point where the reader can't in good faith deny it's obvious. + I do bet the balance of pros and cons is shifted where there is real pain like the Holocaust, and real fear like antisemitism (which both really exists, and is exploited to terrify those who were propagandized into genuinely believing the propaganda)
Agree that countering propaganda is never entirely pretty, and even when you argue everything 100% thoroughly and rigorously in 100% good faith, you'll still see many bad faith propaganda replies anyway lol
E.g. when you write a thoroughly fact-checked book published by University of California Press in 2005, Alan Dershowitz will still pop up and try to get Governor Schwarzenegger to ban it, and then when he says no, lie that you you called your Holocaust-survivor mom a kapo
Hahaha, fucking Hasbarists man, right?
My Jewish-sounding real name does confuse them in real world arguments a good bit.
Weird that so many separate individuals here write on the exact same topic and with the exact same style as that perma-banned user LearnsHebrewHatesIsrael.
Strongly agree.
It's also possible that there's something about the anti-Israel media these people consume that causes them all to write in the EXACT SAME WAY*, but yeah sockpuppets is a more parsimonious explanation.
(*very out-of-place on ACX, I should add)
Hmmm, so the 5 to 7 people I saw the last couple of weeks arguing against Israel's demented forever war are all secretly me? man, I'm fucking awesome.
> anti-Israel media these people consume that causes them all to write in the EXACT SAME WAY*
It's infinitely amusing you're insinuating you're more "Rationalist" than me with a snide sub-remark about an out-of-place style (as if you have personally read and stylistically reviewed every one of the 1500 comment on average to each ot) but you didn't stop to think of Outgroup Homogeneity Bias on your part as an explanation for why all anti-genocide posters sound the same to you, rather than the usual facile one of "anti-sementic media !!!".
What a bizarre and self-righteous pair of characters.
Hmm, I had a number of comment exchanges with LearnsHebrewHatesIsrael ( later LearnsHebrewHatesIP ), some heated, some not, and this commenter seems different to me.
I've made my argument in the thread for why the original post and in particular my exchange here feels strikingly similar to exchanges I had with both LHHI and suspected sock accounts, but I will also acknowledge it's ultimately subjective, and whether or not these feel like sock accounts might be based on the kinds of exchanges one had with LHHI.
Many Thanks! You may well be right. I'm just getting a somewhat different impression - but my memory of LHHI may have faded.
I don't think Never-Again-All-Over-Again sounds like LHHI. LHHI sounded smarter and more furious. Anyhow Never-Again has a blog, and LHHI didn't.
I disagree, obviously, and apparently so do a few others. I think this sounds exactly like LHHI experimenting with being unrestrained.
And that's not impossible, of course! I personally avoid writing with frothing hyperbolic aggression, but that's a choice. I'm fully capable of writing at a far lower level, as are you and most of the other people here.
But last...where's this user's blog? If one clicks through his username on a mobile browser here on Substack, his profile page says he hasn't published any posts yet, and he has no subscribers? Again, the mobile browser version of Substack is crap, so it's totally possible I'm missing something there.
At some point, I have to ask whether it matters. If it doesn't walk exactly like a duck or talk exactly like a duck, because it walks and talks a bit more like a stork with avian rabies, of what use is the duck hypothesis?
If ducks are already banned, then it might be easier to decide to ban another duck than to debate whether or not to ban rabid storks as well.
Naw, you're right about the blog. I was on my phone, just glanced at the page, & mistook the list of blogs the person subscribes to for a list of posts. But I still don't think this sounds like LHHI. As I said, LHHI sounded cleverer and angrier than this person. You're calling the present posts instances of LHHi experimenting with being unrestrained? But when LHHI was unrestrained he was *way* more intense than this person! He called people murderers, said various people deserved to die, swore. You could almost hear him shrieking when you read his diatribes. So that's one difference.
And if you look at the other blogs the person subscribes to, they're all nerdy tech blogs -- no politics. And 2 of the blogs look geared to beginners, and I don't think his work was that low level. At one point I had a long exchange with him about AI training and capabilities and he knew a lot.
I acknowledge my accusation is effectively impossible for me to prove, but based on the occasional exchanges I had with LHHI, this strikes me as being *exactly* the same personality generating slightly different approaches to the text. The formatting of the original comment, (including always proactively including at least one footnote link to create an impression of credibility), the snide, personal-attack-y responses to any push-back, and the general style and tone of LHHI being less intellectual and more bombastic; it all just feel *intensely* LHHI to me. Even the username feels very much in LHHI's confrontational style.
But for me, perhaps the most noticeable signature of an LHHI (or sock) post is the obnoxious lack of acknowledgement or apology for the way his personal tribal affiliation imbues every statement with a fundamental assumption of black-and-white-good-and-evil. That stands out amongst a population of rationalist ACXers who tend to be very careful indeed about acknowledging and interrogating their personal biases.
> "But when LHHI was unrestrained he was *way* more intense than this person! He called people murderers, said various people deserved to die, swore. You could almost hear him shrieking when you read his diatribes. So that's one difference"
But it's also possible that LHHI doesn't want to have his sock accounts banned, and so he's not *quite* rising to the level of what got his "main" account perma-banned (although I believe the first comment in this thread warrants it, sock puppet or no).
Yeah, is someone compiling a list? I think I recognize maybe two sock puppets currently at work, but I haven't been paying enough attention to be sure.
Or perhaps I've been paying too much attention, but that's what the list is for.
I blocked Hindi's Ghost, as I was fairly certain it was a sock (in addition to being obnoxious).
Annoyingly, blocking apparently removes one's ability to see *any* content related to the blocked party - including one's conversations with third parties, like the exchange I'm having with you - so I'll wait until this thread naturally dies off to block.
The topic is a 12-days-old war, and there hasn't been any banning in the last 2 open threads which covered this period, so how could that user have written about a topic that began happening after they're banned?
> so many separate individuals
Perhaps it's outgroup homogeneity bias. There is only so many ways to phrase "Killing civilians is bad and respecting international law is good", when 2 individuals of the same convictions run out of ways in the same way, you tend to notice. When they successfully find different ways to phrase it, you tend to overlook it.
Weird that so many separate individuals here have the exact same pattern of affecting confusion at accusations of sock puppetry and then concluding with deliberately insulting condescension!
So weird!
I didn't start this by meaning to be condescending, but instead of apologizing like the normal me, your response is strongly shaping me into "If The Shoe fits" as the reply of least resistance.
Why did you assume condescension in my words? I understand why a Pro-Israel Hasbarist could be offended at the (very) mild exasperation, but you didn't participate in this thread or any other one on Israel-Palestine on either side of the issue so I'm confused what offended you in my reply.
> Weird that so many separate individuals
Weird that you keep bringing up how So Many People are sock puppets of me (or, apparently, another user entirely) without actually mentioning the actual usernames. I suppose if you make a direct accusation, those people can chime in here and offer to you whatever will satisfy you that those accounts are different. Or maybe I can point you to meaningful differences of opinions in comments those accounts made before your reply that reasonably point towards different minds controlling them.
As I've replied in suspiciously - even *impossibly* - identical exchanges with commenters who very much appear to be sock puppets of LHHI...nah.
I've reported your initial comment to Scott in the hope you will promptly be banned. I urge those who share my objections and/or suspicion to do the same.
Then I urge everyone to block this user so their content is at least invisible on an individual level.
Scott may not use his ban hammer often, but we can certainly use our own.
The resemblance is uncanny.
So I'm not the only one to have noticed that.
Everyone seems.to be retconning It as about nukes, but it was largely about chemical.and Biological. weapons at the time. Iraq had and used such weapons at previous times , but the UN was successfully dismantling its programme at the time of the invasion.
They repeatedly lied that he was building nuclear WMDs too. And remember “we cannot wait for the smoking gun to be a mushroom cloud.”
It's true that they lumped this together with other kinds of weapon. Presumably that was so they could motte-and-bailey it (“see those are WMDs”).
By the time of the actual invasion, Bush had mostly given up on the nuclear weapons charge, but that was after various attempts (aluminum tubes supposedly used for centrifuges, forged documents showing that Iraq had made an unsuccessful attempt to purchase yellowcake) had been discredited. The key sentence from Bush’s announcement of hostilities: “The danger is clear: using chemical, biological or, one day, nuclear weapons, obtained with the help of Iraq, the terrorists could fulfill their stated ambitions and kill thousands or hundreds of thousands of innocent people in our country, or any other.”
It's also that the chemical weapons really are the odd man out from the ABC weapons. Occasionally used nasty warfighting tools with questionable tactical uses but without the vast apocalyptic implication of the others.
Nuclear weapons are the same way. If used in moderation.
(Russians think differently, but they've always made weapons/armament with the philosophy that "we'll be fighting on our soil, which is precious.")
This is a bombing campaign, not a ground invasion. These countries will try to build nukes no matter what because of the leverage it would give them, and the only way to guarantee that they won't develop them is to make sure they are physically incapable of doing so. The less countries there are that have nukes, the better. That's all there is to this.
Well said!
This is a bombing campaign, not a ground invasion. So what does that have to do with countries not having nukes and "being physically incapable of doing so"?
For all the talk about Iraq as an analogy here, pretty much everyone is missing the most important difference. It's unclear whether Iraq seriously wanted to acquire nuclear weapons, and they were certainly nowhere near any such goal, but Iraq was certainly rendered physically incapable of building nuclear weapons for many years.
Because there *was* a ground invasion. Iraq was bombed, invaded, conquered, forcibly regime-changed, and placed under military occupation for about a decade. *That*, is something that can take a nation off the list of potential nuclear powers.
Just the bombing, is mostly just for Feeling Good, like you're a Big He-Man.
The left used to be associated with nuclear disarmament but now it seems to be: nukes in the hands of the US - bad; nukes in the hands of brown people - good!
Show me someone saying this.
“Seems” is usually a sign of pretty weak evidence.
Nuclear countries bombing non-nuclear countries until we force them to become nuclear-armed is not "nuclear disarmament."
Where is the Iran-built nuclear power plant, I wonder. After all this peaceful time.
Oh come on. Ask Google "Where is the Iran-built nuclear power plant" and Google tells you where in its large-font main result. In Bushehr, on the Persian Gulf coast.
What are you even trying to do?
You’re adducing a German-built and Russian-rebuilt plant from the 70s as evidence for the peaceful nature of Iran’s nuclear efforts these past 30 years?
Walk me through it.
There's a trope in armchair geopolitics that gives all agency to the West and none to anyone else. The West isn't forcing Iran to become nuclear-armed. They want to because they want the power, and the West is trying to stop them.
Iran has spent the past twenty years conspicuously *not* becoming nuclear-armed, even though they could have done so at any time during that period. They clearly value having the option on the table, but until fairly recently they valued it mostly as a negotiating point and an emergency capability.
It is not entirely clear how they feel about that now. And you're right, it's entirely their decision and their responsibility. But the US and Israel have deliberately and substantially changed the incentive structure.
Even worse, Obama signed a deal to stop Iran from making nukes and then Trump ripped it up out of spite, leading to the current predicament.
There's a trope in Western internet commenters where the West has zero agency, and the fact that we keep attacking and trying to dominate Middle Eastern countries has zero to do with the fact that they want to deter us.
Obviously Iran has agency too, even in ways that aren't about the West (e.g. the Iraq-Iran war was historically important).
It's bizarre that you pretended I denied this obvious fact, to deflect from the West's agency.
There's a history here ...
https://en.wikipedia.org/wiki/List_of_wars_involving_Iran
Well yes, Iran (aka. Persia) is a really old country.
Yeah I get that they don't like Pax Americana, and theoretically more of those places having nukes would make us have to respect their autonomy more than we do. But not only is it dangerous to proliferate such weapons, it just turns those nations into little Americas in their own neighborhood, they were never gonna need to nuke their tiny rivals but they still become regional hegemons that get more of what they want on the big stage and can leverage that advantage to stay there. It's not any real kind of equity between peoples as they'd simplistically imagine it.
You've mistaken me for someone who wants them to have nukes.
I wrote my reply to that guy before you started your subthread discussion, so I wasn't in any way discussing your views.
Ah ok, thanks for the clarification
>"These countries will try to build nukes no matter what because of the leverage it would give them"
False. There was an anti-nuclear Iran deal, which Trump withdrew from because of his ego.
In fact, the US has consistently *punished* nuclear disarmament. Saddam had no nuclear program after 1991, so the US lied that he did and destroyed Iraq in 2003. Gaddafi gave up his nuclear program after 2003, so the US destroyed Libya in 2011.
The anti-nuclear Iran deal was to end by about now, only with Iran having a much better economy and be better able to produce nukes, possibly shortening their nuclear timeline while making their people better off.
Though I agree with Trump that the Iran nuclear deal was a bad deal, I disagree that he should have pulled us out of it. Because what one president does concerning another country ought not be unilaterally undone by another president, because it undermines the integrity of the United States.
It isn't just anti-nuclear though, the other threat is to the dollar's pre-eminence, Saddam rashly started dealing oil in Euros. Gadaffi had some idea of starting a bank of Africa, which would be independent of the dollar. Also, I think in the early 2000s Iran founded an oil "bourse" to sell oil in currencies other than the dollar, so they've been long overdue for a clobbering. No doubt an agreeable side effect for the US of the latter is to persuade Putin that Trump can take decisive action and isn't all mouth as Putin may have believed.
> In fact, the US has consistently *punished* nuclear disarmament.
Everyone always punishes disarmament. That's the point of having armaments.
---
When the Cambrian measures were forming, they promised perpetual peace
They swore, if we gave them our weapons, that the wars of the tribes would cease
But when we disarmed, they sold us, and delivered us bound to our foe
And the Gods of the Copybook Headings said: 𝘚𝘵𝘪𝘤𝘬 𝘵𝘰 𝘵𝘩𝘦 𝘋𝘦𝘷𝘪𝘭 𝘺𝘰𝘶 𝘬𝘯𝘰𝘸.
The ending of that poem also seems apt:
And that after this is accomplished, and the brave new world begins
When all men are paid for existing and no man must pay for his sins,
As surely as Water will wet us, as surely as Fire will burn,
The Gods of the Copybook Headings with terror and slaughter return!
The whole poem is great. This context seemed to call more for the disarmament verse.
The conclusion seems better for a UBI thread.
The WMD evidence against Iraq was clearly ginned up. Do you believe that Iran is not illegally enriching Uranium? There seems to be no dispute on that point.
Do you mean to say you believe that Iran *is* illegally enriching Uranium?
Because two days ago, the President of the United States and the Prime Minister of Israel assured us that Iran's nuclear facilities had been "totally destroyed". If they're still enriching Uranium, what was the point?
FWIW, I don't believe Iran is enriching uranium *today*, though I'm not hugely confident in that. Better to lay low for a few days or weeks and figure out how best to proceed. But I'm betting that Iran will be enriching an awful lot of uranium, 8-12 bombs' worth, later this year.
No of course not today. But a month ago? Definitely.
>I'm betting that Iran will be enriching an awful lot of uranium, 8-12 bombs' worth, later this year.
I'm betting that Israel already has a plan to prevent this.
Anything beyond blind faith to support that wager?
A long history of Israel defending themselves effectively and proactively.
1. Bombing Saddam, who had no nuclear program of any kind in 2003, helped convince Iran it should have a latent capacity to change its mind later if needed to deter us. If we bomb Iran for knowing it might need to deter us later, then we're begging it to start deterring us now.
2. The US's motives have nothing to do with legality or self-defence, as proven in 2003 in Iraq (as well as by destroying Libya in 2011, to punish Gaddafi for cooperating and giving up his nukes program in 2003, etc).
3. Iran's enrichment is legal under NPT when used for nuclear energy. Though since Trump withdrew from the anti-nuclear Iran deal in 2018, it has stepped it up, and recently IAEA said it was in violation of something related to inspection / declaration, but that's not at all proof of actually building a bomb.
4. US spies and the IAEA agreed that there was no evidence of Iran actively building nuclear weapons.
5. The only country in the middle east that actually *has* nukes is Israel, and few think anyone should bomb Israel just to destroy its nukes (which tbf would also be suicidal). Israel also isn't a signatory of NPT and doesn't allow IAEA inspections (technically its nukes are an open secret, for tricky Israel-US reasons).
6. The US isn't actually scared of an Iranian first strike (which Iran knows would be suicidal). It's scared of not being able to totally dominate the Middle East anymore. Unfortunately that same will to dominate is why it keeps, over and over, giving them a reason to feel the need to get nukes.
Enrichment to 60 percent is not needed for nuclear energy.
US spies?
Does US has spies?
Or only SIGINT.
I mean intelligence agencies. Idk the operational details which I assume are probably classified anyway
I think the difference here is that Israel has not declared an intention to wipe Iran off the map by any means necessary, and does not fund Jewish terrorist cells that launch rockets into Iran and suicide bomb Iranian buses. For all Netanyahu’s faults, he seems unlikely to launch a nuke into Tehran tomorrow, whereas if the Ayatollah were to get a nuke the fate of Tel Aviv is far less certain.
Not sure what is your point here. Obviously, Iran is not the good guy here, but the fact that Israel is somewhat better than Iran does not make it the good guy either.
Also, it is a bit funny to portray Israel as one avoiding civilian casualties as much as possible (this is understandable, though -- as long as they can get away with everything by yelling "human shields", why take the risk?).
I’m not saying Israel is the good guy. I think both Netanyahu and the Ayatollah are evil psychopaths who have visited untold misery on the region. But Israelis have more of a chance of getting out from under Netanyahu, and if I had to pick a place to live, I would choose Israel.
There is zero chance of Iran destroying itself by nuclear first-striking Israel, which Iran knows has nukes.
Again: Iran is a country with goals and strategies, not your cartoon propaganda’s suicide death cult.
(The chances of Netanyahu nuking Iran are also low, I hope. But a risk there for Israel would be if Netanyahu starts to fall for his own propaganda—that he starts to actually think Iran is a suicide death cult immune to deterrence logic—so he mistakes their deterrence for aggression, nukes them in the hopes that Pakistan’s deterrent umbrella is a bluff, and then it isn't a bluff.)
You don’t think so? The regime in Iran is widely despised by the people. What if the Ayatollah was facing revolution from within and the threat of Reza Pahlavi replacing him, while Israel was striking their military sites from the air? You believe in the capacity of evil people who have nothing left to lose, to exercise restraint on the behalf of humanity?
If Iran has a nuclear deterrent in your scenario then why is the US ignoring their deterrent and trying to destroy them? Lol goofy
1. Whatever Iran's motivations are, they indisputably want to become a nuclear power. I don't think you have to blame the US for that. Most countries would like to be a nuclear power. In any case the origin of their motivations is irrelevant. This is realpolitik not therapy.
2. So? Our motives are our own self-interest. That's sufficient. There are many good reasons to not want an unhinged Muslim theocracy to have a nuclear weapon.
3. Iran is enriching to 60% which is far beyond the 5% needed for energy. They've also denied inspections. "That's not at all proof of actually building a bomb" is suicidally naive in my view. What's proof, Tel Aviv disappears? Given the stakes, proof isn't necessary. Preponderance of the evidence suffices. This isn't mock trial, it's the real world with real consequences. Better to kill ~1000 Iranians on a reasonable hunch then risk the death of a million Israelis.
4. Again, willfully naive. They're doing enrichment which is the hard part.
5. Israel is a stable ally. It's in our interests for them to have nukes and they've demonstrated that they can be responsible. Also they already have them so there's nothing to be done. This illustrates why we have to be proactive in preventing Iran from obtaining them. Once they do our options narrow.
6. Yes, we're pursuing our own rational self-interest. I don't know why you find that objectionable. Existential fear isn't a prerequisite for military action. We're the global hegemon. We decide what's appropriate.
>"is suicidally naive in my view. What's proof, Tel Aviv disappears?"
You'll recall Bush's focus-grouped slogan in 2003, to stop anyone questioning his lies to justify his criminal aggression: "We cannot wait for the smoking gun to be a mushroom cloud."
You'll also recall that Iran isn't a suicidal death cult, or a country of idiots who don't know about Israel's nukes. Obviously Iran isn't going to destroy itself by nuclear first-striking Israel.
>You'll also recall that Iran isn't a suicidal death cult
Yeah, right (quotes below from https://www.jns.org/irans-sickening-use-of-child-soldiers/ )
>“An Iranian government representative admitted in a closed-door sub-commission hearing that children did participate in the war against Iraq,” Refworld continues. “In a series of rulings issued in the autumn of 1982, Ayatollah Khomeini declared that parental permission was unnecessary for those going to the front. … Iranian officers captured by the Iraqis claimed that nine out of ten Iranian child soldiers were killed.”
>“Boys as young as nine were reportedly used in human wave attacks and to serve as mine-sweepers in the war with Iraq,” Refworld found.
>Often bound together in groups of 20 by ropes to prevent the fainthearted from deserting, they hurl themselves on barbed wire or march into Iraqi minefields in the face of withering machine-gun fire to clear the way for Iranian tanks.
You said:
>Obviously Iran isn't going to destroy itself by nuclear first-striking Israel.
You do _NOT_ know that.
Just to be clear:
Your attempted proof that Iran's leaders want to die (to the point that nuclear deterrence logic doesn't work on them)........
..........is an article about how they send others to die instead, while they're safe at home (on account of how they don't want to die)?
But Iran funds a bunch of suicidal death cults (Hamas, Hezbollah, Houthis)
Yes Bush was likely wrong. An argument being wrong for one set of circumstances in 2003 doesn't make it wrong for a different set of circumstances in 2025.
>Obviously Iran isn't going to destroy itself by nuclear first-striking Israel.
That's not obvious at all. In any case, it's in our interest to prevent them from obtaining nuclear weapons. That's really the only thing that matters.
I'll also point out the issue with Iraq II isn't really that they didn't have WMD, it's that we were completely and naively unprepared to rehabilitate the Iraqi political structure. If it had gone as well as our post-WW2 occupation of Japan then no one today would care that Bush lied about WMD. This is a results business. As long as we prevent Iran from obtaining WMD without significant cost to us then that's the only thing that matters.
I think 2 and 6 are clearly wrong, but they're readings on the motives of US and Iranian leaders which are hard to bring evidence for or against.
I'd like to push back on 4. According to the IAEA, Iran has amassed 900 lbs of 60+% U-235 uranium, the largest stockpile of highly enriched uranium by a nation without a nuclear weapon in the world: https://www.aljazeera.com/news/2025/5/31/iran-increases-stockpile-of-enriched-uranium-by-50-percent-iaea-says
That's Bayesian evidence that Iran is working towards nuclear weapons; we would certainly be more likely to find large stockpiles of highly enriched uranium if Iran was aiming for nukes than if it wasn't. (And Iran's accusation of the IAEA looking at forged Zionist documents isn't credible to me; Iran has approximately zero credibility here. Nations intent on proliferation always deny it, and Iran specifically shamelessly lies to the media all the time.)
This seems like yet another confirmation of the Law of No Evidence: https://www.lesswrong.com/posts/j9HoG56Y6KuopSzdn/law-of-no-evidence
> few think anyone should bomb Israel just to destroy its nukes (which tbf would also be suicidal).
Would it? Whatever nukes Israel has, it lacks the ability to launch them very far, so if there were ever a US President who decided "actually we really do care about the non-proliferation thing" there's not much the Israelis could do to respond.
Israel has intercontinental ballistic missiles, though numbers and details are quite fuzzy. Israel also has submarines capable of launching nuclear cruise missiles, and while it would take them a few weeks to sneak across the Atlantic, they're reasonably quiet subs in a big noisy ocean.
I suspect it would, as the United States would likely retaliate in its behalf. Since Israel isn't actually a NATO member, the US is under no obligation to do so, but it seems very likely, since the US gave Israel nukes in the first place.
> since the US gave Israel nukes in the first place
Wasn't it France?
So it seems. I thought we actually gave them nuclear weapons, but it seems they developed them themselves with a secret reactor with French assistance.
So my backup to the estimation that the US would retaliate to an attack on Israel was wrong, but I will still say that, given the support the US has provided to Israel in the past that they would still retaliate. Poking Israel is like poking the US, only a lighter poke.
I assume it has submarines and second-strike capability. Ofc it's hard to know specifics given the ambiguity policy.
(The point is more illustrative, like "few here would think Iran has the right to try to attack to destroy Israel's nukes."
In real life, if it was the US who decided that, it wouldn't need to attack. It's Israel's biggest enabler, so all it would have to do is say "no more settlements, no more military aid, no more occupation, time to withdraw to the green line and make peace with your neighbours and sign NPT. The PLO and Arab League are already down, and if we release Marwan Barghouti he can convince the rest of the Palestinians.")
Here's a precedent / parallel.
In 1981 Saddam had bought a French-made nuclear research reactor, and France-supplied highly enriched uranium (the standard fuel at the time). It was above board, legal, under IAEA supervision and everything.
So Israel bombed it, trying to stay the only nuclear power in the region.
But the predictable and actual effect was that it *convinced* Saddam he *did* need a nuclear weapons program after all, and that it should be secret. The Iraqi secret nukes program, created as the consequence of Israel's counter-productive attack, was only ended after 1991 by the UN.
After a successful US invasion, it should be noted.
(Not advocating a US invasion of Iran by any means, but I felt compelled to point out that that 1991 date didn't occur by accident)
Yes. (That was different in the sense that it was UN-approved—which might also affect how Saddam would interpret what it means for future threats and deterrence—but nevertheless, yes.)
If only they’d had UN approval in 2003, things would have worked out so much better.
Yep. You can't just say "some guys claimed this about another country 20 years ago but were wrong, therefore these other guys are also wrong about this other country."
You can say "Given the known fact that Iran was not building nukes and about to nuke us; and the known fact that the same people are telling the same lies about it for the same reasons as they lied in 2003; it's probably deliberate, not by accident, and anyone who falls for it again is not misguided but a sadistic warmonger."
You could say it but it's obviously wrong. They are not the same "lies" by the same people or for the same reasons. Moreover, Iran clearly was "building nukes"- i.e. taking steps to make it possible to build nuclear weapons very quickly, that had no possible peaceful purpose.
Having nukes is a fairly good guarantee against being attacked. Its the main reason any country has them. Since trump killed the JCPOA, it became obvious to Iran that guarantees made by the US were not to be trusted. This is all a fairly obvious outcome of escalatory brinkmanship.
Very much agreed, especially with that last part. Between Russia invading Ukraine and this action against Iran, if I were in the leadership of a non-nuclear country right now, I would be looking into every possible realistic avenue of getting nukes. It's sure looking like "might makes right" is the guiding principle for most of the world's leading powers today, and I wouldn't trust anything they said to the contrary.
Which is, ultimately, very sad. While still far from perfect, humanity seemed to have made pretty remarkable progress in the past few decades of leaving behind the violence of our past. Now we seem to be backsliding.
I do not really understand what you say. Principles are for intellectuals. The reality on the street is, people do not kill their neighbors inside countries, because there are courts enforcing laws. Now international relations are always anarchy, there is international law and the international criminal court, but crucially it can only be enforced by war. So internationally, courts cannot enforce law without a war first. Milosevic had to lose a war first, then when to the ICC to be charged by the violation of international law.
So no, might does not make, but enforce right. Iran violated international law, one of the most important ones (non-proliferation), and it can only be enforced this way.
" The reality on the street is, people do not kill their neighbors inside countries, because there are courts enforcing laws."
But that isn't the reality on the street. That has never, ever, in human history been the reality on the street. That "reality" is a story made up by cynics to justify their cynicism or by monsters to assure themselves that everybody else is secretly as broken inside as they are.
Here in actual reality, the vast, vast majority of people would not kill their neighbors even if they were absolutely sure they would never be caught or punished. There is a small minority that would, and a larger minority (but still very much a minority) that *might* if circumstances got out of hand. But it simply doesn't take very many willing murderers to ruin things for everyone else, so the world with laws and courts ends up much better than the world without.
Regardless, none of that is much relevant to international relations at all. Countries are not people. They do not have the same decision processes as people, the same constraints, vulnerabilities and limitations as people,. nor the same virtues and strengths as people. They are very different things, and the analogy to civic law breaks down very quickly.
Lots of international agreements work just fine without any sort of violent enforcement mechanism. They work because cooperating and playing nice often benefits all parties involved. Even when a country encounters a specific case where they would benefit by pushing the "defect" button, the short-term payoff is rarely worth the loss of future opportunities that accompanies that hit to your reputation. A lot of leaders and people in high government positions understand that very well.
Unfortunately there are several prominent leaders who do *not* seem to understand this, or who rationalize away that understanding whenever they see a defect button that they *really* want to push. In the longer term their countries will do worse and lose power and influence against the countries who consistently cooperate. (If human civilization overall survives and flourishes into the longer term.)
Now the Iranian regime is a pretty terrible government, and they themselves are not so great at avoiding the "defect" button. But in large part that's because the rest of the world--and the U.S. VERY much in particular--has shaped them that way. Consider that the current regime took power after Iranians had spent nearly two decades getting systematically fucked over by an autocratic puppet government supported by the U.S., all because Iranian oil was more important to Uncle Sam than Iranian lives or Iranian civil rights. Consider that the last time Iran made an agreement with the U.S. regarding its nuclear program--a mutually beneficial agreement--that agreement was torn up *by the U.S.* within a few years of its signing, for no good reason whatsoever, followed not too longer after by the U.S. assassinating one of their top generals on the flimsiest of justifications. From Iran's perspective "international law" no doubt looks like a joke of a concept: no protection against foreign abuses, but often cited to justify them. [1]
So yes, principles are for intellectuals. And intellectuals have spent the past few centuries taking over the world: very slowly at first, but much faster in recent decades. I'll admit that the anti-intellectuals have made a very, very impressive showing this year. A really awe-inspiring bid to drag us all back down into the, to stage a grand comeback tour for Pestilence and Famine and War[2], to really redecorate the place with that retro Dark Ages look. But they haven't succeeded yet. There are still lots of us out here who *do* have principles and are willing to stand up for them.
Relevant further reading if you made it through that wall o' text and can still manage more.
Regarding Cooperation in Modern International Relations Specifically:
https://acoup.blog/2023/07/07/collections-the-status-quo-coalition/
Regarding Cooperation in All Human Affairs More Generally :
https://slatestarcodex.com/2014/02/23/in-favor-of-niceness-community-and-civilization/
[1] If one wants a better understanding of that part of the world, consider that for any reason an Iranian might feel anger at the international community and certain countries in particular, a Palestinian has two or three such reasons. As a people they've been on the receiving end of an absolutely staggering string of abuses and insults, going back more than a century. (And yes, obligatory disclaimer, they've been on the giving end of some as well.) And plenty of non-Palestinian Arabs *can see that* and don't necessarily have a lot of reason to expect that Britain or Israel or the U.S. would treat them any better if they were in a similar position.
[2] Death being a timeless classic that never goes out of style.
>Here in actual reality, the vast, vast majority of people would not kill their neighbors even if they were absolutely sure they would never be caught or punished.
Okay, I sort of over-exaggerated this. Let's say the vast majority would commit financial fraud or take bribes. Culture matters. As a textbook European I don't really get why Americans are so brimming with holiness... I think it might be a Calvinist thing, I think Catholic cultures more accept that everybody is a sinner and it is basically okay.
>Lots of international agreements work just fine without any sort of violent enforcement mechanism.They work because cooperating and playing nice often benefits all parties involved.
This is true for private individuals and corporations too, yet we need law enforcement. In fact because of the principal-agent problem, individuals and corporations ought to need less enforcement, as national leaders have way way bigger principal-agent problems. A marketplace is always better from a principal-agent viewpoint than an election. Like if term limits mean Trump cannot be re-elected, what exactly does he have to lose?
>Even when a country encounters a specific case where they would benefit by pushing the "defect" button, the short-term payoff is rarely worth the loss of future opportunities that accompanies that hit to your reputation.
See above. It is not a country, but the leader pushing the button. Hitler pushed the biggest "defect" button ever, yet a few years after the war, most everybody happily cooperated with democratic West Germany. Just 6 years later, the European Coal and Steel Community was signed - imagine France trusting Germany so much to pool their coal and steel i.e. military industry resources!
>because Iranian oil was more important to Uncle Sam
Actually it was British Petrol.
>And intellectuals have spent the past few centuries taking over the world: very slowly at first, but much faster in recent decades.
I wish it was true. The reality is, the bureaucrats did. I mean an intellectual is someone like Chomsky. While these bureaucrats are as unprincipled and aggerssive as ever. Look at what happened in Raqqa. A war of annihiliation, 2017 https://www.amnesty.org/en/latest/news/2018/06/syria-raqqa-in-ruins-and-civilians-devastated-after-us-led-war-of-annihilation/
And these are the intellectuals with principles?
Israel is supposedly nuclear and still got bombed, by Iran, on October 2024.
But Iran got bombed because they were trying to go nuclear. Shouldn't countries draw the exact opposite conclusion- that if they try to get nukes, they will be destroyed? Surely if Ukraine had tried to get nukes, Russia would have destroyed them at that time anyway, right? And probably if Russia invaded earlier, they would win easily.
Agreed! ( Particularly if, while enriching uranium to 60% U-235, they spend 40 years chanting "Death to America!" )
Iran *got bombed* because it was suspected of trying to go nuclear. Iran was not *destroyed* over this. The Iranian regime was not destroyed over this. The Supreme Leader was not killed or imprisoned or even deposed over this.
See also North Korea. And Pakistan, which was pretty clearly in bed with Al Qaeda and Osama Bin Laden.
The Iraqi regime, which we were pretty certain didn't have nuclear weapons and wasn't going to be making nuclear weapons any time soon, that *was* destroyed, and Saddam Hussein was dragged out of his spider hole and hung by the neck until dead.
The Libyan regime, which we knew for a fact didn't have and wasn't going to be making nuclear weapons, which had done the "right thing" by thoroughly and verifiably dismantling its nuclear arms program, was destroyed, and Muammar Gaddafi was anally raped to death on Youtube. Libya has been a failed state ever since.
You really think the conclusion Evil Dictators(tm) are drawing is that they should *not* nuke up ASAP?
I suspect the only reason North Korea was allowed to get nukes without getting couped or having their nuclear program raided/bombed is because they have a lot of artillery aimed at Seoul.
The primary takeaway any Evil Dictator should have from the last 40 years is that power speaks, and the West is super consequentialist about things, so make your threats as credible and devastating as possible. Nuke up, artillery up, etc.
The threat of artillery bombardment of Seoul is vastly overrated. North Korea has bignum guns, a small fraction of which have the range to reach the northern suburbs of Seoul and none of which have the range to reach downtown Seoul. This gets misreported as "North Korea has bignum guns and they can bombard Seoul into oblivion!". They now have heavy artillery rockets that could do so, but that's a recent development. Twenty years ago, it was only a modest number of Scuds and the like, roughly what Tehran and Baghdad were inflicting on each other in the 1980s.
North Korea was "allowed" to get nukes because everyone else was stuck in the Buck Turgidson "ignorant commie peons don't understand technology like our boys do" school of threat assessment. I met far too many people, far too highly placed, who were assuming that North Korea couldn't *really* build nuclear weapons five years after their first fully successful test. And since "everybody knew" North Korea can't really build nuclear weapons, we didn't really feel the need to do anything. Then it was too late.
That mistake, at least, we aren't repeating with Iran.
Interesting. But you think this really won't be enough to slow Iran significantly, right? I think I saw elsewhere you were predicting nukes within 12 months. Are centrifuges that easy to quickly obtain and deploy? Or are you just figuring that we didn't eliminate enough of their current stockpiles and equipment?
"But Iran got bombed because they were trying to go nuclear."
Or because they were behind Hamas, and, you know, Oct/7.
Or because they shoot at Israel last year (twice).
Or because they are an evil theocracy.
When the official explanation is changing so frequently, one cannot help but have the feeling that none of them is true.
Some decisions are overdetermined. Multiple independently sufficient motives that all are based in actual facts don't undermine each other's legitimacy. Israel has many, many reasons for bombing Iran, most of which boil down to "they hate us and have been waging a (mostly, but not entirely proxy) war against us for decades."
I think the question is not why Israel hates Iran (or, when we are at it, vice versa), but that what is the concrete reason for the actual ("pre-emptive") attack. And I don't think it is irrelevant, for several reasons. First, a rational expectation is that they end the war when their goals are met. If the war is about nuclear warheads, they don't start bombing police stations; if the war is about destroying the Theocracy of Evil, that is an entirely different matter.
And I don't think "we hate their guts, we are mortal enemies for like forever, that's why" would have met with an applause on the international stage.
But communicating one thing first, then another second, then a third later gives me the impression that they are throwing whatever plausible explanation crossed Netanjahu's mind to see what sticks.
(Heck, they could even give a list why, but AFAIK this is not what happened.)
Of course, the fact that Israel's current government lost a lot of its credibility in the last ~20 months does not help either. Overall, their communication is aimed to cover up whatever they can get away with. Which is not worse than the median in the region, not at all -- but enough to raise doubts about the official communication.
I think that's a kind of a silly standard to hold Israel to. Iran and Israel have been at war for decades now, and the only reason Israeli and Iranian troops haven't directly fought much is that there are neutral countries in between them. Instead, Iran has used proxies (and as recently as last year, its own military) to bombard Israeli cities with thousands of imprecise terror weapons. There's been no ceasefire since those attacks. No armistice. Not even a softening of relations in any way.
In other words, I don't think this is reasonably framed as a pre-emptive attack at all. The correct frame is that they were already at war. There's no other name for a state of affairs where one country is launching thousands of missiles at another. They're at war, have been at war, and throughout the course of that war Iran has mostly been the aggressor. (Admittedly, that's probably because Iran has a lot more ability to cultivate proxies than Israel does, not some moral superiority on Israel's part.)
And when you're in a long-standing war, the reason to strike back at enemy military targets is: A) you can, and B) the juice is worth the squeeze. All of Netanyahu's statements make a lot more sense when you realize he's explaining specific war tactics, not trying to justify starting a war.
> When the official explanation is changing so frequently, one cannot help but have the feeling that none of them is true.
Or because those things are related.
(You could make a similar argument about WW2: "First they said it was about Czechoslovakia, then they said it was about Poland, then they said it was about France, now we have this story about concentration camps... the constant change of the narrative makes me suspect that the so-called Nazis are just a boogeyman invented by American media.")
You are right, let me make it more clear. When I wrote, "none of them is true", I meant it in the sense that none of them is the real reason of the attack, not in the sense that they do not hold some truth. Of course they are mostly true -- but that they are the reason for the attack seems bogus after such swift changes.
"They are related" seems a fuzzy concept to me, though; I mean, of course they are related in the sense that all has something to do with Iran, and all are things which sound quite bad for the international public. And this is exactly why they can be ideal BS PR for such a bombing campaign.
Which, again, do not make them untrue; they just might not be the real reason for bombing. (Also, you can't stop Iran to be an "evil theocracy" by a bombing campaign.)
Considering your examples, would the allies state they go to war with Nazi Germany because of the concentration camps, that would be outright BS as well. It does not mean, though, that concentration camps existed, or that the war was the reason these stopped operating, or that it is good that they stopped operating.
"But Iran got bombed because they were trying to go nuclear."
This has by no means been adequately demonstrated. They didn't dismantle their nuclear program entirely. And I'd say they were obviously interested in keeping the option of building up to weapons capability in a medium-short, rather than long time horizon open[1]. But the claim that they were actively trying to build a bomb at the time they were attacked is one that needs significant support that I haven't seen yet.
[1] Which to my understanding is a state that many non-nuclear developed countries are perpetually in, simply by virtue of having the right mix of industry and technology and scientists. It's actually a little bit scary how *simple* nuclear weapons are when you get right down to it.
>But the claim that they were actively trying to build a bomb at the time they were attacked is one that needs significant support that I haven't seen yet.
The IAEA looks convinced that Iran prevented the IAEA from ensuring that Iran was not building a bomb.
https://www.reuters.com/world/china/iaea-board-declares-iran-breach-non-proliferation-duties-diplomats-say-2025-06-12/
And they enriched uranium to 60% U-235, while power reactors only need 5%. The _point_ of making such highly enriched uranium is a nuclear _weapons_ program.
I sent a thank you note to Trump after he authorized the bombing of the Fordow uranium enrichment site. I consider bombing it to be the right choice.
None of that particularly contradicts what I had just written. Avoiding international inspection and enriching uranium to 60% are of course both consistent with a country actively working towards a bomb. But they're also consistent with a country that would like to keep the option of getting there quickly, but without larger the provocation of actually assembling the weapons.[1] I don't especially have an opinion either way: it's plausible that they were trying to get to a bomb, it's still far from certain. As a general rule, you shouldn't trust motivated claims that come from unreliable sources even when they align with an other-wise plausible reality. Good lies have grains of truth. But just because someone is a habitual liar doesn't mean they're never truthful.
[1] I'll note that if this is the case, they either miscalculated how much of a provocation their intermediate efforts actually were, or simply underestimated how ready Netanyahu and Trump would be to jump on an excuse for belligerence.
Many Thanks!
Ok, I think we are just splitting the difference between "steps directed towards making a bomb" (including enriching uranium to 60% U-235) versus "final assembly of a completed weapon" (or maybe even "arming a nuclear bomb" ?).
>it's plausible that they were trying to get to a bomb, it's still far from certain. As a general rule, you shouldn't trust motivated claims that come from unreliable sources even when they align with an other-wise plausible reality.
Look, Iran hasn't denied that they have 60% enriched uranium. That, _alone_, makes it clear that they are/were trying to get a bomb. 60% U-235 certainly isn't intended for power reactors!
Regrettably, as https://www.iranwatch.org/our-publications/articles-reports/irans-nuclear-timetable-weapon-potential says
>It is also important to consider that Iran could use its stockpile of 60% enriched uranium to make weapons directly without the need for further enrichment. There would be limitations in the delivery of such weapons, as discussed below.
( Basically, they could build a gun-type bomb with their existing 60% U-235 with no further enrichment, albeit it would be heavier than a 90% U-235 weapon. )
Sure, but its worth noting that Iran was much farther down the path toward weapons capabilities than any other non-nuclear-weapons nation. See the IAEA report that Iran has 900 lbs of uranium enriched to 60+%: https://www.aljazeera.com/news/2025/5/31/iran-increases-stockpile-of-enriched-uranium-by-50-percent-iaea-says
That's the largest stockpile of >nuclear power grade uranium in the world, aside from nations that already have nuclear weapons. And according to the IAEA, it's increased by 50% in the last few months.
I can't see an explanation for this that isn't attempting to either A) get nuclear weapons or B) make getting a nuclear weapon something they can achieve in the very short term.
I agree with the last paragraph, but I think A) and B) are pretty importantly different here. In particular, I think a world in which we were confident the answer was B) and not A) would be a world in which the bombing would be *exactly the wrong move.*
Re: footnote one, any country with a domestic nuclear power program (not just a reactor, but producing their own fuel) could stand up nuclear weapons within 2-3 years. The main players there (aside from the countries that already have nuclear weapons) are Germany, Japan, Canada, Sweden and South Korea.
Thermonuclear weapons (ie H-bombs) would take longer, but that's more because designing and building an H-bomb is notoriously difficult - an implosion-type A-bomb is a pretty straightforward engineering problem once you have the enriched uranium and you know how to machine it (both of which you have as a result of the domestic power program).
There's something involving "lensing" in making an H-bomb which I have deliberately avoided finding out about because I would like to be able to travel the world without constraint, but it's famously a hard problem.
"There's something involving "lensing" in making an H-bomb which I have deliberately avoided finding out about because I would like to be able to travel the world without constraint, but it's famously a hard problem."
Ah yes. It happened exactly once in my education that a professor unwilling/unable to fully answer one of my questions because "some of the details are classified." The part of the answer that he did give was pretty much "something involving lensing."
The details of the Teller-Ulam design (for an H-bomb) are “born secret”, ie they’re secret under US law even if you reinvent them yourself.
Some of them were published in a magazine called “The Progressive” in 1979, but how accurate or complete they are is unknown.
It was notoriously difficult, because both the USSR and UK programs had partial information from the US (in one case by spying, in the other given voluntarily) and still struggled to get it right, and the other two open programs (France and China) also took a long time.
I think you're mixing up plutonium and uranium A-bombs. Uranium A-bombs don't even require implosion; just two chunks of uranium you slam together. If you can make bombs, and you have a critical mass of highly enriched uranium, they're child's play.
Plutonium on the other hand has pretty low neutron production, so you need implosion explosives to trigger them. That's apparently a difficult problem, at least to the point that it made North Korea have multiple fizzles.
Jeffrey beat me to the punch, but my understanding was that the implosion was to ensure that the plutonium had time to react fully before the explosion spread it too far apart.
I believe that similar considerations are what make H-Bombs difficult: simply packing deuterium around a fissile core will give you extremely high temperatures and pressures, but with such an abysmal confinement times that your triple product remains quite low.
>Plutonium on the other hand has pretty low neutron production,
Huh?
>so you need implosion explosives to trigger them.
Umm... AFAIK, the problem is that the Pu-240 impurity in a Pu-239 pit has a _high_ spontaneous fission rate, so it produces a _high_ neutron background rate, which can trigger a premature chain reaction when the pit is only partway to being fully imploded, and that gives you a fizzle. That's much of the motivation for using high explosives to make the implosion fast, and reducing the chance of a fizzle (you also want to compress the plutonium as much as possible, to get a given mass of plutonium plus reflector plus tamper as supercritical as possible to get the yield up)
Saddam completely gave up his nuclear program after 1991, and they punished him for it in 2003.
Gaddafi gave up his nuclear program in 2003, and they punished him for it in 2011.
(Plus all the other coups and massacres and wars of criminal aggression that had nothing to do with nukes. E.g. overthrowing Iran's secular democratically-elected PM to install a dictator to protect oil profits in 1953; btw that, plus his overthrow in 1979, is why Iran is a theocracy now.)
North Korea has nukes, so the US only sanctions it instead of invading or bombing.
The US has, through its actions, tried hard to prove that no one is safe from US aggression without a nuclear deterrent. This is very foolish.
> overthrowing Iran's secular democratically-elected PM to install a dictator to protect oil profits in 1953;
Of all the inaccuracies in the Standard Internet Commenter View Of History, the idea that Iran was meaningfully democratic prior to 1953 is one of the most pernicious.
The reality was... very complicated. There were elections but they were regularly rigged, politicians had a tendency to get assassinated if they didn't agree with the Shia fundamentalists, and the 1952 election ended with only 79 out of 136 seats filled because Mosaddegh did a literal "Stop the count!" move once a parliamentary quorum was reached and before the votes from rural seats which would have opposed him could be tabulated.
The post-1953 system also had rigged elections, though for a while it seemed to be heading in the right direction with multiple competitive parties (albeit all Shah-approved) then in 1975 they just said "fuck it, actually let's just have one party and outlaw all the others".
I think 1953 is best seen not as a transition from democracy to non-democracy but just another speedbump in the shitshow of rigged elections, assassinations, coups, revolutions, party-bannings, and neverending conflict between royalists, Islamists, communists, landlords and the military that characterises the last century of Iranian politics.
Right. The world is full of crappy dictators who are allowed to stay in power because they don't cause any particular trouble.
While Ayatollah Khameni is hiding out in a bunker, the Emir of Dubai is banging bikini models on his yacht, because making friends is better than making enemies.
Yup. After WWII there was a chance for multilateralism and international law and non-proliferation—but the US preferred hegemony to survival.
sure the Soviets were all for that
I think you are right on this.
On the other hand, it is also obvious that the US made sure to have the overseas empire of the UK collapsed. (It was happy to see the collapse of the French one as well, but actively contributed to the former.)
Correct. US and SU were competing for the colonies, but agreed that they will have them, not the UK and FR. So basically textbook great power politics.
Hegemony is survival. The best way to guarantee survival is to remove the tools others have to hurt you.
Foolishness. No, spending decades trying to dominate and crush and dominate and crush is not a good way to survive.
There are only two options. You are either friendly and cooperative, or totally crush them. What does NOT work is some intermediate level of violence, and that is IMHO the mistake the US and Israel tends to make. That is, use enough violence that they hate you, but not enough that they fear you.
You seem to be skipping over the first 70 post-WWII years.
1953 - overthrew Iran's secular democratically-elected PM, to install a dictator to protect oil profits
(forcing on Iran a US puppet dictator, to be overthrown by a homegrown Islamic Revolution in 1979, is why Iran is now a theocracy. It was so dumb)
The quest for domination did start quite immediately, I'm afraid
> 1953 - overthrew Iran's secular democratically-elected PM, to install a dictator to protect oil profits
https://open.substack.com/pub/astralcodexten/p/open-thread-387?r=izqzp&utm_campaign=comment-list-share-cta&utm_medium=web&comments=true&commentId=128821580
Uh...really? "The US preferred hegemony to survival"? I mean, it's not like there were any *other* countries that were pursuing hegemony after WWII...
I honestly wonder if you realise that you have, with a comment like this, absolutely ensured that 90% of readers, many of which may have been highly open to your viewpoint, will now permanently disregard everything you say on the matter. More generally, I'm just amazed at how many people trying to advocate for a position have an unbelievable talent for appending a single sentence that compresses "I have a complete lack of perspective", "I have no awareness of the slightest factual nuance" and "I'm a fanatical ideologue" into that sentence.
>"Uh...really?"
Yes really. (You know about all the coups, all the overthrowing democracies to install dictators, all the terrorists and massacres, all the undermining of international law, right? In the nuclear age this isn't just wicked, it's an existential risk.)
>"It's not like there were any *other* countries"
Did I say there weren't? But if you're American, you should understand what's going on so you can be an effective citizen. (And it is the only superpower so it has an enormous impact.)
>"will now permanently disregard"
I hope they prefer survival, over encouraging nuclear proliferation so they can pwn david bahry in the comments section of astral codex ten dot com.
You're the one who said that after WWII there could have been peace and stability, if not for the US. The only sensible way to parse your comment is as saying that the US on its own ruled out peace.
If you're now trying to claim that you only meant the US played *a* part, that's a pretty blatant motte-and-bailey. Otherwise, you would have to say that a comment saying "there was a chance for world peace, but Iran chose jihadism instead" would get total unchallenged agreement from you.
The US played an enormous and consistent role, which it is still playing, then. If you want to compare the amount it did vs what the USSR did then feel free (I'm historically curious, though it's little relevant to talking to Americans on an American blog about what America is doing).
Re: the first, if you didn't know, start with
- 1953 - Iran - overthrew the democratically elected PM, replacing him with a dictator to protect oil profits
- 1954 - Guatemala - overthrew the democratically elected President, replacing him with a dictator to protect fruit profits
- 1965/6 - Indonesia - helped the incoming dictator Suharto massacre a million peaceful leftists
Etc
The general pattern never stopped. For some books check out Vincent Bevins' "The Jakarta Method" (reviewed at: https://www.theamericanconservative.com/the-jakarta-method-how-the-u-s-used-mass-murder-to-beat-communism/) or Chomsky and Robinson's "The Myth of American Idealism" (reviewed at: https://foreignpolicy.com/2024/11/15/chomsky-foreign-policy-book-review-american-idealism/).
> - 1953 - Iran - overthrew the democratically elected PM, replacing him with a dictator to protect oil profits
According to Melvin's post upthread, this is an extremely misleading description of events.
Again, you're either saying that the US played *a* role among others in destabilising the post-WWII world, or you're saying it played the primary role. If the latter, I suggest you look up the Yalta pledge. If the former, this is an almost textbook motte-and-bailey discussion.
I strongly agree with this post
some science + money gossip:
"Colossal Biosciences" is the company that pretends to de-extinct dire wolves and so on. It's really just living paleoart—fun, but not remotely what their hype claims. E.g. the "dire wolves" were really just grey wolves with a handful of genes edited to make them big and white etc.[1] For fur colour, they couldn't even use the actual dire-wolf gene, b̶e̶c̶a̶u̶s̶e̶ ̶t̶h̶e̶ ̶g̶r̶e̶y̶ ̶w̶o̶l̶v̶e̶s̶ ̶a̶r̶e̶ ̶s̶o̶ ̶d̶i̶f̶f̶e̶r̶e̶n̶t̶ ̶f̶r̶o̶m̶ ̶d̶i̶r̶e̶ ̶w̶o̶l̶v̶e̶s̶ ̶t̶h̶a̶t̶ ̶t̶h̶e̶ ̶d̶i̶r̶e̶ ̶w̶o̶l̶f̶ ̶g̶e̶n̶e̶ ̶w̶o̶u̶l̶d̶ ̶h̶a̶v̶e̶ ̶m̶a̶d̶e̶ ̶t̶h̶e̶ ̶g̶r̶e̶y̶ ̶w̶o̶l̶v̶e̶s̶ ̶d̶e̶a̶f̶ ̶a̶n̶d̶ ̶b̶l̶i̶n̶d̶.*[2] The misrepresentation of what they're doing isn't just cringe—it's dangerous, given how Trump wants to use it as an excuse to try to remove protections from endangered species.[3]
My supervisor Dr. Vincent Lynch, apart from his research e.g. on cancer resistance in elephants [4] and other cool stuff, is a critic of Colossal's hype (while admiring some of the genuinely important biotech that Colossal is refining, which could be redirected to other genuinely useful purposes).[5]
The gossip part: it looks like Colossal's finally started using intimidation tactics. First, some bizarre hit piece articles appeared trying to discredit his critiques (but his critiques are standard—Colossal itself admits that calling gene-edited grey wolves "dire wolves" is just "colloquial"). Second: when he suggested that the articles are AI-generated and might be paid for by Colossal and its billionaire cofounder (obviously plausible given the incentives + clearly disclaimed as opinion i.e. protected speech + clearly related to public concern given Trump's intent to remove protections), Colossal had their lawyers send an intimidating "shut up about it" letter to his front door.
Ironically, Colossal's behaviour is more humiliating for them than scientists' critiques ever could be. The hype train was rolling and not going to stop; people want to believe that we have dire wolves and will have mammoths. But people don't like billionaire Elizabeth Holmes-ass behaviour to silence critics.
[1] https://www.newscientist.com/article/2481409-colossal-scientist-now-admits-they-havent-really-made-dire-wolves/
[2] https://www.businesswire.com/news/home/20250407444322/en/Colossal-Announces-Worlds-First-De-Extinction-Birth-of-Dire-Wolves
[3] https://www.washingtonpost.com/climate-environment/2025/04/10/trump-endangered-species-protections-dire-wolves/
[4] https://elifesciences.org/articles/11994
[5] https://www.livescience.com/animals/extinct-species/colossals-de-extinction-campaign-is-built-on-a-semantic-house-of-cards-with-shoddy-foundations-and-the-consequences-are-dire-opinion
*Added: this part was probably my misinterpretation, inherited from some early reporting. More likely the issue was that these genes are risky to mess with in general (e.g. in the event of imperfect CRISPR edits?), rather than a specific incompatibility with the dire wolf version. Thanks to Catmint for pointing this out. https://www.astralcodexten.com/p/open-thread-387/comment/129267704
This is just the very first thing they've done. Of course it's not reaching all goals.
Alexander Graham Bell's phone was *dogshit* compared to iPhone 16, but fortunately people kept improving the technology.
> For fur colour, they couldn't even use the actual dire-wolf gene, because the grey wolves are so different from dire wolves that the dire wolf gene would have made the grey wolves deaf and blind.[2]
This is as far as I can tell a miscommunication between scientists and laymen. Actual quote is:
> The dire wolf genome has protein-coding substitutions in three essential pigmentation genes: OCA2, SLC45A2, and MITF, which directly impact the function and development of melanocytes. While these variants would have led to a light coat in dire wolves, variation in these genes in gray wolves can lead to deafness and blindness.
For context, many species have minor variants of these genes which affect coat color (usually lightening it). For example, a variant of OCA2 causes blue eyes in humans, while a variant of SLC45A2 is responsible for palomino horses. White spotting patterns in dogs are associated with MITF variants.
However, in mammals, maybe vertebrates or higher, major disruptions can cause negative side effects. For example, complete deletion of MITF is lethal, while major disruptions result in an all-white coat and often deafness. Certain mutations of it in mice are also capable of reducing eye size, which probably doesn't help with vision.
OCA2 and SLC45A2 are much less serious. An OCA2 variant is the most common cause of blue eyes in humans. Blue eyes are known to sometimes cause minor difficulties with vision especially involving bright lights, however the effects are small as you can find out by asking any blue-eyed human. The causal mechanism is differently reflected light by the iris, so this is not gene-specific. SLC45A2 and can also cause blue eyes. I do not know of any cases where full blindness was caused in any species by any mutation to either of these genes. MITF too can cause blue eyes, but this is not the concerning thing about it.
I do not know of any studies of OCA2 or SLC45A2 showing a causative link to blindness. Closest I found was "There was no statistically significant difference in vision between patients with OCA1 versus OCA2, or between patients with and without mutations. Patients with two mutations tended to have worse vision than those with one, but not statistically significantly." from https://www.sciencedirect.com/science/article/abs/pii/S1091853116301823 (do not be mislead by the word "blindness" in the introduction - it may be referring to a different gene)
For more on MITF mutation viability in canines:
> In humans, deleterious MITF mutations cause disorders of vision and hearing, including the Waardenburg and Tietz syndromes [9], [12], [13]. Deafness has also been recorded in white dogs, where approximately 2% of white dogs (sw/sw) present with bilateral deafness and 18% are unilaterally deaf [14]. The majority of mutations reported in mice and humans that cause severe pleiotropic effects are generally loss-of-function mutations affecting the coding regions [15]. This is not the case with dog MITF alleles. A comparison of S and sw haplotypes, using BACs from an S/sw heterozygote, across the 100 kb canine white spotting candidate region revealed 124 sequence polymorphisms, all of which were located in non-coding regions [6]. This demonstrated that the extreme white coat colour phenotype is controlled by one or several regulatory mutations. This hypothesis is strongly supported by the fact that coloured patches on white spotted dogs display normal pigmentation. Thus, this suggests that the canine MITF variants primarily affects migration and survival of melanocytes during development, but have no or only minor effects in mature melanocytes in the hair follicle; pigmentation of the hair requires MITF protein expression [16].
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0104363
That paper also has a table of MITF variants found in wolves. So, this isn't a deal where gray wolves have something special going on that makes MITF variants uniquely harmful, but rather that major disruptions to the gene _can_ in some species have harmful effects, and (here I enter the realm of speculation, as I know the theory but not the practice) perhaps they were concerned that CRISPR would not modify the gene accurately enough to include just the desired substitution without damaging the rest.
I think you're right. Although e.g. Zimmer's NYT article said "they did not introduce the remaining five [mutations], because previous studies had shown that those five mutations cause deafness and blindness in gray wolves"[1]—which is might be where I got that interpretation from—the press release itself only said "variation in these genes in gray wolves can lead to deafness and blindness."[2] So yes, I now agree it's probably more that they considered these genes risky to mess around with in general, rather than a specific incompatibility with the dire wolf version.
[1] https://www.nytimes.com/2025/04/07/science/colossal-dire-wolf-deextinction.html
[2] https://www.businesswire.com/news/home/20250407444322/en/Colossal-Announces-Worlds-First-De-Extinction-Birth-of-Dire-Wolves
[added: I've now edited the OP]
Hm, will look more into this later.
Sure, I love talking about animal color genetics so hit me up with any questions :)
Ooo, here are some! Pick your favourite:
- What's your favourite weird colour pattern mechanism?
- Are zebras white with black stripes or black with white stripes?
- Why do some kittens have mittens 🥺
Leopard spotted horses, which aside from being cool molecularly are also a very old pattern which may have helped with camouflage during the ice age. The first part is a retroviral insertion disrupting the TRPM3 gene. By itself this causes a normal-colored foal except with striped hooves, which as it grows gradually gets more and more white hairs throughout the coat. But combined with a certain variant of another gene, the whole horse can be white with dark dots. Nobody has actually proven the mechanism, but the setup and the effect are similar to spotted corn, so it probably also works by transposon. If that's so, then the second gene is a special activator for the retroviral mechanism, allowing it to copy itself out of the DNA and into other parts. So maybe in some cells the transposons copy themselves right out of the DNA and restore the gene's function (the colored dots), while in others they copy themselves further or otherwise mess up that section of DNA even more (the white regions). Speculation, but the spotted corn definitely does it that way and horses are cooler than corn.
Zebras are brown with black and white stripes: https://arthistoryanimalia.com/2023/01/31/animal-art-of-the-day-for-international-zebra-day-1-george-edwards-george-stubbs-and-the-first-british-zebras/ Like this horse: http://www.skhantoniow.pl/waldi.htm then you take the brown areas and you turn them dusty-tan like this: https://upload.wikimedia.org/wikipedia/commons/c/cd/Cheval_fjord_00003.jpg while you also take those stripes and extend them to cover the whole body. Legs too: https://upload.wikimedia.org/wikipedia/commons/e/e5/Somali_Wild_Ass.JPG
Kittens have mittens because it's very cute and it even rhymes! What's going on there is not white being added but color failing to be added because the cells for it don't grow in that part of the skin. For color you need melanocytes, and melanocytes develop from melanoblasts, which start at the spinal region and during early embryonic growth spread across the skin of the whole animal. Except if the genes are not working right, it doesn't do that, and instead the cells only reach some areas. Because the paws are far away from the spine, they are one of the last places where the pigment cell precursors arrive, so if things get slowed down it's easy for the paws to miss out on getting color. (And this starts before the tail has grown out, so the tail tip also counts as far away from the starting location. And the path on the head is around the sides meeting in the middle, so if it doesn't quite make it you get a little pink nose with white strip and then you have to really hope the cat will let you kiss the nose.)
Quality reply! Thank you.
This would read much better without the TDS, and especially [5] which is just more TDS and boo outgroup.
If you are going to write about scientific topics, stick to scientific arguments. Science + politics = politics.
I'm sorry you were triggered by hearing relevant info about what the administration is, in fact, trying to do and for which the administration is, in fact, citing Colossal's claims as justification.
I'm not sorry you are triggered that your unscientific horse manure was called unscientific horse manure.
There is a very interesting discussion to be had regarding the scientific merits of endangered species protections and the unintended consequences of such actions (protections seldom work as intended and often inhibit more productive solutions). I enjoy those discussions.
Sneering leftist drivel composed solely of reducto ad Trump has never once been productive and actively makes the world a worse place. Do better.
If you think it would be "horse manure" for the Trump administration to want to remove protections for endangered species, and to cite fake de-extinction as a justification—
—tell the Trump administration. https://www.washingtonpost.com/climate-environment/2025/04/10/trump-endangered-species-protections-dire-wolves/
Okay, why are you assuming that the Trump administration knows this is fake and is cynically removing endangered species (I presume you are hinting at 'so rich developers can bulldoze habitats and make YUGE profits which they funnel back to GOP politicians!' rationale for that).
All the flippin' "trust the science, bro" messaging from the liberal/left side about many issues, so if a company comes along with "look, we are using SCIENCE to do this thing", why not believe them?
I think we shouldn't believe Colossal, as it stands, but if you're just skimming something then "hey, extinct species can be brought back, that's great!" does translate into "and so the at-risk species today are no longer so at-risk since we can clone them and reintroduce them as necessary". That does not require hand-rubbing glee at "now we can kill off all those pesky animals and make megabucks" but rather believing the hype around science and technological advancement.
I could easily see Trump, for one, being delighted by this as a sign of how great and how advanced America is, here's an American company that can bring back extinct species from the past, we're the greatest! (Personally I blame Michael Crichton and Jurassic Park for putting it out into pop culture that we can do stuff like this).
Expecting this particular administration to not take a big public dump on your particular hobbyhorse is hilarious, by the way. They took away pennies! (Yes, that's a deliberate strike against the coin collectors). I don't mind you yelling about the Trump Administration being big meanies, but you're mistaking distractions for "the Administration really cares about this."
Oh look, you are trying to pass off more political opinion as some sort of scientific discussion. Very much horse manure, I said do better, not worse.
Anyway, I've had enough of your lashing out at facts in order to protect your feelings about the administration. I will not respond after this. Bye.
Good, perhaps a productive discussion can now occur.
Their writeup [2] describes solving a challenging and multifaceted problem, for which a reasonable definition of success is unclear. Obviously it's self serving, but it is a startup that needs to pull some funding by embellishing their project, and the claims seem reasonable and honest. So far so good.
So from [5] I was expecting a 10,000 words technical rebuttal, instead I got Trump Derangement Syndrome and evil mustache twirling capitalists. Is this guy really a biologist? I see nothing of substance that actually engages with Collosal's claims.
What a strange comment. He mentions Trump *once*—"The Trump administration has announced it will remove protections from many endangered species, citing Colossal's de-extinction of the dire wolf." I'm sorry you consider it a derangement syndrome to inform you of relevant facts.
(Also, as I already explained: they slightly edited a tiny handful of genes in a grey wolf, and pretended it was a dire wolf. That's what happened. I'm not sure what additional 10,000 words you're looking for to re-explain it.)
I don't care who cites their work, I care about the merit of their work.
The link in [2] explains the tradeoffs they had to navigate and their reasoning. The link in [5] does not engage with that at all and instead only implies ulterior motives, which is low key confirmation that their decisions were correct as even an expert critic cannot engage on merits.
I didn't say anything about "who cites their work."
As I explained, the issue is the dishonesty. If they want to make living paleoart that's their business. They shouldn't pretend it's de-extinction.
The merits of their work are poor. Dire wolves are an entirely separate genus, not species but genus, from gray wolves. They edited 14 genes towards the dire wolf sequence. This is not the amount of change you do to get to a different species, let alone genus. Even domestic dogs are all considered one subspecies, and they've got way more genetic variety than that! Plus if you read between the lines, and look closely at the choice of genes and the fact that GRR Martin is their board's cultural advisor, it becomes pretty clear that the goal was to make the wolves look as much like the game of thrones wolves as possible. Like, why did they specifically add non-dire-wolf genes known to cause white coat color? Not because dire wolves were thought to be white, previous research suggested they may have been kind of reddish, but because the Game of Thrones wolves were white.
Yeah, it's a hard problem, and yeah, the level of success they got is actually technologically impressive. But the level of hype and marketing BS around it is unacceptable and would not be tolerated in any company making products for consumers. This is like 'Our new phone will impress 500x more girls when they see you using it' levels of accuracy.
And when I first looked into this, I thought "Ok, the dire wolf stuff is poppycock, but at least they're still doing good work cloning endangered red wolves." Nope. The wolves they cloned weren't even endangered, they were red wolf-coyote hybrids that are doing fine. You could argue that conservationists should be more accepting of hybrids as worthy of conservation, but you have to actually make that argument, not just claim that the animals _are_ red wolves.
Yup. In Jurassic Park terms: this isn't making a dinosaur with a bit of frog DNA to fill in the gaps, this is making a frog with a bit of dinosaur DNA to fill in the gaps
""The Trump administration has announced it will remove protections from many endangered species, citing Colossal's de-extinction of the dire wolf."
I've tried to track this down, and all I'm getting at the moment is:
https://factcheck.afp.com/doc.afp.com.39WX79D
"Secretary of Interior Doug Burgum claimed on X that the "revival of the Dire Wolf" signaled "the advent of a thrilling new era of scientific wonder, showcasing how the concept of 'de-extinction' can serve as a bedrock for modern species conservation."
Burgum doubled down during a live-streamed town hall on April 9, 2025 saying: "If we're going to be in anguish about losing a species, now we have an opportunity to bring them back".
"Pick your favorite species and call up Colossal," he claimed, as the current federal government continues to target protections of various endangered species, often tied to misleading narratives."
So not an official government pronouncement, just social media hoo-hah. Unless you can point me to official government department website saying they are officially scrapping the endangered species list?
You know, I can't help feeling that if we had President Kamala and one of her cabinet officials making a Brat Summer announcement of "extinction can be reversed!", the same guys tut-tutting about this would be all "yay science!" even though it would be the same company and the same con artists making claims.
Burgum is a member of Trump's cabinet. This shows clear stated intent to remove protecting regulations, even if it's not an official order yet. https://x.com/SecretaryBurgum/status/1909345951069651032
Re: your second speculation, our objections to Colossal's dishonesty are older than the Trump administration plan. Burgum’s recent statement, only two months ago, is the first time any of this has actually had any connection to Trump at all.
(And obviously it wouldn't be acceptable for Kamala to do this either. It's weird how quick people are to assume common sense must be partisan political signaling.)
You, and/or the guy whose article you linked, were all "the Trump administration has definitely done this thing!"
Now you've rowed back to "okay they haven't done it yet but this guy's tweet is an indication that they're gonna do it".
Wake me up when they *have* done something, not when "this guy who hates my political enemies wrote a hyperventilating article about how they're gonna do all the bad stuff".
I think there's more than enough room to give Colossal a good kicking without dragging in "and Trump will do villain stuff because of them!!!!"
I said "wants." The article said "announced." And, in fact, the tweet announced what they want to do, exactly as we said.
I know it triggers you to hear any unpleasant information about Trump, but please deal with that on your own time.
re: "challenging and multifaceted problem, for which a reasonable definition of success is unclear"
If it might be possible to someday successfully clone a real dire wolf with an actual dire wolf genome—instead of slightly editing a grey wolf genome—cool. But that's not what they did and not what they tried to do.
Although there might be an area where the line of success is fuzzy (starting with a dire wolf genome and filling in a few small gaps with grey wolf DNA?), what they did here doesn't remotely approach anything like that.
And, of course, it might be that it's still too hard (e.g. having to edit way more genes, to get to >90% dire wolf genome if that's our criterion). If it's too hard then it's too hard; that wouldn't be anyone's fault, but people should be honest about it.
I agree their "dire wolves" look like regular Arctic wolves, but from their woolly mice, it seems pretty clear they're developing the tech you'd need to turn elephants into mammoths. (I consider whether "taxonomy experts" would deem that TRUE de-extinction entirely unimportant.)
I do think Colossal is going to be the Theranos of this field. I don't think we're going to get mammoths out of whatever process they're using, but the woolly mice are undeniably cute (and I'm very resistant to cute animals), so if they pivoted to "exotic pet creation" they'd do everyone a favour by stopping pretending they're going to save endangered and extinct species.
I also think they just jumped right on dire wolves because of Game of Thrones, not because of the real extinct animal. The ordinary person who ever heard of dire wolves probably has an image of the GoT beastie in mind, so "actual wolves just bigger and whiter" is sold successfully as 'this is a dire wolf'.
Yup. They even got GRRM to post about them lol
I do wish they'd just call it living paleoart. Living paleoart is fun!
Yes, they’re kind of misrepresenting what they’re doing,
One possible use for that technology is filling the gap in an ecosystem left by a species going extinct; not bringing it back, but putting in a functional replacement. But that would likely raise the issue of the animal went extinct in the first place because other parts of the ecosystem it depended on are gone, too.
>(I consider whether "taxonomy experts" would deem that TRUE de-extinction entirely unimportant.)
You don't think there might be more to a mammoth than a hairy elephant?
Even if you accept a hairy elephant as a mammoth, it doesn't get us any closer to actually de-extincting other interesting species that aren't a simple aesthetic variation on a surviving one.
No, if they, or anyone else, have any specific criticism – like, head too small, tusks too straight, ears the wrong shape, or anything else – I'd be perfectly happy to concede that as a shortcoming, and then it's a matter of taste whether you think it a significant enough failing that it no longer counts as a mammoth. But the current level of criticism I've seen is on the level of "it wouldn't have the SOUL of a mammoth," and THAT is what I have contempt for.
I agree there'd still be a long way to go to sauropods, pterodactyls, etc. that don't have any living species close enough to hack together a facsimile from, but think it DOES get us closer.
I think it's less about having the "soul" of the extinct creature and more that these creatures likely differed from their non-extinct counterparts in ways that aren't externally visible or detectable from fossil evidence, such as social behavior, diet composition, effects on the ecosystem, vocalizations, brain and organ structure, etc.
And in that case labeling a modern animal genetically modified to look like an extinct animal as the extinct animal is misleading. You can't learn about actual dire wolf behavior from studying genetically modified grey wolves, and (since some people want to clone mammoths to rewild them in Siberia) you can't precisely replicate the effects on the ecosystem of a herd of woolly mammoths with a herd of genetically modified elephants, which might eat a rather different balance of vegetation and have a significantly different effect on the landscape. And people react strongly to this kind of 'deextinction' because there are suggestions we don't need to work that hard to conserve species that are currently alive, as long as we have the future option of making something that superficially looks like them.
This isn't "it doesn't have the soul of a mammoth."
This is "it doesn't have the anything-at-all of a woolly mammoth, even remotely, except for a tiny handful of gene edits, compared to the much vaster number of genetic differences between an actual woolly mammoth and an Asian elephant."
For the "direwolves," it was literally only 20 edits in 14 genes.
The mammoth plan is similar. It's not a plan to clone it from a mammoth genome; it's a plan to slightly edit a tiny handful of genes in an Asian elephant genome.
The tech is useful, I agree, including probably for fertility and genetic medicine.
I'll leave aside whether, hypothetically, the tech could ever be used to actually make a real mammoth with a real mammoth genome (instead of a slightly-edited hairy elephant). The point is that the "dire wolves" are not remotely dire wolves; and their current plan for a "mammoth" is exactly the same kind of not-remotely-a-mammoth.
If it was just their private fantasy, it would be unimportant. Or even for entertainment (I don't mind if Peter Jackson hires a hairy elephant to play a mammoth in a movie). But remember that Colossal and the Trump administration are both pretending this is *literal* de-extinction, and the Trump administration is already trying to use this as an excuse to remove protections from endangered species.
Are there any new/prospective parents who have experience with Alpha School a/b tested with home schooling? Reading the review, it really seems fantastic to a degree I didn't realize existed with off-the-shelf schooling options.
The current essay contest experience could be improved to a significant degree:
1. Everything being dumped into massive google drive documents kind of sucks; headings get weird, the pages are laggy, things have to be sorted, etc, and I'm sure it's a big pain in the butt for Scott and his assistants to copy and paste everything into the documents, especially as the number of entrants rises.
2. People submit essays for many reasons, but a major one is wanting attention/feedback/comments from a large, engaged reading population. The current rating system gives you no way to write comments for the reviews you rate, because it would be a massive headache to sort and collate comments. I had things I wrote down about each essay I rated, but didn't have anywhere to actually put them other than a separate ACX comment.
It seems like the solution is adapting an existing blogging platform or making a bespoke website for this purpose. I'm picturing a website where each essay is uploaded as a post by its writer, where anonymous comments can be left below each post, and where there's an integrated rating system on each post.
I know that someone had proposed an ELO system in a previous open thread, where instead of rating each post, you read two essays and say which one you liked better, and that over time results in an ELO score for each essay. It'd be great if that could also potentially be part of this, because right now, rating is tricky; I wasn't sure how to rate the first few essays I read, because I didn't know how they stacked up.
I'd be interested in working on this project, but have minimal web dev experience. I'm quite good at AI wrangling, but would love to see if there's anyone else who wants to work with me on making the essay contest experience better, who ideally is more experienced in website development.
That is a great idea--people who worked hard to write something basically always want feedback, and I would definitely often post feedback if there was a way for it to be attached to the specific review.
Wanted to say I like your/that person's Elo* system and agree with your rationale.
My main concern is potential abuse by raters who don't necessarily read every essay but might rate every other essay as worse than the one they liked (out of a couple they actually read, say), or as better if they happen to really dislike one, etc.
*not ELO, it's Arpad Elo's last name not an acronym, sorry to nitpick
>It seems like the solution is adapting an existing blogging platform
Why not Substack itself?
Because it doesn't have a well-functioning commenting system, for one.
I'm not sure if it has the functionality we need; a volunteer could certainly upload each essay as a post onto a new sub stack, but I think comments that only show up after the voting is done would be nice, and I think rating should be integrated into each post, which seems tricky to do through substack. If there's a way to accomplish those goals, though, then yeah, substack would work well.
Parts of this seem like they would be relatively easy to do by a volunteer. For example, taking a huge google drive document, and splitting it into many small ones.
For commenting the submissions, a possible solution is a new thread on ACX where someone makes a top-level comment containing the name of each entry and a link to its document, and everyone else will reply to these comments. (Scott would create the thread but keep it "only available to those who know the secret URL", send the URL to the volunteer who would create the top-level comments, and then Scott would switch the thread to "public".)
So I guess what needs to be done is for someone to volunteer (it could even be me) and propose the specific kind of help to Scott, maybe publicly in a related thread, so that other people know what is going on. Well, the next time we do reviews.
The ELO system... yeah, that's much more work.
"someone had proposed an ELO system"
Essay No. 1 - it was excellent, giving this one a 'Last Train to London' rating
Essay No. 2 - solidly workmanlike, 'All Over The World' ranking
Essay No. 3 - will appeal to pretty much everyone, 'Mr Blue Sky' status
I laughed until I 'Turned to Stone' and felt the 'Fire on High'
But 'Hey! Boy Blue is Back'
I agree that the formatting on the entries could have been better, perhaps with each entry linking to its own post, with comments disabled/hidden until after the rating period ends. I too wanted to leave comments on some of the essays I read, though they would have been of more interest to the author than to Scott.
The question of rankings is interesting - ELO doesn't seem like the right tool for this unless you assign pairs of essays randomly to readers. (You don't run a chess tournament by allowing the audience to choose who plays who.) Conventional voting theories don't work because there are too many candidates for voters to be acquainted with (so ranked choice won't work) and you allow voters to participate as often as they like. It's an interesting maths problem actually, though someone who knows this stuff better than me might come along and say this sort of problem is solved already.
I think the Elo idea is assuming that you'd indeed present people with two random reviews. It's already the case that choosing reviews randomly is encouraged, but I imagine that it'd be desirable to have some kind of option for letting people rate a review of their choosing, and I don't know how that'd work.
I'd also be tempted to try to use something like the Smith set rather than Elo, but I don't know whether sample sizes would be big enough for that to be viable. It would at least solve the problem of how to make ranked-choice voting work when most voters don't know most candidates.
It's not clear to me that the best selection of finalists consists of the most popular reviews. You risk lacking variety. (I also don't know how you would find a Smith set, or how that would help here.)
An alternative approach would be to attempt to find the 'most X' reviews for various choices of X: surprising, well written, profound, thought provoking, etc.
I like this suggestion. It would be nice for non-finalists to be able to get some feedback.
Could be good to take a median review from the previous year to act as a standardbearer. "This one was a 5, rate others accordingly".
That seems like it flattens reviews too much to even be useful. Was the mid review mid because it's too controversial/not controversial enough, poorly structured, scope too broad/narrow etc. For book reviews the parameters are narrower so *maybe* it makes sense, but even then it's questionable imo. For open ended "review anything", the median means very little.
Seems kinda mean to the author of the officially mid-tier review.
I'm a webdev (and the person who does the random review script), and I'd do this if Scott wanted to, but I have no idea whether he does.
why not use Substack itself for this?
Substack offers very little customization, so while it would be possible to make a page for each submitted review, I'm not sure how it would help with any of the other features of the proposal. Also the process of creating the pages would still be fully manual.
First off, thanks for the work you've already done; I used pretty much exclusively the random review script to read reviews. It is up to Scott, of course, but it does seem like the current system involves a fair amount of work for him, with all the copy and pasting and compiling scoring information; I'd be happy to email him, asking if this project would have his blessing.
And if it does get his blessing, I'd be happy to help in any way that I can, though I'm unsure how much help I'll actually be.
I'll post my favorites here, to show my appreciation to the authors:
11 Poetic Forms: I'll remember what a Sestina is for the rest of my life.
Airships: Cool and steampunk. Would like to understand the technology better.
Arbitraging Several Dozen Online Casinos: in terms of potentially relevant new skills learned, this was #1. It's not really about "arbitrage" though, or am I misunderstanding that word?
Two Years Of Parenthood: Heartwarming!
Getting Over It With Bennett Foddy: Wouldn't have read it if it weren't for the random review script. I liked the connection to the author's experience as a writer.
Also, a shout out to the first Sermon on the Mount review. I liked how it adds some contrast to Scott's post on early Christian growth and strategy.
I gave that review a poor rating because I consider it epistemically irresponsible to make such strong/spicy claims about matters of historical fact without providing more evidence than the review did. In particular, I would have liked to see some links to academic biblical scholars (not just theologians) who prefer the essay's interpretation, so that I could get a sense of how mainstream or fringe it is in the field.
I see what you mean. The historical context presented in the review seems plausible to me; things like a left hand taboo and laws governing forced labor. That said, whether Jesus or Matthew really had these things in mind remains a matter of interpretation, and I guess no amount of biblical scholarship can provide a definitive answer to that.
It's an isolated reading. I've heard the transgressive reading of those three before, but there are four statements in that section, and the last one is "give to the one who asks you, and do not turn away from the one who wants to borrow from you." I haven't seen someone offer an anti-Rome version of that one.
Kind of like how people will put together a cause-and-effect explanation of the Ten Plagues of Egypt, but then don't bother explaining that time Moses and the Egyptian wizards turned their staves into snakes.
I'm sure one could make a case that high interest rates were a tool for oppression, and that moneylenders were heavily regulated ;-)
But you are right that it would be quite a stretch.
For folks living in NYC who are voting in the Dem primary for mayor, I wanted to a) remind you that the election is tomorrow, please go out and vote! and b) strongly encourage you to rank Mamdani over Cuomo.
I disagree with both of them on policy, but of the two it seems obvious to me that Mamdani is legitimately interested in making the city a better place, while Cuomo is corrupt, self-interested, and (following the Trumpian approach to politics) is an all-together indecent person.
The Democrats need fighters. But those fighters must be as virtuous and honorable as they are aggressive, people who are so clean that nothing sticks to them so they have full ability to go after everyone else, people whose very presence on the debate stage shines a light on the hypocrisy of those who would seek to use public office for anything other than the good of the country.
Cuomo is not that guy. Mamdani is.
I make the case for Mamdani (and for decency in politics generally) more extensively here (https://theahura.substack.com/p/decency-is-more-important-than-policy)
A point you could have made is that Mamdani, like Bill DeBlasio, might moderate significantly in office since there’s a limited appetite for hard left policy in the city (as opposed to primary voters).
Also, the Democratic establishment really needs to start backing better candidates than these uncharismatic insiders. Just look at how much money got poured into Cuomos’s lousy campaign!
I purposely wanted to avoid policy discussion, because it's not germane to the larger point. I'm rather disappointed how many people here seem to be willing to vote for someone who is obviously a bad person if it means small concessions on their policies of choice.
Well, it's done.
I have time until November to think up some falsifiable predictions about the next four years.
From what I understand, from all of Mamdanis promises, two that he can execute on without too much interference are a) expanding rent control and b) repurposing subway station into homeless help centers.
Both of these are tragically bad in my opinion. The latter seems easier to do, so my first prediction will be that the homeless situation in NYC will begin to resemble that or Seattle or SF (before the latest elections)
A competently corrupt politician is often preferable to an enthusiastically well-meaning but inept one. The corrupt politician has an interest in not befouling his own nest; the enthusiastic idiot often won't even notice he's doing that.
If you want reform, find reformers who understand how things actually work.
a few folks already posted this, kenny's steelman is here (https://www.astralcodexten.com/p/open-thread-387/comment/128812105) and my response
> I agree this is the steelman, but implicit in the steelman is the assumption that the earnest person is also dumb or blind (because they wouldn't realize or be willing to realize that their policies aren't working right). I think it's extremely rare that someone is dumb/blind, earnest, has ideas that are so bad corruption would be better but is effective enough to actually implement those policies over the objections of others, AND is able to mount a successful campaign. Bayesian prior of all of those together is extremely low. (Trump gets almost all of these except earnest)
I don't particularly want reform. I want politicians that are going to act like adults and try and solve problems. I disagree with your use of the word "often". I would say "very very rarely"
The idiot doesn't have to be *effective* to be a menace. If their dominion is facing real and serious problems, being ineffective and blocking more effective people from solving those problems, is sufficient.
Or is the contention that NYC doesn't face any really serious problems and doesn't need capable high-level governance?
Again, I understand in theory that all of this is true, but I think you have a steep hill to climb when you're openly admitting that the counterfactual is someone who is openly corrupt -- and presumably, also effective (at being corrupt). My prior is extremely high that decency will lead to better outcomes on average than corruption. And unless you think that being a decent person is somehow highly correlated with being extremely ineffective, your prior should also be equally high.
John already climbed that hill at least partway by pointing out the corrupt person still has an incentive not to "befoul his own nest", and also has the means. He might take bribes all day, but if he's also smart enough to understand that his own welfare depends on that of his city (as does the decent incompetent), and is also skilled enough to maintain that (as does *not* the decent incompetent), then the city will likely continue to do better than if the decent incompetent is in charge.
You could argue that that still isn't enough to close the gap, but you haven't made that argument yet. Meanwhile, there are multiple examples from history that ought to lower that extremely high prior you possess.
Sorry, no, the burden of proof is on the guy who's arguing that corruption is fine. I don't feel the need to do the work of "looking at history" or whatever if other people aren't going to make that argument.
Also, at this point you're basically arguing a tautology. "The person who does better will do better"
Fine, I'm willing to concede that I can't defend putting a person with the IQ of a banana in office vs a corrupt person with the IQ of an average person.
While I broadly agree about NY political corruption, you do realise you are inviting investigation into whether Mamdani has any skeletons in the closet? Cuomo and his scandals are already known, but is Mamdani as squeaky-clean as you make out?
I'm not particularly interested in this election but simply reading the news gives me, whether I want it or not, a lot of coverage. And honestly? all the hagiography around Mamdani makes me think that if elected, he'll be an AOC type: yeah, talk the talk, be very good at holding on to their seat, maybe try a few Big Things but get slapped down or have them watered down, and settle comfortably in to being a career politician.
I don't mind a bit of the ould socialism myself, but expecting (pace Jacobin) "a promise of a fundamentally different way of doing politics for the working-class majority of a city facing an out-of-control crisis in the cost of living" makes me smile. Every hog, dog and divil running for election promises that. Then they get into power, have a look at the tangle and mess facing them that has been left behind by all the previous administrations, court cases, regulations, etc., and realise that it's going to be business as usual for them.
Even Elon and the chain saw couldn't cut through it all! I don't think New York has much wiggle room for sweeping changes to how the money works in order to make the working-class very much better off.
Speaking of that Jacobin article, really, Mamdani? Really?
https://jacobin.com/2025/06/new-york-times-mayoral-mamdani
"Far from “show[ing] little concern” about crime, public safety has been a prominent part of his platform. Mamdani’s ambitious plan for a Department of Community Safety pledges to build out social services and mental health care to help humanely address the root causes of crime."
Now, when was the last time I heard about a Committee of Public Safety? 😁 He might want to rethink the name there. And yeah, I'm sure a corps of social workers is going to stop violent/crazy criminals in the streets.
https://en.wikipedia.org/wiki/Committee_of_Public_Safety
> any skeletons in the closet
I mean, he's running as a socialist and campaigns with the slogan "globalize the intifada." What kind of skeletons do you imagine that could be worse?
You could try to muster some rape accusations, but those won't stick. Maybe you could find (or "find") video of him singing along to rap in high school or something, but that's weak sauce.
He doesn't seem to have been in politics very long, so what did he do before?
There's already been a little kerfuffle over "intafada". He was a rapper as a young man? Then somebody must have sample lyrics of him rapping about bitches and hoes and drugs and guns (unless he was an exceedingly wholesome rapper) 😀
Swearing about your granny? Having your (actress playing your) granny swear? Oh, my!
https://www.theguardian.com/music/2019/may/10/mr-cardamom-madhur-jaffrey-nani-video
"How do you get an octogenarian south Asian screen icon to swear in a rap video? According to hip-hop MC Zohran Mamdani, a little context can be helpful. When he came to prepare world-famous actor and TV cook Madhur Jaffrey for her role – lip-syncing along to the words of his expletive-laden tribute to his maternal grandma, Nani – he had a little explaining to do.
Mamdani, who performs under the name Mr Cardamom, can remember Jaffrey asking him: “In the lyrics, who is this ‘mother fucker’ for?” he says. “Fuck this person, fuck that person; now whose ‘fuck’ is this?”
The guy may indeed be squeaky-clean, but if someone wants to dig deep enough and far back enough, it may be possible to find something that could have a bad interpretation put on it.
Update: they found something that might work. He did the Pocahontas thing of lying about his race to get affirmative action benefits.
> He was a rapper as a young man
> hip-hop MC Zohran Mamdani,
Wtf?! I didn't know that: I was simply using "singing rap lyrics" as an example of the kind of thing normal decent people get smeared for*.
*Mostly by the kind of people now preaching about how decency is so important.
OTOH, if he adopts Mme Guillotine as a core part of NYC's justice system, *that* could work to keep crime down....
> you do realise you are inviting investigation into whether Mamdani has any skeletons in the closet
good?
> he'll be an AOC type: yeah, talk the talk, be very good at holding on to their seat, maybe try a few Big Things but get slapped down or have them watered down, and settle comfortably in to being a career politician
I'm pretty neutral on AOC, but, like, is your take that AOC is very corrupt or something? "a career politician" did she do something bad? or are you just annoyed about the aesthetics?
Net net, I purposely did *not* focus on either Cuomo's or Mamdani's stated policy positions, because the point I want to make is that being a decent person is a necessary precondition to talking about policy at all. I still think that's true, digs about the French Revolution aside.
Okay, if Mamdani avoids photo ops of him bawling his eyes out outside a parking lot, that would be an improvement.
In a race between someone who is a corrupt petty sex abuser and someone with aesthetics you don't particularly like, I personally would go with the mid aesthetics
Not the aesthetics, it's the opportunism. Turns up with a 'Tax the Rich' designer dress at the Met Gala. Yeah, really getting down with the proles there, Sandy!
Though I do have to respect that she's managed to retain her seat and get her career well-established, unlike the rest of the Squad who have either faded or got bogged down in sectarian controversy. She's dug herself in for the long haul as representative for that Bronx/Queens district, successfully turfing out the old Democratic party guard there.
Honest, Well meaning socialists did a lot more damage than corrupt capitalists.
I would recommend asking the right questions. First of all in the original Marxian analysis, the problem is with the system itself, not capitalists as persons. Now, libertarian economists tend to see the market also something like a machine, except that they see it as a good one. So I guess both agree capitalists do not have much personal responsibility, the market is a machine, if there is money on the table, someone will take it.
OTOH socialists dictators DO have a lot of personal responsibility, because this was not a machine like that. Stalin could have easily decided to not create the GULAG.
I think that several people have this take, and it always seemed like handwavy vibes-based feelings grounded in GOP scaremongering more than anything based an actual political outcomes. If you have any actual examples of this (and I mean relevant examples, not, like, "This cambodian general called himself a socialist in Khmer Rouge") I'd be open to seeing it.
But, also, fine, rank any of the other candidates, just not Cuomo.
San Francisco
San Francisco, a place that’s so undesirable to live in that you can rent a 2-bedroom apartment for … oh, hmm, $4,000 a month.
Sorry, what was your point again?
Only a modern-day progressive would cite unaffordable rents as evidence of success
Even though you're being snarky here (which is fair, you're responding to a snarky comment) I think this is obviously a silly take under the hood. Progressives are famously supposed to be the people who *dont* understand market dynamics. But here you argue against market dynamics while tarring progressives for...understanding them? Very backwards.
As an aside, dinging vectro for being a 'progressive' like its a big scary dirty word really just says more about you than anything else. Ooo vectro is a progressive...what are you gonna call him next, vegetarian?
increasingly of the opinion that people don't actually know what policies mamdani ran on or have any idea who the major players are in the big cities in the US
What *about* San Fran? Are you saying Mamdani is like Lurie, a moderate? Or like Breed, an establishment Dem? What policies do you think are the same between these people? Have you lived or visited SF over the last 10 years? What is your understanding of the problems that SF faces?
How did that work out for Brandon Johnson?
I think Michelle Wu or Lisa Hidalgo are equally reasonable comps who have done extremely well, so its increasingly unclear that Johnson isn't just a fluke or a particularly bad candidate/mayor
I dunno enough about Chicago tbh. I've heard the comparison a few times and agree that it's possibly a good comparison. Do you know more / have an informed opinion? Is Johnson unpopular because of policies? Gaffes? Scandal?
https://www.slowboring.com/p/progressives-need-to-reckon-with is a decent summary of the case against Johnson.
Thanks. I think just reading this, with nothing else to go on, it seems like the biggest problem with Johnson is not the ideology, but rather being too ideological. He seems like a crusader more than a leader, someone who does not know how to make trade offs because making a trade off of any kind would be impure in some way. The left absolutely has these people; left activists even more so. But I think this is true of any political affiliation or party -- you have the people who are shrewd leaders who actually know how to do things, and the people who are ideological warriors who are best when they are pointed at certain things by people in the first group. One of the big failures of MAGA is putting all of the ideological warriors in charge. The Dems have historically not done this in public office (Bernie, for eg, famously got knifed by the Dem party). It's likely that there is more appetite than ever to elect warrior types into office on the Dem side, and I agree that would be terrible, as it seems to be with Johnson.
But it's not at all clear to me that people like Mamdani (or even AOC or Sanders for that matter) are so ideologically uncompromising that they do not know how to hire good people, read the room, and be outcomes driven. This matters because on the ground, the policies of Johnson and Mamdani _don't_ seem that similar. Yes, there's a general tax and spend attitude, and I agree that is similar across the two of them, but I think the similarities more or less end there. It was Cuomo who was more embedded with the Unions, and more likely to cut favorable deals with them in exchange for their backing. And it was Cuomo who would regularly refuse to create a reasonable budget for the city and state, mostly to fund vanity projects or things that were obviously unworkable. In these things, Cuomo and Johnson are quite alike.
I think there is a way to do tax and spend that doesn't cause the city to burn. The question is whether Mamdani is too much of an activist to be able to thread that line. And that frankly remains unclear to me. But I think he's talking about the right things, and the ways in which the city can make things easier for constituents by *removing* red tape (eg here: https://x.com/ZohranKMamdani/status/1878853557111414795)
What about hugo chavez from Venezuela?
what about him? Why do you think this is a reasonable comparison, beyond Chavez also calling himself a socialist? who are the corrupt capitalists you're comparing against in Venezuela that would have done better?
If you want to point to Venezuela without any further explanation, I think it's equally fair for a socialist to point to the nordics with equally little explanation.
There are enough corrupt capitalists in south america to choose from. They are and were bad, not as bad as Chavez.
Which policies the Nordica have do you find socialists? They are pretty capitalists, though with a big safety net.
Calling the nords socialist or socialist-lite is pretty common i think. From wiki (https://en.wikipedia.org/wiki/Nordic_model):
> This includes a comprehensive welfare state and multi-level collective bargaining[2] based on the economic foundations of social corporatism,[3][4] and a commitment to private ownership within a market-based mixed economy[5] – with Norway being a partial exception due to a large number of state-owned enterprises and state ownership in publicly listed firms.
> While many countries have been categorized as social democratic, the Nordic countries have been the only ones to be constantly categorized as such. In a review by Emanuele Ferragina and Martin Seeleib-Kaiser of works about the different models of welfare states, apart from Belgium and the Netherlands, categorized as "medium-high socialism", the Scandinavian countries analyzed (Denmark, Norway, and Sweden) were the only ones to be categorized by sociologist Gøsta Esping-Andersen as "high socialism", which is defined as socialist attributes and values (equality and universalism) and the social democratic model, which is characterized by "a high level of decommodification and a low degree of stratification.
Social corporatism, mixed economy, and norwegian state ownership are often seen as 'socialist'. Also I think generally welfare state stuff is seen as socialist too in the US. I think that's kinda dumb -- imo healthcare is obviously its own thing and I think universal healthcare in the states would immediately solve so many problems its honestly silly we havent implemented it yet -- but it is true that about half the country thinks obamacare is socialist (and half those think its communist!)
For better or worse, I think that 'socialist' in America has come to mean 'strong safety net and high taxes'. Which is more or less what mamdani is proposing, like I haven't seen him saying stuff like "we should nationalize oil"
> There are enough corrupt capitalists in south america to choose from. They are and were bad, not as bad as Chavez.
I think this is kinda weasely. You made the claim that Chavez was a good comparison to Mamdani! Can you back it up more than this?
Based on your comment history I dont believe your verdict is informed by his decency. You prefer the socialist to the moderate because hes the socialist (and being coy about policy disagreements) Whether you think thats fair or not, I think readers have high bs sensitivity to decency based arguments that are just laundering political preferences. See GOP preference for decency during clinton/bush era vs now.
In my experience socialists are very prone to equate agreeing with them politically = being a decent person. And historically socialist regimes have treated political disagreement as one of the most heinous crimes (and recruited even family members as informants.) I think it’s an intellectually and morally bankrupt political philosophy, but it seems to stick around in institutions of higher education and in the minds of their students.
It’s interesting that we seem to have learned only the lesson of fascism, and not the lesson of communism, from the 20th century. No one ever says we need to try fascism again, Hitler just did it wrong - but I hear it frequently about communism from seemingly earnest people who are quick to dismiss Stalin and Mao as aberrations.
There are decent and sensical reasons why communism is never going to be fully purged from the Overton window despite being a bad idea: https://balioc.tumblr.com/post/614257600500989952/communism-in-its-simplest-conceptual-form-is-an
> agreeing with them politically = being a decent person
This is a far more widespread position than you imply.
Sure, it’s people on both sides, and it sucks when any of them do it.
But hey, I disagree with Scott politically and I still think he’s a great writer, thoughtful, intelligent and I avidly read anything he puts out. I also enjoy Freddie DeBoer even though he is “far left.” He’s honest and that’s more important.
Yes, but I wasn't actually making a "both sides" point here: You said it was only "socialists," and my point is that it's not just the fringe, but a mainstream position among Democrats as well.
Well, Scott still lets me post here, but it's true he's one of the rare classical liberals who still strongly defends free speech. I wouldn't last 2 seconds on Reddit or BlueSky. And yes, it's definitely worse on the left than the right - as much as liberals like to complain about Elon, they are welcome to do so and post their hottest takes right on Twitter/X (although many have left because the mean rightists keep replying to them and Elon never seems to do anything about that)
a) claiming to have special insight into the motivations of a person you don't know beyond their stated motivations is...bold, to say the least. Very bayesian of you. (Claiming to have special insight into the motivations of "readers" -- also very bold)
b) my comment history is motivated by a disdain and disgust of the Trump administration and everything it stands for. That you think 'hatred of Trump == socialism' says more about you than it does me.
c) vote for literally anyone other than Cuomo, it doesn't have to be mamdani he's just the person who's second. If Lander or Adams were second, they would be better choices. Literally *anyone* would be a better choice than the multi-time sex pest with a penchant for corruption.
Why do you feel the need to defend the multi-time sex pest with a penchant for corruption? "Our policies at any cost" is fucking stupid, sorry. I care about voting for decent people, because at the end of the day my own policy preferences are likely to be misinformed, the policy positions of candidates on the campaign trail are *definitely* misinformed, and when push comes to shove I want my public leaders to put their constituents over themselves. Cuomo is so selfish he's willing to split the vote just to stay relevant in politics. It's disgusting, as is defending him because you're stuck in the cold war.
In case it matters, I'm pretty economically centrist. More taxes on large corporations and the wealthy. Universal health care. Way fewer regulations, especially on construction. No taxes for small businesses.
I think most of Mamdani's policy positions are mixed to bad. I don't like rent freezes or rent control, and I don't get the groceries thing. Cuomo's policies are shockingly similar, though no one ever really reports on this. Doesn't matter though -- the main point is that decency is a prerequisite to even being considered a candidate. Mamdani isn't going to be cowed or bribed by Trump because, and I can't stress this enough, *he's a decent human being*. If Cuomo thinks being pals with Trump will benefit him, he will turn immediately -- just like Adams did before him. No thanks. I want to actually like my leaders.
"I dont believe you" is something ppl are allowed to say and doesnt require them to believe they have special insights into your motivations. And if I dont believe you, others probably dont either, because im not so special.
I offer no defense of Cuomo, and the things you say about him are probably true. But sorry, I dont believe you, and also I think mamdani would be bad (see I can just say that outright)
I agree, and further suspect this stance that one's policy preferences shouldn't outweigh "decency" would be abandoned as soon as the differences become significant enough*, probably by arguing that anyone who held those positions ipso facto lacked decency, and so voting for the corrupt bully with your policies over his honest opponent becomes the better choice after all.
*The opposite position on ALL of gun rights, abortion, affirmative action, LGBTQ+, and immigration, say.
> is something ppl are allowed to say
If the best argument you have to advance your point is that you literally are allowed to say it, you might as well just concede there's nothing here. If you want to make claims about my political positions as a means to disparage me, at least have the decency to back it up with evidence. You claim to have ample evidence with my post history after all. Or is "socialist" just a bad word you're using to tar by association?
That was a rebuttal to your own reasoning.
"Based on your past actions,I suspect your motives"
"You dont know my motives as well as I do"
"So what, the whole point is I think you are being dishonest"
It is true, I havent taken the time to prove you are being dishonest about your motives. And I wont because, my goal is not to take you down.
I thought you might feel like my skepticism was unfair, I said as much in my first post, I am just giving you my honest feedback. I cant take your argument touting a leftist above a moderate (fine I wont say socialist) on grounds other than their leftism serious because I see you commenting like leftists do. There are people who criticize Trumps terribleness without giving a leftist impression. You arent one of them, and it affects how others receive your arguments.
> I see you commenting like leftists do
I'm sorry you don't like my aesthetics. I'm not a leftist, but you're welcome to your opinion. Death of the author, consider my position despite your dislike of the leftist that you think I am.
Disclaimer: I am also a socialist, and a Boston-born Red Sox fan who has never enjoyed visiting New York.
That said, the "moderate" (I don't agree with that self-appellation; this faction is socially liberal but seems every bit as committed as the radical right to fucking over the less fortunate) preference for Cuomo speaks incredibly poorly of that faction in NYC. He's a former governor who resigned in disgrace and is best known for killing the elderly in nursing homes during the Covid pandemic... and changing laws to give unsafe nursing homes immunity from lawsuits. You can't even defend him on idpol grounds, given his history of sexual harassment and homophobia (vote for Cuomo, not the homo!) or party loyalty (he worked to hand the state senate to Republicans).
What's the case for him - he's rich and famous?
NYC has ranked choice voting and there are other non-socialists on the ballot.
>homophobia
Seems an odd claims given his pivotal role in legalizing same sex marriage in NY. https://www.cityandstateny.com/politics/2019/06/how-new-york-almost-didnt-legalize-same-sex-marriage/177228/
The OP should probably have chosen a different moderate candidate to recommend placing above Cuomo in order to appeal to socialist skeptic voters with a sense of decency, as yes its ranked choice. Instead the OP makes the case for Mamdani, hence my skepticism.
It doesn’t matter who else you put above Cuomo - it will likely come down to Mamdani and Cuomo and anything else you do is likely primarily aesthetic and expressive.
Last time garcia was 3rd first round down 11pts from adams. She finished 0.8 behind Adams. The ranked choices can accumulate and I would not take it as a given the top 2 first rounders are the last 2.
Polls haven’t just looked at top choices - there have been some that try to measure head-to-head support of several candidate pairs to see who are likely to be the final two. Last time, polls showed Garcia and Adams and several others as possible finalists, I recall.
20 million dead Americans say Cuomo did the right thing!
What? They didn't all die?
I guess covid19 wasn't all it was cracked up to be.
Cuomo got too popular, so like Spitzer, he had to go. Unlike Spitzer, Cuomo declined to fall on his own sword. The democrats are in a bad place, where they believe that anyone getting "too popular" (read "bigger than the current TopGuy") needs to go.
Grade B people hire Grade C people.Grade C people hire Grade D people (so as not to be outshone). And suddenly, Democrats!
Less of this style, please.
Would you like me to elaborate on the plans that Cuomo was following?
Or would you like to discuss the absolute dearth of "coming superstars" in the Democratic party?
Mamdani is 33 years old and the entirety of his political experience is 4 years in the State Assembly. He hardly seems qualified for the job.
We know from prior experience that Cuomo is evil and self interested, so nearly any breathing person is better, as we don't have proof that they would act in a way that as bad as possible for the city. Is that not obvious, or is proving those foolish leftists wrong more important than acting in self interest?
>Cuomo is evil
It is hard to take seriously anyone who uses terms like "evil" to describe someone.
> is proving those foolish leftists wrong more important than acting in self interest?
If I wanted to prove foolish leftists wrong, I would vote a foolish leftist into office and watch what happens. As for self interest, I own a home in NYC, so it is in my interest to vote for the guy whose policies are least likely to limit the rise in housing costs,which again would be a foolish leftist. But I want to do the opposite which means I might have to vote for Eric Adam's in the general election, regardless of who wins the Dem primary.
> I would vote a foolish leftist into office and watch what happens.
We even already have a cautionary tale to point to in Chicago.
Just to clarify, your position is that Mamdani will bring housing prices down?
No, my position is the opposite. As a homeowner, it is in my interest to have housing prices increase.
...and therefore you're voting for Cuomo?
sorry if I'm being dense, there's too many double negatives in your initial post and I'm not sure what exactly you're saying. You're a homeowner, therefore you want housing prices to increase, therefore...?
Evil and self-interested in a politician spells Nixon to me. As he was a smart guy, he did stuff that was mostly good for America. That your description of "why not to vote for him" dovetails so nicely with Nixon doesn't speak well to your discernment.
"Cuomo is like Nixon" is a fantastic metaphor, thanks 🙏
I've definitely seen Cuomo/Nixon comparisons, and from Democrats.
Did you really just say that a rule that advocates voting against Nixon is ipso factor worse than a rule that advocates voting for Nixon?
Yes? He's not Hitler. "A smart enemy is easier to deal with than a dumb friend."
I'm not going to justify your low effort comment with a proper response but you sound like this;
Jews for Hitler is perfectly reasonable! Sure, he might be evil and self interested but so was Nero! He brought untold status to Rome by acting in an self interested manor, so therefore I'll look for that quality in all future candidates. This non-intuitive conclusion is extremely satisfying! I am very smart! I'm so smart that I value looking smart over anything else, including my own self interested. I prefer to look out for the interest of capital and billionaires to signal to other Very Smart people just how smart and clued in I am!
https://en.wikipedia.org/wiki/Association_of_German_National_Jews
And you were saying? Unfortunately, this was quicker to cite than actual Jews in the Nazi Party.
I do not in fact believe that the "Road to Zero" is going to lead anywhere good. At this time, neither is solar power nor wind power (without batteries, the grid needed burns more natural gas, not less).
See, you're making an actual argument here: "Cuomo is Emanuel come again" -- and that's making an argument, as a reason not to vote for him. It's no longer "He's evil" but "His constituency wants things that I do not want."
> I'm so smart that I value looking smart over anything else, including my own self interested
The wikipedia link you cite is not exactly beating the allegations. Somehow, you've found yourself arguing things like Jewish support of Hitler means that people should vote for Cuomo. I think you should take a long look in the mirror. "Owning the libs", or "winning internet points", or whatever your actual ideology is, isn't worth corrupting your soul.
On the flip side, if you're openly of the opinion that jewish support for hitler was reasonable, I'm just going to block you.
o undoubtedly. We don't know whether Mamdani will do a good job or not. But we know that Cuomo won't, so the choice seems obvious.
Also, I'd rather an inexperienced but decent human who can pretty easily learn on the job, than a corrupt 70 year old who is going to wield power for his own petty bullshit while lining his own pockets. Again, the choice seems obvious.
>But we know that Cuomo won't,
Not only don't we know that, your article made no such claim. Indeed, it ridiculed those who would make their choice based on which candidate would pursue better policies.
One of the many points of the article is that you cannot trust what Cuomo says or does, because even though he says he will pursue abc policy he's actually just going to do whatever is best for himself.
If you think Cuomo is going to do a good job, make the case. You seem to admit that he's a corrupt and indecent person, so I feel the bar for defending him should be pretty high.
> It ridiculed those who would make their choice based on which candidate would pursue better policies
Right now, I think your stated position of "I'm voting for the corrupt and altogether indecent guy because I want housing prices to go up" is a pretty horrible take. There's a suite of ethical reasons why its bad, but honestly I think its bad on your own stated merits -- your housing price isn't going to go up as fast if the mayor does a bad job, and Cuomo is going to do a bad job because he's corrupt and petty.
Or, put another way, "I think the guy who's going to siphon funds for pet projects and political retribution is going to make my community better" is a wild take.
>One of the many points of the article is that you cannot trust what Cuomo says or does, because even though he says he will pursue abc policy he's actually just going to do whatever is best for himself
That is a claim, but whether it is true is another matter.
>You seem to admit that he's a corrupt and indecent person,
I did not admit that. My point is that policy matters.
>Right now, I think your stated position of "I'm voting for the corrupt and altogether indecent guy because I want housing prices to go up
I actually said the exact opposite.
>"I think the guy who's going to siphon funds for pet projects and political retribution is going to make my community better" is a wild take
1. I don't accept your premise.
2. Even granting your premise, as someone else pointed out re Nixon, your conclusion does not necessarily follow from your premise.
The steelman of the case for Cuomo over Mamdani is that the corrupt 70 year old who wants petty bullshit while lining his pockets might still be preferable to someone who will earnestly attempt a lot of policies that will cause massive problems.
I don’t live in New York and thus haven’t thought hard about what the candidates are promising to attempt, so I don’t know how to evaluate whether Mamdani’s promises are that bad, or whether Cuomo also has some equally bad promises. But sometimes, the corrupt person is better than the true believer.
Also a corrupt 70 year old is less likely to stick around long term.
I agree this is the steelman, but implicit in the steelman is the assumption that the earnest person is also dumb or blind (because they wouldn't realize or be willing to realize that their policies aren't working right). I think it's extremely rare that someone is dumb/blind, earnest, has ideas that are so bad corruption would be better but is effective enough to actually implement those policies over the objections of others, AND is able to mount a successful campaign. Bayesian prior of all of those together is extremely low. (Trump gets almost all of these except earnest)
> so they have full ability to go after everyone else
See, THIS is actually a reasonable argument, that decency (or the appearance of decency, which is the same thing in this business) is a good weapon for attacking Trump/national Republicans with, and that translates to longer-term federal policy gains worth the shorter term local policy hits.
That is NOT the argument you make in the essay. There, you seem to espouse a doctrine divorced from reason, leaning fully into voting as a kind of religious ritual to affirm sacred values instead of a bloodless transaction granting power (and the concomitant opportunity for abuses thereof) to a candidate in exchange for him enacting your preferred policy.
I...direct quoted from the article?
Yes, but you leave out the absurd bits that are the main thrust of your essay (summarized well in the title, so kudos for that) in the parts you excerpted here, and which change the meaning entirely.
K
Well, it looks like it's too late to organize review reviews under one comment, so I'll just throw them up as a top comment. Although I didn't read many in the first place.
The Bo Burnham review had bad chronology problems. There were several points during the review that I thought we had finished with the main album and were jumping up to focus on a follow-up album, only to realize that the jump ahead had been an unmarked aside and that we were still only partway through the first album. If all those jump-aheads had been in parentheses (or not there at all) the review would have been stronger.
Phoenix Theatre had too much backstory at the beginning, and also monstrous paragraphs that are ugly to the eyes; the Word Count tells me paragraph 2 has nearly 400 words. I don't remember how far I got into it but I don't think I got to anything actually about Phoenix Theatre.
Skibidi Toilet seemed like a really solid review, but I was only a page in before I realized I hated the concept of the show and actively want to never think about it. It's like seeing a review on a painting of smeared shit. No amount of effort will make it pleasant.
My fiction writer AI has:
-Suggested that I upgrade the AI on my starship to one with self-learning capabilities.
-Described an AI in a different context as having "endless wisdom." (It set my alarm clock for me.)
How worried should I be?
Hi, I wrote the review of Effective altruism/rationalism, and really want to hear what people think of it. No reason is required to not be selected, I assume all finalists are great of course.
But I hoped it would start a real debate about how to institutionalise the movements in a less illiberal and elitist way. Instead, I haven't even seen it mentioned in a comment. I want to understand why:
1. It landed as obvious (we've noticed the skulls). In this case I want to hear whether everyone knows this is a big problem, eg sick of the discussion, or if the reason is that you think it is not actually a problem.
2. It landed as wrong/misguided. Please argue
3. It landed as rude/arrogant, poorly written or not the right kind of text. Feel free to give feedback
To me (obviously, that's why I wrote it) what I point to really is a problem on the ground, so I want to hear people's arguments
I also gave it a rather poor rating. I should first say that I have never really looked into the EA movement. Basically all I know about it is via Scott, so I may have wrong impressions.
But the impression is that the points you raised are not really new. Like, the tension between mainstream/free thinker. Or I would rather describe it as tension between the inside/outside perspective. How much should you trust your own reasoning versus the conclusion of other people, if those do not agree?
I think this is a problem that at least Scott is very aware about, and I have no indication that the EA movement is not aware of it either. The issue is that this is a really, really hard problem. I face it myself a lot of time when I read the news, and I don't have a good solution to it. You push for going more into mainstream/outside view. But I don't think that generically this is very helpful advice, because you are using your own inside perspective here (which is an outsider perspective for EA). You should first consider that your inside perspective may be wrong, and that the EA perspective is right, and I didn't have the feeling that you discussed a lot why you are so sure that you are right and they are wrong. In fact, I would assume that for some concrete issues you are right, for others the EA people are right, and that it is a really, really hard problem to tell which is which. Personally, I am totally not concerned about shrimps, but am I right about that? I don't know! And I totally agree that x-risk assessment is very speculative. I think the EA community is also painfully aware of that, this seems to be one the central issues they discuss endlessly. But is the right course of action then to dismiss x-risks because it is speculative? You seem to say yes, and I see the arguments for yes, but I also see good arguments for no, and overall I find it very unclear what we should do in such a case.
So, in summary my criticism would be the following. Mind that I only focus on the negative points here. Your review was not as bad as it may sound from just listing the criticism.
- I have the impression that the EA community is totally aware of the problem (balancing inside/outside view) you tell them about. But they understand better than you that it is a hard problem, and just pushing generally for "go more outside" isn't very helpful.
- You tell them to go further in direction A. From my perspective, this is just a single data point, your own opinion, and I didn't find that you acknowledge this limitation. Phrasing it nasty, your main argument for direction A was "I think it is so", and that is a weak argument to me.
- This also made the review a bit arrogant. I didn't mind this so much, but I could imagine that people with more skin in the game might.
That review annoyed me and I gave it a poor rating, which I felt a little bit bad about as I'm a card-carrying EA and so potentially biased. I think it's cool that you're interested in feedback though. (For what it's worth, most reviews didn't get any comments, since the thread wasn't really set up to facilitate that, so I wouldn't read much into it.)
I think my three most clearly articulable objections are (1) most of these criticisms could be leveled against any set of political/social opinions at all, making it unclear why EA is being singled out; (2) I can't figure out what the kind of movement you'd like would actually be in favor of (you seem to support some kind of "mainstream" synthesis that I don't think exists); and (3) it makes a whole lot of different critiques in a somewhat scattershot fashion that makes it hard to tell which ones constitute the true rejection. There are also some smaller things that I think the review gets wrong or incoherent, as well as a few "yes we've noticed the skulls" points (where the issue is the lack of a preferable alternative) and a few things that are just kind of grating in a way that marshals people's biases against being persuaded.
I started elaborating on these, but unfortunately I find longform writing kind of difficult, and wanted to get something posted rather than just procrastinate forever on this, so for now I'm just posting the short version. If you're interested in further elaboration I can try to do that, especially if you either ask specific questions or want to do it on Discord or something (because I write more fluently there).
Again, thanks for asking about this, I think it's a good thing to do and not common enough.
Thanks! See my other reply. (1) No, see piecemeal/Utopian, (2) I outlined changes at the end, (3) i'm not rejecting EA, I'm saying it should be implemented in a less illiberal and individualistic way if it wants to achieve its stated goals.
These points must have been poorly formulated for all of the commenters here to think I was doing some hit piece on EA with no alternative. I get some of it, but still not really. Send me as DM here if you want, happy to discuss further
I went ahead and sent a DM.
My private notes on this one went:
> Where (other review) was too personal, this was too abstract. One of the charges laid against EA was: “If they have a job in a non-EA organisation, they care more about leveraging the job to further agreed-upon EA policy goals than to do the job itself well.” No examples were given.
> Then there was a bunch of stuff about how hard it is to have good epistemics. I agree that epistemics can be difficult, but I was looking for an argument that said “and here are some examples of EA people making these specific mistakes” and the only one I saw was Sam Bankman-Fried, and I’m not sure if I’d count him as EA.
Trying to give a more general comment, though it's been a few weeks:
It felt like this was an attempt at armchair-theorist reasoning from first principles about why the EA movement might have problems, but if we accept this reasoning as correct then we have a fully general argument for why nobody should ever try to make the world better in any way. I think that making the world better is important, and I think it would be very hard to move me from that position without lots of practical evidence.
If an argument wanted to convince me that EA was bad on net, I think the argument would have to start: "Here are the N largest and most important EA ventures, and here is evidence for why each of them caused more harm than good."
Thanks, this is useful! Here's my response:
I agree it should have had more data/examples. This is partly lack of time, partly not wanting to call out smaller actors, and partly not obvious how to document it. I ended up relying on my personal experience, counting on people recognising behaviour they have also seen (see other comment in this thread).
As a journalist, when I meet EAs that know I share their background, they tend to assume I will strategically write whatever is good for EA's concrete policy goals, colluding with them rather than doing good journalism.
Reading these comments it also seems like I underexplained some points. I'm not making a generalised argument, and I'm not saying EA is a net bad (??). I'm arguing for Popper's piecemeal social engineering rather than Utopian social engineering.
EA is great for global health and other areas with tangible benefits. My criticism is specific to politics, illiberal attitudes and institutional design in the movement. When you cheat in politics, you legitimise cheating for your opponents and we get a race to totalitarianism. This can be institutionally addressed along the lines I described.
I understand the call for evidence, but how would the world look if I was right? I would say in that world the movement would remain a subculture, get a cultish shady reputation with bad media coverage and a scandal or three as it moved further into politics, rather than becoming more well-known and gradually accepted. These are data points, albeit post-fact.
Sorry this for rambling, didn't have time to write this properly. I agree that a full argument would dive more into the specific areas but this is what I had time for. Theory is also a thing
> counting on people recognising behaviour they have also seen (see other comment in this thread)
I am not an EA and do not read their forums. I think this explains why I didn't get much out of your review!
Did you think about the readership of the essay competition when you were writing? You're not going to win many 10/10's while dismantling a person's ideology for them.
I read and reviewed your essay. I think there's an element of (1) there, nothing you said struck me as especially insightful. I thought there was an irony in your main critique being that the EA movement mostly produces online critiques rather than engaging in the world to improve it. And you... produced an online critique?
Interesting, thanks! The first point surprises me - I thought rationalists were all about criticism. That they want new ideas more than feel-good.
On the second, so do you agree that this is a problem or not? I haven't seen anyone in the movement grapple with this, the closest I found was this: https://forum.effectivealtruism.org/posts/dsCTSCbfHWxmAr2ZT/open-ea-global
That discussion is still all about how to further EA causes. "EAG exists to make the world a better place, rather than serve the EA community or make EAs happy" says the Utopian social engineer, breaking some eggs for the omelette.
I don't agree with the third point. You have to go inside to tell people to go outside.
Also I'm not saying EAs don't engage with the world to improve it, in a sense I say they overdo that part. I'm saying EA is too private in its epistemology, but too optimising in its practice. The combination is dangerous
Your surprise that your critique didn't land well comes across as a little bit naive. Have you tried giving someone negative feedback in the real world?
3 is definitely in play; there's a lot of flippant language, even as early as when you say wind turbines are allowed "to shoot up", as if they're just weeds sprouting out of the ground. Very much has the feel of someone wanting to tweak the nose of the movement. The new moniker, the absense of evidentiary hyperlinks in your list of observations despite them being prevalent before and after, it has the tone of a hit piece. I'm dropping off before page 3.
I see, thanks!
I didn't want to tweak any noses, but definitely felt (perhaps too strongly?) that it is important and people don't realise it while writing the review.
Interesting about the observations, they were not meant as proof. I don't think there's any data for this, but please tell me if there is (ACX survey?).
Maybe I should have made my thinking explicit: The reader, presumably familiar with the movement, can validate the observations for themselves. If they don't ring true, the situation is less bad. If they do ring true, this is a problem, and the review is trying to show why. Given my personal experience, I would expect the latter.
It rings as focused solely on news-worthy parts of the movement. If you found a soup kitchen, you get an article for that. If it runs for fives years and serves lots of people, maybe you get another article if nothing else interesting happens in town. If one of your workers bonks a homeless child on the noggin with the serving spoon and hurls insults at them, congrats, you've made every headline.
Anyway, the movement was heavily influenced by Yudkowski's writing, and Yudkowski's writing has a lot of cultishness attractors (see: entirety of MIRI), so it's not really a surprise that there's cultish subgroups doing stupid stuff. As long as GiveWell, 80,000 hours, and other impactful groups stay sane, we should be fine. But I don't know what you expect me to do about the fact that other people come up with insane policy proposals and call themselves EA.
As for not trusting the mainstream, I just posted a comment in this open thread discussing a fact that was incorrectly reported by pretty much every journalist that touched the subject, because one of them talked to a scientist who said something vague and it was misinterpreted. Lots of people here are probably experts on some niche topic and have had similar experiences. If we see journalists getting stuff wrong in areas we know a lot about, why should we trust them for the rest?
As someone who's only familiar with the movement through what Scott and the commentariat post here, my accepting the points requires my trusting the narrator. The hostility of the writing style requires much more upfront evidence than a more neutral wording could get away with, as I would trust a more neutral piece to bring up its supporting evidence further in.
Interesting, I'm surprised it landed with such hostility
Wondering if I should go ahead and publish my review (which didn't make it into Scott's post) and thank the people calling it out in these comments, or keep to anonymity on the off chance the review gets elevated into a last-minute honorable mention position, as it did previously.
Either way, I'm really pleased with the quality of the reviews as a whole, and that there are loads of people in these comments calling out reviews other than these 10+ finalists and honorable mentions as their favorites. Obviously I'm sad mine didn't make it, but to see even one person name it as their favorite still warms my heart. See you next year!
Logistics continue to be a crucial element to warfighting.
https://jsargentr.substack.com/p/last-week-in-logistics-16-june-to
"Strikes against Russian industrial capacity continue. The Nevinnomyssky Azot fertilizer plant in Stavropol Krai, one of the top producers of ammonia and nitrogen fertilizers in the country, halted production after drone attacks. Another attack was charted in Orenberg, 1500km from Ukraine, and likely home to strategic missile base. In an effort to blunt the drone campaign, authorities cut mobile services to vast swathes of Russian territory. It ends up being a map of strategic priorities. Oops."
That's clever, but assuming that internet was shut down in whole regions, knowing that your target is in Sakha (area 3+ million square km) doesn't give you a lot of additional info.
My plan had been to review Fresh Fruit for Rotting Vegetables by Dead Kennedys, but other projects took enough of my time that once I got to the writing stage I quickly deduced I wouldn't be able to finish by the deadline. Perhaps I will do something on a related subject for a later contest.
Is your interest in it musically or lyrically focused? I find the second record "Plastic Surgery Disasters" to be more interesting in lyrical content as it contains numerous examples of fringe theories and sentiments that underwent political realignment in the past 40 years, and I like some of the more complex compositions, but I certainly understand the more punk purist appeal of the debut record.
FFfRV is more meaningful to me, personally, and its balance of edgy/dark songs like I Kill Children with political ones like Holiday in Cambodia would have also been something to consider and contrast. Due to its general reification there's also more written material on how it came to be.
I've gathered the courage and agency to start writing about Agency, and my path to understanding it: https://onlyluck.substack.com/p/good-luck
Maybe someone will find value in my exploration, and by subscribing, serve as an external motivator for me to write more.
Would love to see Bukele promoted to finalist! Thought it was really interesting, with analysis by someone living in the country of interest and well written. Have discussed it with several people in real life since reading it. Definitely keen to read and discuss further.
The Musk Algorithm was also great distillation of real world experience.
I really enjoyed it, learned a lot, and have shared it with others. The Miracle of El Salvador is an example to the whole world of how fast change can happen, even if it's failed for decades
I did think, however, that it played up the Dictator angle in a hype-y way, possibly for drama and readability. At the end, it says clearly that the author intends to just "monitor the situation" and maybe there's no dictator worries here. But ALONG THE WAY, it uses literary devices to make things sound much more 1-sided.
For example, it uses a dramatic and sudden "end of paragraph/section" after saying something about how Bukele immediately had a political opponent tried for corruption. Only significantly later in the article does it then add, "Oh, yeah, the opponent was literally on camera doing crazy corruption, so this was the reasonable and obvious legal/justice thing for anyone to do."
It would have been less of an emotional roller coaster to say this right next to the "mic drop moment" about Bukele going after the opponent, but more honest.
I lived in El Salvador for 2 years as a Mormon missionary during the highest homicide rate years, and am meeting/talking with lots of Salvadoranean people (eg at Kimbara bar) in San Francisco. As we face AI risk and hopelessness, I hope the example of El Salvador's Miracle can give us hope that outlandish claims about rapid results can actually be real and viable.
(I don't claim to know how to discern if the "extremely tough on crime" approach is worth the immense improvement in safety (not in detail), but a 91% approval rating makes me think I should cede judgment to those who have lived there.)
I really blew it on the review. I didn't read the contest note clearly and started work on an actual book review. DQ! I'll finish it up when I have time and get it in early next year.
Tried to play catch up with a review of Milgram's obedience experiment, in the context of popular fictional character archetypes, but spring is not a good time for writing in my line of work (Apple Grower). I'll finish and post that sometime, but I was off to a good start.
Pre-contest note, in anticipation of what I thought would be a book-review contest, I began a review of Tristram Shandy. When I discovered was a non-book review contest I tried to adapt my review into an argument that Tristram Shandy isn’t really a book. But it was too complex and I gave up the attempt. This is my review of that attempt.
Is Milgram's another of those "does not replicate" experiments, kind of like Lord of the Flies?
After reading the wikipedia article: no, Milgram's experiment seems on firm grounds.
Is "Lord of the Files" an experiment? I always thought of it as a novel.
It's propaganda, as taught (I imagine so is the Milgram experiment). If we wanted to laugh at the author and say "this story is so hilariously wrong that it's not even worth considering "good science fiction" as it doesn't get humans right," then we'd do that instead. Lord of the Flies mischaracterizes children in such a way that we're taught that "babysitting services" (like our current government schools) are essential to teaching children how to not behave like ... well, deer. Deer murder cats and dogs for no reason other than they're aggravating. Deer push their fawns (babies) into dangerous circumstances, before Mom gets anywhere near. Deer overpopulate without regard to population pressure (which rabbits don't do, mind). Etc. (yes, I may be biased. Our deer are exceptionally stupid here, in no small part because they eat toxic plants).
Milgram himself ran several replications of it, mostly because he had already built the lab. Some factors and results were stronger than others. For example, proximity of the authority figure was a big deal, people delivered fewer shocks when they were told to over a phone.
Punchline is this study started a wave of reform in Psych testing, with new ethics codes that protected subjects from the kind of stress and emotional harm that were evident in his work. So, replicated? can't say.
here's the link to his contemporaneous documentary film
https://youtu.be/rdrKCilEhC0?si=QFqZ9X2taBoIhMED
In 2012, Australian psychologist Gina Perry investigated Milgram's data and writings and concluded that Milgram had manipulated the results, and that there was a "troubling mismatch between (published) descriptions of the experiment and evidence of what actually transpired." She wrote that "only half of the people who undertook the experiment fully believed it was real and of those, 66% disobeyed the experimenter".[26][27] She described her findings as "an unexpected outcome" that "leaves social psychology in a difficult situation."[28]
Pulling this from wiki. I suspect this was designed (and paid for) to provide people psychological cover for "just following orders" -- "everyone else does it" sounds good on paper.
Uhm, this is a pretty misleading excerpt. Why don't you also cite the paragraph directly below it?
"In a book review critical of Gina Perry's findings, Nestar Russell and John Picard take issue with Perry for not mentioning that "there have been well over a score, not just several, replications or slight variations on Milgram's basic experimental procedure, and these have been performed in many different countries, several different settings and using different types of victims. And most, although certainly not all of these experiments have tended to lend weight to Milgram's original findings"
Also, I think there are a lot of reports of participants of the original experiment which also confirm it beyond the pure publication. My understanding is that this is pretty well-established. Other than, say, the Stanford Prison experiment or the Robbers Cave experiment.
EC/VC lawyer here. Happy to connect if you still need help drafting/structuring.
Good contest, there were some really thought-provoking entries in the ones I read - and I got to very few of the ones that ended up being finalists, so I still have plenty more to look forward to.
I wrote the Skibidi Toilet review. I had a ton of fun writing it, glad it got some positive reception and informed some people about one corner of the vast and weird landscape of modern storytelling. :) I'll post a polished-up version to my own blog soon. (Update: It is here https://eukaryotewritesblog.substack.com/p/skibidi-toilet-review )
To the commenters who vocalized a disdain for learning anything about the topic or even being vaguely compelled to think about it: Hey, your time is your own and you should spend it as you see fit. Maybe next year, you can write a review of rank cowardice so the rest of us can decide if it's right for us.
After reading some comments, I read your review, and it was good enough that I actually watched Skibidi Toilet from the playlist provided. The series is interesting, especially with the background provided by your review., though it's kind of hard to follow because it seems to follow its own physical laws, like getting your head cut off doesn't kill you, except when it does.
You did explain that the technology is never explained, and I found that a bit irritating, too. In the 4th episode, somehow the people in the restaurant get turned into toilet people by no visible means. In 41, the toilets are compelled to commit suicide, but this is apparently never again used against them.
I thought your review was excellent. I thought you had done the most out of anyone to present the worst possible topic in the best possible light, but Skibidi Toilet was always going to be a tough sell.
Eagerly awaiting the next time someone disparagingly mentions Skibidi Toilet in a discussion about The Youth of Today and I can "Well ackshually" them
Thank you for the review, it was entertaining and informative! The haters are missing out (on the review, not necessarily on Skibidi Toilet).
Yours was the first review I read (bc scrolling and... "there's a review of TOILETS?!?!"). It was an interesting choice, and I thought your review gave a lot of context to an extremely weird meme.
> Maybe next year, you can write a review of rank cowardice so the rest of us can decide if it's right for us.
Savage (also, hilarious)
I enjoyed your review, thanks for writing it. I was hoping its author would comment so i could say that.
For what it's worth, I read your review immediately upon seeing it in the table of contents, and enjoyed it thoroughly.
Amen.
Iran has had their militarily capacity greatly degraded in recent weeks/months. Setting their nuclear program aside, we know that their air defense and missile launching capabilities have been largely curtailed. However, if somebody thinks this is a great time for regime change coming from within--which is the only place it should come from, self-determination being near sacrosanct to me--I have this thought. The security and police apparatus that is so effective at keeping the Iranian people from succeeding at their protests is roughly intact.
So, any notion that we are suddenly in good position for a popular counter-revolution in Iran had better buckle-up, because it will be a rough ride. Sure is fun and easy to bomb targets--and eventually that means people--but intervening to help a local/national uprising does not appear to be on anyone's menu. Presidents that say "go ahead, we have your back" have not been credible, and so should be taken with salt.
Inviting such an action through a preemptive leadership decapitation action would be a black swan sitting on a black box. no-one knows what will happen squared. Not a great comp, but no one seems very happy about Libya post-Ghaddaffi (mostly due to Wagner/Russian interference, IMO), but Iran is a much higher base-rate competence country that Libya was, so who knows.
Still, don't do it.
Relevant prediction markets: https://polymarket.com/event/khamenei-out-as-supreme-leader-of-iran-in-2025 and https://polymarket.com/event/will-the-iranian-regime-fall-in-2025
Currently at 37% chance Khamenei isn't supreme leader by the end of the year, 18% chance that the Iranian regime falls by the end of the year. Since the former is implied by the latter, would it be reasonable to assume that there's a fairly high chance that if Khamenei is gone then the government will be overthrown? Or am I mixing up causality somehow?
I don't think this mathematically follows. There could be a world with a 19% chance that Khamenei is killed by an airstrike (or dies naturally), and conditionally on that, a 100% chance that the regime survives. Plus a chance of 81% that Khamenei survives the year, and conditionally on that a roughly 22% chance that the regime falls. All this is mathematically consistent with the numbers you gave, but then there is zero chance that the regime falls if Khamenei dies.
All this is purely mathematically, of course, and not a model of the real world. But we can say (taking the numbers at face value) that the probability of the regime falling if Khamenei dies is AT MOST ~50%. Because there is at least a 19% chance of "Khamenei dead, regime in place" and at most an 18% chance of "Khamenei dead, regime falls".
Ah, that makes sense.
Clarification about the "self-determination" you consider sacrosanct: would you object to the US arming a few dissident groups inside Iran with vast amounts of guns, tanks, fighter jets, and missiles to do with as they please?
Fighter jets come with a huge logistics tail. Giving them to "dissidents" isn't really credible.
Sure, but I'm not suggesting this operation be covert or deniable in any way: this would simply be kickstarting a civil war, and using the rebels to fight a proxy war. What is the part you think isn't credible?
(Also, I'm not actually suggesting this pointless procedure of adding extra steps to the regime change operation; I don't give a shit about "self-determination" in the sense of preferring local tyrants over foreign ones.)
No
> However, if somebody thinks this is a great time for regime change coming from within--which is the only place it should come from, self-determination being near sacrosanct to me--
If it comes from without, how is that any less self-determination than what they have now? Iran isn't a real democracy. Anyone living their is having their life determined by someone else as it is.
Yeah, this is a good point. Reminds me of Iraq -- of all the negative thinks one might say about the invasion, "but imperialism" is the weakest one, since the result was an increase in self-determination (which is not the usual result of "imperialism")
Okay, let me address your 3 points in reverse order. 3: Well, all societies have either rules or anarchy, so yes, other people have a lot to say about how you live. Not much for me to disagree with there. 2: Iran not a democracy? Not generally thought of as one, and yet...they do have elections that have actual political consequence. Yes, the Mullah branch controls who is allowed to run, but within that scheme there are still extremists and moderates--the current PM being the most moderate candidate in the last election. They also have women with the right to vote since 1963! Not bad. On the whole, I rate them as slightly less democratic than Israel, which would have a considerably higher grade if so many of their residents weren't denied citizenship/suffrage. But I digress.
3: They got pretty pissed off in Tehran when a woman was murdered in police custody. They went to the streets and made a very big deal of it, and were beaten, jailed and even killed for their trouble. These people want something better, and I hope they can find their way to it, although right now might not be the ideal time. I will say that in the photos of Iran after the bombings, of the people I see only ~half the women have their hair covered, the rest don't like that.
I once heard on some news show a dissident in Syria--this was 10+ya--say that Assad would fall when Syrians made him fall, and any other interference was quite possibly disastrously unhelpful. His preference has aged well.
> but within that scheme there are still extremists and moderates--the current PM being the most moderate candidate in the last election.
Suggesting that they want someone more moderate than the Mullah branch allows.
I admit I haven't done any real research into that. Just a quick google. But that response isn't the defense you seem to think it is.
"that response isn't the defense you seem to think it is."
Not sure just what it is that you think I am defending. What do you think I am defending?
I'm not really coming at this from an ideological stance. If you think I am, run it by my and I can confirm or deny. If you want to know what I think--about Iran, American Power, the Iraq war, The Afghanistan war--you can ask; you will get a better answer than if you guess.
It sounded like you were trying to say that they have a halfway function democracy.
Oh, okay. That sounds about right, actually. In terms of functionality, Israel is far superior. Other than some ongoing assaults on the independence of the judiciary, it certainly serves as a working, stable democratic government. Iranian civil society is utterly hobbled by the big system-small system dichotomy of the Appointed mullahs and the elected representatives. The elected ones don't really have a fair shot--though they do often try! More's the pity.
Ah yes the fall of Syria,… to Al Queda. Good work there.
That, sir, is an untruth.
where's the lie?
I would accept 'former Al Quaida' and even that is a bit of a stretch, you make the alliances you can in a terrible war. AQ ia not shy, they would be glad to claim they were running the gov. of Syria. That has not happened--tho not a dispositive on its own--and they are not running the country as such under a different name. A great gov? I dunno, we will see. They have chased Russia out and degraded their ability to make trouble, but that doesn't mean he isn't a semi-secret AQ agent. Trump recognition far from the worst thing he has done, policy-wise
That's probably why Israel has been targeting low to mid level police and police leadership recently. I understand they have been intentionally assassinating the equivalent of the local police captains, though I haven't seen a lot of detail on it.
Well, shit...They are setting up for a decapitation, if so. What are your sources?
It was an unsourced discussion on another forum. I do not have original posts about it, unfortunately.
Today there were strikes on the Basij (a group of state-sanctioned gangs who beat and rape women who protest the regime), and a notorious political-prison: https://www.timesofisrael.com/liveblog_entry/israel-confirms-hitting-basij-hq-evin-prison-destruction-of-israel-clock-in-tehran/. Not sure about police-assassinations...
I thought this was an excellent take-down of Curtis Yarvin's recent allegation that the Allies starved hundreds of thousands of German prisoners of war to death in 1945:
https://arctotherium.substack.com/p/right-wing-pseudohistory-part-1
Yarvin's original X thread was done in excellent Yarvin form, replete with rhetorical wink-winks and "if you read any real history, this wouldn't be a surprise" vibes. I admit to hoping it was complete BS, but trying to hold my judgement for a while, and I'm glad someone did what seems to be a thorough and even-handed job of refuting it.
Is this article accurate?
https://en.wikipedia.org/wiki/Disarmed_Enemy_Forces
I agree that Yarvin is, oh gee, let’s just call him a bozo for the sake of brevity, but arctotherium goes on to carp about ‘the elevation of the Tulsa race riot to a massacre’ in his conclusion by citing an article in The Unz Review which is edited by Ron Unz, who per Wikipedia is an antisemitic holocaust denier and a white nationalist so I won’t take arctotherium’s word as gospel.
I would not trust an article by Ron Unz to tell me much of anything, not because of his politics or morals, but because he is very careless with getting the facts right and very inclined to reach for a cool conspiracy story.
My impression is that while Ron Unz is personally a crank, his stable of writers includes a weird mix of fellow cranks and people actually interesting to read who care about the truth (or at least used to - I haven't paid much attention there since both Audacious Epigone and Anatoly Karlin left). I think that Steve Sailer is the only person I can think of who is both an Unz Review contributer and a regular commenter at ACX, and hope someone wouldn't dismiss what he had to say just on the basis that he also wrote there.
I had a similar reaction. It made me curious about the rest of the blog, since that choice was so wildly at odds with the well-argued and well-sourced article that preceded it.
So I clicked on the first article in the archive that caught my eye, which turned out to be nothing short of appalling. It was a self-righteous bout of verbal diarrhea about how immigration is ruining Canada and how the U.S. ought to step in to coerce those misguided Canucks onto a better path. As a Canadian-American who has lived in both countries (and fully intends never to go back to the U.S.), I can now confidently say that I would no piss on the author if he were on fire.
If I had the option of dual citizenship the Sunshine Coast of Vancouver Island would be appealing right now.
Yeah, that was my reaction when I saw his reference to Tulsa. I clicked on the article, but no way was I going to read it with all my BS alarms going off. Thank you for going the extra mile, it smelled that way to me.
The German POWs in Texas and Oklahoma had such a pleasant time working on local farms that some returned for reunions, for years.
Yarvin is very capable at *sounding* like he's deeply knowledgeable about a given topic. So it helps to apply actual logic to his nonsense.
.....
This is bad. The author is bad. To be fair, I *think* Yarvin is trolling.
We know that former Nazis were treated incredibly poorly and anyone familiar with sources from the period would be aware of it. My personal familiarity of this comes from the diaries of John Rabe, the Nazi who is credited of with personally saving 250k Chinese during the Rape of Nanking. You can read his wiki here (1), watch a trailer for the movie about his life here with Steve Buscemi (2), and buy his diaries under the title "The Good Man of Nanking", which I would recommend.
We know he was basically starved to death. We know this from his diaries but we also know his situation was so dire that in 1948 the mayor of Nanjing traveled through Switzerland to Germany and the people of Nanjing sent him food until his death. Allow me to repeat, a ~64 year-old humanitarian was essentially starved so badly that the people of Nanjing China, during the Civil War and Maoist period, were sending him food. If that was happening to an old humanitarian Nazi bureaucrat...yeah, I can imagine some pretty bad stuff occurring to actual soldiers.
Yarvin is bad here because I *think* he's basically trolling and this is Trump's Hatian cats all over again, where there probably was not a million person execution of captured German soldiers but their living standards and treatment was probably pretty horrific and now we're all talking about this. Yarvin *probably* wants this because it serves his purposes for the Skeletor of the 20th century to be Communists, not Nazis.
This response is worse. It presents no actual evidence except official statistics provided by the very people accused of genocide/abuse, cribbed from a book written by some other guy. Give Yarvin this, I believe he's actually read and interacted with direct sources and now he's spinning/trolling to get clicks and influence. This bozo just cribbed a book someone else wrote about it with, I don't want to say no evidence, but certainly not persuasive evidence. You just...if someone accuses the Soviets or the Maoists or the Japanese of genocide, it is not persuasive to show official Soviet or Maoist or Japanese figures declaring that they totally didn't commit a genocide. The bar is low but you have to actually cross it.
Please don't share stuff like this in the future. It was dumb and it made me angry.
(1) https://en.wikipedia.org/wiki/John_Rabe
(2) https://www.youtube.com/watch?v=4IPnkLA5__Y
John Rabe was not in any of the prisoner camps. He was starving as a civilian, and as the author made very clear a *lot* of people were starving and going without food in Europe in 1948. John Rabe going hungry as a civilian in Berlin is not evidence that hundreds of thousands of Germans were starved to death in Allied prisoner camps.
The author makes a lot of good points. If this happened, why doesn't anybody claim that it happened until 1989? Why did the West German commission whose sole job was to track down every German that died during or after the war not find hundreds of thousands of starved prisoners? Where is the evidence that such a thing occured? Yarvin doesn't seem to present any.
Lots of details like this get classified and only become public decades later. Operation Keelhaul is another good example of this kind of thing, which seems like it only became known in the 1970s.
https://en.wikipedia.org/wiki/Operation_Keelhaul
If an old bureaucrat and family were being starved, why wouldn't that count as evidence that former Nazi soldiers were starved? Like...the Americans aren't going to starve the former SS guys but they will starve some guy's daughter? From memory I think one of them was 10 and another was a teenager at the time but that's old memories from the diaries.
It's not uncommon for events like the Holodomor or Dresden bombings to remain controversial, unproven, and generally uninvestigated outside of extremely niche communities for decades.
Again, what probably happened is that a lot of former Nazis got treated extremely badly after the war. How bad? Who knows? Also, not to be too much of a jerk, who cares? It would have hardly been the worst crime of the period, it probably wouldn't have cracked the top 10 even if they had killed a million. Yarvin is *probably* overselling it for political ends and this guy is underselling it using some of the worst logic ever.
The old bureaucrat and his family wern't "being starved", they were just starving. Nobody was keeping them from getting food, there just wasn't enough food to go around and everyone who wasn't a German got priority on food aid. Which is the author acknowledges and nobody disputes it. This widespread hunger is not the same as the claim that hundreds of thousands of German prisoners starved to death and the Allies covered it up.
The Holodomor was controversial, but people were reporting on it as early as 1932. Similarly, the Associated Press was reporting on the destruction of Dresden within weeks of the actual bombing, and there was a lot of people call it a war crime at the time. Yet we have no contemporary claims of mass deaths by starvation in Allied prisoner camps, not for many decades.
I agree that what probably happened is a lot of former Nazis got treated extremely badly after the war. We know this because we have a ton of evidence to support that fact: personal testimonies, newspaper articles, written government policies, etc. We don't have any of that for Yarvin's claim.
No, he was being starved. It's very explicitly noted in his diaries and it so shocked the people of Nanjing, a people who had literally lived through the Rape of Nanjing, that they sent him food to keep him and his family alive for two years during their own civil war. We have, to the best of my knowledge, multiple accounts of his starvation by independent sources and no contrary accounts. Nor was the mechanism confusing, he and his family were explicitly and covertly barred from any employment until his death, with contemporary accounts noting them living on wild seeds.
Nor was the Holodomor widely known. It was denied in the pages of the New York Times as it was occurring (1) and this was endorsed by FDR at the time for fairly obvious realpolique reasons.
How badly the post-war Nazis were treated and how many died is a source of ongoing controversy, especially since it's so politically charged. As an example, the best estimate of the number of civilian casualties during the Dresden bombings has varied from 135k to 250k to 25k (2) in my lifetime alone. A lot of that is political/culture war stuff but a lot of it is that it's...just actually really tough to estimate.
Without any specific insight into this particular controversy, the idea that out of 3 million Nazi soldiers, that only a few thousand died in US custody while contemporary reports of civilians at the time being in dramatically more dire straights is...not super credible and would be kinda horrific if we actually did treat Nazi soldiers better than German civilians. Conversely, a couple hundred thousand dead is...plausible but highly unlikely. But if the Dresden bombing casualties can range from 135k to 25k, then it makes sense that the Nazi....whatever this is, the Rheinwiesenlager could vary from 30k to 300k.
Having said that, for all his trolling, at least Yarvin is familiar with the direct sources and providing 3 party reports. The article originally shared doesn't, it's bad for reasons well elaborated on already.
(1) https://en.wikipedia.org/wiki/Holodomor_denial (see Duranty)
(2) https://www.nbcnews.com/id/wbna26977893
Rabe was officially de-nazified on 3 June 1946, but grateful Nanjing residents sent him food packets starting in 1948, presumably meaning he was still starving at that time. At least from 1946 to 1948, it wasn't the Allies preventing him from getting a job.
We seem to have different definitions of "being starved" (being banned from working is not the same thing at all in my eye as deliberately starving someone) so we'll just have to agree to disagree on that. The point remains that one person being kept from working and as a result starving is not evidence that hundreds of thousands of German prisoners starved to death in Allied camps and the Allies covered it up. It is evidence that former members of the Nazi party went through bad times after the war, which was never in dispute.
The Holodomor was not widely known but it was *known at all*, known to the point that the NYT put out articles denying that it was happening. Where is the equivalent for the mass starvation of prisoners? Where's the newspaper article saying that such claims are exaggerated or untrue? We don't have a thing for this until the 1980s.
We probably did treat Nazi soldiers better than German civilians when it came to food because the US Army was responsible for feeding their prisoners but not that responsible for feeding random civilians. For one thing all the prisoners were concentrated in camps where you could send food shipments to and distribute them, while the civilians were scattered all over the country.
I read it too, and there are reasons for concern with this author (like, defending residential camps, calling Tulsa a 'race riot?). But the most compelling stats he drew from came from the West German government of the 60's and 70's.
That's what it was called for about a century after the event. Which way do you prefer it be revised: up towards "genocide", or down towards "mostly-peaceful protest"?
Shankar and Melvin, you are reading me wrong, which is probably the result of my bad writing. An author above, during a lovely takedown of Curtis Yarvin, dropped a reference to an article 'debunking' the events that happened in Tulsa. To wit: he was changing the events from a massacre to a 'race riot,' in a decidedly both-sidesy sort of way. I didn't feel like giving it more of my time, attention, and I guess I already failed there, but I don't like people trying to airbrush a massacre away. Ditto with the Residential Schools. They happened and it was bad that they happened, and its okay to say that.
In what sense was Tulsa not a race riot?
Shankar and Melvin, you are reading me wrong, which is probably the result of my bad writing. An author above, during a lovely takedown of Curtis Yarvin, dropped a reference to an article 'debunking' the events that happened in Tulsa. To wit: he was changing the events from a massacre to a 'race riot,' in a decidedly both-sidesy sort of way. I didn't feel like giving it more of my time, attention, and I guess I already failed there, but I don't like people trying to airbrush a massacre away. Ditto with the Residential Schools. They happened and it was bad that they happened, and its okay to say that.
"Race Riots" isn't an inherently both-sidesy term. Most race riots in history have tended to be entirely one-race sorts of affairs.
The problem with rebranding the "riot" as a "massacre" (which is an extremely recent phenomenon, wikipedia changed it on Feb 7 2020 https://en.wikipedia.org/wiki/Talk:Tulsa_race_massacre) is that nothing about it sounds massacre-ish.
The defining aspects of a massacre are its speed and one-sidedness, and this particular event doesn't match that. You could call it a very bad riot or a very small battle, but not a massacre.
Yes, the West German government's retrospective in the 60's and 70's is more trustworthy than the American's. Still:
#1 Independent 3rd party accounts are better, and Yarvin provides those in his...sigh...Twitter thread here (1). An account the original author ignores and seems to get confused with a report from the Army Medical Department. (the Andersonville Prison reference)
#2 Postwar Germany was politically...not the most unbiased of locations. Holodomor denial from the Ukranian Soviet Socialist Republic is better than the USSR but...ya know, not exactly unbiased.
As an aside, I tried to see if he referenced the red cross thing anywhere, I don't think he does, but he does claim that Pope Benedict was in one of these camps and...no? Certainly not my area of expertise, correct me where I'm wrong, but Benedict's Wikipedia shows him in Neu-Ulm and Fliegerhorst Bad Aibling (2) but those are not in the list of 19 Rheinwiesenlager camps Yarvin is discussing (3).
To be clear, if I have not, I have not heard of this particular claim before. It seems similar to the bombing of Dresden or the Holodomor. Old events that are highly charged, highly ambiguous, and where it's likely something bad happened but how bad and how intentional it was is quite vague. Both events also that kinda lurked in academic corners until they became fuel for the culture war. I doubt Yarvin's specific claims but the abuse of Nazi and German people pops in up in every contemporary account I'm familiar with and also...just makes too much sense. That's just what victorious armies do. What I do resent and get angy about is just bad argumentation.
(1) https://threadreaderapp.com/thread/1930232102986842384.html
(2) https://en.wikipedia.org/wiki/Pope_Benedict_XVI
(3) https://en.wikipedia.org/wiki/Rheinwiesenlager
Sounds like the guy is out of ideas and pretty desperate to get attention.
EDIT: Or maybe, if this is intentional, he decided to target a dumber audience than previously. Now the ideas are less abstract and more dramatic, and there is less need for plausibility.
I wrote the Deathbed Ballads review. It didn't make the cut, but least one person seems to have liked it, which I appreciate. Last year, I reviewed Martin Luther's "Bondage of the Will", which looks like it narrowly missed the Honorable Mentions, and I got some good feedback on it after the fact which I still plan on using to make a revised version to publish elsewhere.
Some other topics I considered reviewing this year:
"Taxman" by The Beatles. I wrote a paper in college analyzing this song and tracing out specific references in it to elements of the UK tax system in the 1960s. I think it would only need minor reworking, but I thought the same of my college paper on Bondage of the Will last year and had wound up redoing it from scratch. I decided against it because it was much more of an analysis than a review.
"Identity", a YouTube video where the creator, Abigail Thorn, comes out as a trans woman to her audience after already being established as a moderately big-name YouTuber before starting her transition. I watched it a few months into my own transition and found it very moving and applicable. I think it's very well-made and has a number of subtle details I'd be interested in talking about. I decided against reviewing it mostly because I didn't think I could do so anonymously: I have been very active in comments discussions here about trans issues and have cited Identity more than once in those discussions.
Analysing songs from the American Civil War as a corpus for hints at how soldiers thought about the conduct and motivations of the war. I had originally chosen this as my topic and started writing it, but abandoned and pivoted to what I hoped would be a lighter topic because it was very quickly turning out to be too verbose and research-heavy. Not sure how successful I was (gestures vaguely at my 14000 word actual submission). It was also shaping up to be too much an essay and not enough a review.
One other topic I won't share publicly. The reason I didn't review it this year is because it overlaps substantially with a book review I want to do in a future year and I think it works better if I review the book than later than if I review one major topic discussed in the book now.
I still remember the review of "Bondage of the Will", and found it really impressive! I haven't read the Deathbed Ballads review, but may still do.
Thank you! If you do read Deathbed Ballads, please let me know what you think.
I did read it now. To be honest, I found it hard to connect to it. A good part of that might be myself. I am not a music person. I don't listen to radio, and never have done. I thought I might try it because I knew one of the songs (France, and that actually in many versions). But without knowing the others, it was just a very long list of parts that I couldn't relate to.
I found the topic also a bit less, well, relevant. For Luther, I was already going into the review with the premise that he is an important person that I would like to understand better. For the songs, I would have needed to be convinced that they are important, and there wasn't so much convincing done. Perhaps they really are not so relevant, and perhaps they are, but in the latter case I don't know about it.
I still think it was objectively well-written. You made the most of the topic, and I do buy you conclusion with two origins and found the table interesting. So to me the problem was not in how the review was done, but rather what the topic was.
That's good feedback. Thank you for taking a look, and I appreciate the critique.
I loved that, and voted it for it. The green fields of France is need a great anti war ballad.
"And the Band Played Waltzing Matilda" is another first-rate anti-war ballad, also written by Eric Bogle about a different facet of WW1.
That was one of the few reviews I read, thank you for the write-up "Tie me Kangaroo".
Just too much overall material, I was overwhelmed.
You're welcome.
>Just too much overall material, I was overwhelmed.
Yes, that's completely fair.
Taxman. 1 for you 19 for me. Pretty easy to understand complaint. Even for a kid who had yet to pay any taxes in 1966. One of George’s pre-metaphysical songs. No sitar in that one.
Can anyone here who supports the Israeli and American war against Iran explain to me your reasoning? I'm kind of baffled by the support for it. AFAIK the official rationale of Israel/America is that Iran is building a nuclear weapon and so they're engaging in self-defense to stop them, but U.S intelligence says they're not building nukes https://apnews.com/article/us-iran-nuclear-intelligence-b506d130e474c00f6bd653d3d5a8d31a . In particular I would like to know
1. Whether you think Iran is actually building a nuclear weapon
2. Whether you think the war against Iran is legal under international law
3. Whether you think diplomatic alternatives to war have been adequately explored
4. Whether you think regime change is viable and desirable
1. Yes
2. Maybe?
3. Yes
4. No
Basically, with all of Iran's proxies in the region declawed at the moment, this was the best time to act with impunity and Israel took it. Killing their scientists and demolishing much of their nuclear research infrastructure hopefully has set them back far enough that they won't be a player. The US intelligence leak to which you're referring was being treated skeptically. The Mossad clearly believed otherwise, as their attacks had no other clear objective and did not seem tailored towards regime change since they left the non-military authorities mostly untouched. The whole thing seems to have been crafted to clear the way for the action the US took.
I don't think imposing regime change from the outside is viable, or necessary, or desirable. I didn't really care about Iran in any way other than preventing them from having a nuclear weapons program. If I had to prefer one nation be the regional hegemon down there, I'd certainly prefer the Saudis over the Iran/Qatar axis, and I'd prefer secular monarchies over Islamist theocracies, but as long as they've been de-fanged it's not really something I'd want to sacrifice US war material and lives to get involved with and that would likely be counterproductive anyhow.
Big fan. This has been the cause of a very very large 'pro trump' update for me.
I was broadly in favor for pro-democracy reasons (I felt america had gotten captured by bureaucracy and it was very difficult for voters to effectuate change, and here was a guy who very publicly said what he would do, was non-ideological, and would try to get it done) but disliked trump in particular for usual reasons, and here we have some actual leadership / good results. I claim to have supported it before 'it all worked out clearly better than we could have reasonably hoped, which maybe it hasn't yet but probably has.'
1. I think Iran was trying to get as close as it could to being able to make a nuclear weapon, between a week and year from it. They had 60% enriched uranium that we know of. Given their history of sponsoring terrorism against us and our allies, already a very big reason to not want this. Their strategy was approximately 'get to a week within a nuke, while having many many missiles, use that as deterrence while funding terrorism.' Given how much Israel weakened them by destroying their proxies in syria/lebanon/yemen, I think it was clearly a good idea.
2. I...don't care about international law, at all. International law says if someone declares credibly 'I will gain power, and once I do I will kill you' you have to wait for them to get sufficiently powerful that they feel ready to go to war, to try to stop that. It's horrible tool in this context. In theory, great for minimizing casualties in 'reasonable wars between equal powers who might have wars in the future and it's a repeated game', horrible for religious crazy regimes with nukes or nuclear ambitions.
3. I think diplomatic alternatives to war were sufficiently explored to discover that the difference between what the Iranians would agree to, and the expected result of bombing their stuff, was extremely large and in America's favor. Theocratic dictatorships are extremely bad at modeling the will/capacity of their enemies. There's just...very little you can do in this situation. Iran for religious reasons wants to defeat their enemies and in so doing redeem Islam by proving they have gods favor. It's hard to negotiate with that unless they are extremely practical after you prove 'they are not going to succeed.' So you need to 'succeed' sometimes.
4. Uh, I think regime change of the form 'the 80% of domestic civilians, and the 98% of iranian diaspora that hate the evil regime rise up and install a secular democracy' would be great. I think that's unlikely to happen because the regime is so good at repression. It's also not our problem. Ultimately, people need to liberate themselves. It's also very chaotic.
We did what we wanted to do: we blew up their nuclear program and a substantial portion of their intercontinental ballistic missile program. That's good enough for now.
The most popular beer in Iran is BEER. Now, how's that for repression? There's also underground dating scenes where middle class men and women get together and dance in Tehran. How's that for repression?
(Bottom line: learn a little more about a country before you assume it's Communist china -- do as we say or you're re-educated. Iran is a very delicate country, full of many twists and turns. It's fascinating, in its own way).
How many women have been tortured and killed by the police forces for showing their hair or skin in public? How many non violent protesters have been imprisoned.
Yes, I agree head scarves are not currently being enforced on the streets of tehran. I assure you what you have stated is not new information to me. Also China has earned the trust and respect of a majority of its people, Iran has done the complete opposite.
There were (circa a few years ago,haven't checked recently) 12 different ways out of Iran. Some were conservative some were liberal. If you wanted out of the theocratic regime, you could get out. Pretty much anyone could. I'm going to say that women getting tortured and killed is, of course, bad. Is it worse than in the rest of the Muslim world, where a woman cannot get married without her father's permission, and cannot get divorced without her husband's? (In case you missed it, this is prostitution colored Muslim -- and it's still fairer than marriage in the rest of the Islamic world).
China has earned the trust and respect of a majority of its people? I've seen illegals in America -- folks that China has deemed murderable on our soil. They did not look happy, they looked very very afraid. They are in danger of their lives at all times. I can say more about China, if you want.
This is a weird conversation. You're not telling me anything I don't know. I've been to China and am reasonably familiar with its people. I don't even understand what factual point you imagine I might disagree on with you.
1. Does it matter if they're building one this very moment? (This is not to discount the evidence provided by others that they are.) The mere fact that a theocratic regime has taken regular steps towards nukes is terrifying enough and surely unsustainable.
2. Iran's *existence* isn't legal under international law. Doesn't the latter include a whole bunch of things about human rights and freedom of religion? Or did I dream that? Is funding interstate terrorism legal under international law? Is having an official regime-defining ideological goal of wiping another country off the map legal under international law? If any of these things aren't legal under international law, is there a world police that's eventually going to actually enforce this "law" and why they are taking so long?
Does anyone actually utter the phrase "international law" with a straight face?
3. I mean, Iran spent several decades *not* being bombed and...did it stop funding terrorism? Did it stop talking about wiping Israel off the map? I don't even understand the coherence of this question. If you think there is any chance of negotiating with Iran now, it can surely only be because it's under direct military threat. Which is...the very thing you're objecting to.
4. I suggest you look an Iranian woman, an Iranian Christian, a person who's been in an Iranian prison, in the eye and tell then you have doubts about whether regime change is desirable. I dare you.
Viable is an entirely different question. I'm certainly open to arguments about whether regime change is viable or possible or counterproductive. But (and I have the same attitude to restrospective analysis of Iraq) such questions are in an *entirely different universe* to suggestions that even if regime change were possible, viable and low-risk or zero-risk it would still, somehow, not be right. The former is sensible and logical. The latter is total moral bankruptcy and moral cowardice of the highest order.
1. Whether you think Iran is actually building a nuclear weapon
> I don't know. I give it more than 10%, and think that the rest of their international policy is enough at least to remove their "deontological protection".
2. Whether you think the war against Iran is legal under international law
> I care about ethics. I would love to live in a world were "international law" is really a thing - but there is neither a leviathan nor established expectation of reciprocation
3. Whether you think diplomatic alternatives to war have been adequately explored
> No idea. Low prior that a diplomatic alternative is possible given the positions of the parties.
> 4. Whether you think regime change is viable and desirable
I think it is very possible that the regime would fall given the right encouragement, though not something to count on. And that it is overall desirable - both for its own sake and to make the best alternative to agreement worse
I hence strongly support atacks against tyranny-preserving organizations there as long as there are other reasons to fight. It wouldn't be my way goal though, because I want to avoid ugly fight to the (regime) death, and won't pay the price of ground invasion
on (3), "No idea" about diplomatic alternatives to war seems incredibly cavalier to me.
A prior, do you not think that exploring diplomatic alternatives to war is important? What if an extremely feasible, easy, low-cost, highly likely to succeed diplomatic route exists?
I agree that it is important. I have no idea not because i don't care but because I lack access to the information, as do you. I only know the public positions and actions of the parties involved
1) Yes, I think they intended to build nuclear bombs. 60% enriched U-235 makes no sense for a power reactor (typically 5% enriched), and looks only like a step to bomb-grade U-235. And the IAEA said Iran has broken nonproliferation agreements.
2) I'm fairly disgusted by international legalisms. Maybe commercial international law has some value.
3) Iran was great at diplomatic delaying tactics. I view these as delaying tactics while they enriched to bomb grade.
4) (writing from the USA) I don't think _we_ should be trying to change Iran's regime. The current regime has been chanting "Death to America" for 40 years, but, even if we _had_ the power to oust their current regime, (a) We didn't do well with that in Iraq or Afghanistan - maybe we've lost the knack for it (b) Trading the devil you know for the devil you don't know often yields buyers' remorse. It isn't as if we could just kill Khamenei and have Iran miraculously transform into a Persian Switzerland.
The very first sentence you linked to is "A U.S. intelligence assessment says Iran is not pursuing nuclear weapons at the moment but has ramped up activities that could help it develop them."
Can anybody explain to me what the difference is? As I understand it there are no other uses to which the super-pure uranium they are acknowledged to be producing can plausibly be put to. How exactly is that "not pursuing nuclear weapons"?
I've read Democrats whining that there was no rush, that Iran is months away from having a nuclear weapon. To which I just drop my jaw and say, "Months!?!"
One obvious difference is that it may well have been possible to negotiate them off that course previously. It certainly won't be now. If the U.S. is going to treat them as if they are nuclearizing one way or the other, then they're certainly better off rushing for a nuclear deterrent as fast as they can.
The difference is that Iran is using the uranium as a bargaining chip: "if you don't drop these sanctions, I'll enrich to 70%! Then to 75%!" By inching ever closer to the weapons threshold, they hope to extract concessions from the West. If they actually build a bomb, they gain a military deterrent but lose a diplomatic bargaining chip.
I think that one habit people really need to get out of is calling governments by the names of their countries, and vice versa. We need to distinguish between "Iran", which is a huge country of 90 million people, and "The Iranian Government", which is a random bunch of unelected thugs that happens to control a capital city.
The morality here is not in question -- when you have an unelected government ruling over a country, it's moral for absolutely anybody, no matter how foreign or humble, to kill any element of that government.
I disagree with the implication that elections would make the thugs that end up in charge any less moral to kill.
I strongly agree with this:
"I think that one habit people really need to get out of is calling governments by the names of their countries, and vice versa. "
But find this to be absolutely unhinged:
"The morality here is not in question -- when you have an unelected government ruling over a country, it's moral for absolutely anybody, no matter how foreign or humble, to kill any element of that government."
Like, setting aside the fact that "any element of that government" could just as well refer to people paving roads or distributing food as secret police. Even if we're just talking about the top brass, this is unhinged. Make no mistake, I have very little sympathy for those who abuse their power, regardless of whether they got that power through heredity, military force, business acumen or winning elections. But even a very modest knowledge of history should thoroughly disabuse one of the notion that "just kill the bad people" is a workable recipe for improving the lives of the average person.
Iran does have elections, and a parliament. It used to be a secular democracy until 1953? Know what happened?
> country, it's moral for absolutely anybody, no matter how foreign or humble, to kill any element of that government.
Except the pro western dictators to whom we are eternally grateful.
Iran has kinda sorta elections..with theocratic oversight. Also, it's pretty difficult to remove leaders without going through ordinary citizens to get to them.
>I think that one habit people really need to get out of is calling governments by the names of their countries, and vice versa.
Yep
Iran still elects a jew to its parliment. In this, its theocracy resembles Israel's.
*Lebanon
What I don't get is why the US joined in. They had Israel already doing their dirty work for them. They just needed to supply weapons, which they already do and is not against international law, and intelligence, I'm not sure about the law for that but it can be done discreetly. I don't think that would grant them any sort of moral high ground, but it seems like the pragmatic thing to do.
Bibi has played Trump hard. As for the bunker busters, those can apparently only be dropped by B2s, so it's not like the US could "just supply" them.
I have seen the theory that Trump ordered the strikes to take the wind out of Israel’s sails. It is easier to call for a ceasefire when one side has already achieved one of its war goals.
I do think this verges on the “5D chess” school of giving Trump too much credit, and would prefer a simpler explanation. Maybe the intent was to inflict damage to the nuclear program regardless of how long or how effectively Israel keeps fighting?
Alternately, some general showed up to the daily brief with a contingency he wrote in 2016 and said the right things to get the President’s buy-in. I don’t know.
Because the US has deep bunker busting weapons and Israel doesn't (and they can't just hand them over, since they require B2s to deploy which only the US has). For the US it was a cheap free option to finish off buried iranian nuclear sites for free.
The US is signalling to its peer enemies. Iran itself is immaterial, other than they made an easy target (through losses of Hezbollah/Syria/Lebanon).
Where peer means “definitely not peer because Iran isn’t a peer country”. China is probably not quaking in its boots, and it has nuclear bombs. As does North Korea. The way to make sure you aren’t bombed for having WMD is to have WMD.
> China is probably not quaking in its boots, and it has nuclear bombs
They should be at least a little concerned. This is a great tech demo proving that yes, the B-2 is still untouchable, and the US is still quite capable of putting half a dozen of them above your head without you noticing.
They should be afraid of the B-2, and they should be even more afraid of the B-21, which will be like the B-2 but far more numerous.
Hasn't Israel basically flattened Iran's air defenses? Would a B52 have been at high risk dropping our bunker-penetrating bombs on this site?
I mean it proves that the B52 is not touchable by whatever Iran has but that’s it. The Chinese can launch ICBMs anyway and Iran cannot so I can’t see that happening.
does it prove that? The Iranian air defense system had been pretty thoroughly dismantled. The Israelis had supposedly been flying over Iran for ten days (including in non-stealthy F15s) without loss. So B2s getting in and out without loss also seems to tell us very little about the untouchability of the B2, when going up against a competent and intact air defense network.
The B2 may in fact be untouchable, but Iran being unable to touch it seems like very weak evidence for the proposition.
B2s ran over the Pacific too. Did China notice?
The American Military Might got squashed in the Ukraine (on a logistics basis, the Russians make cheaper arms and get a better kill per dollar ratio). One doesn't need to have them quaking in their boots to get them to re-evaluate how likely they are to get squashed by America.
The Germans famously got a much better kill per dollar ratio than the Allies in both world wars. They lost (at least in part) because their enemies had more dollars. In WW2 this became a deliberate policy of the British and then of the USA - prepared to be wasteful of high quality materiel rather than risk high casualties.
Can you cite me some sources on that? (I did a brief googling and didn't come up with anything) My understanding was that the Nazis favored "quality over quantity" in terms of materiel and that the Americans favored the opposite.
> The American Military Might got squashed in the Ukraine (on a logistics basis, the Russians make cheaper arms and get a better kill per dollar ratio).
That's a ridiculous conclusion, unless by "American Military Might" you mean "a half-assed fraction of America's obsolete equipment that was just a few years away from the scrapyard, slowly trickled down over years".
HIMARS appears to be actively under development (software wise) in the past decade (2017 to be specific).
Aside from that, what I mean is what we have for sale to other countries. Obviously they don't get our best stuff. This, same as the first Iraq war, is a tech demo. A bloody, useless, tech demo.
China may or may not be quaking, but they are probably reevaluating how likely Trump is to order military action if they invade Taiwan.
To add more details, if China wants to conquer Taiwan militarily, China for sure needs the US to not intervene. Aircraft carriers based in the ocean east of Taiwan and land based Typhon missile systems in Phillipines and/or Okinawa would prevent China from attacking with land troops across the Taiwan strait.
A good video on the Typhon's possible role against China.
https://www.youtube.com/watch?v=TM2XQ2ssvxE
Yeah, as far as I can see, one big lesson from the Ukraine war is that current technology favours defence over attack. Crossing 100km+ of water in a slow-moving and easily tracked boat while being actively attacked by missiles, artillery, planes, submarines and drones just doesn't seem possible.
> is not against international law,
I don't think this is a consideration, beyond telling your press office to replace references to "international law" with "rules-based international order" in the statements when you decide to flout it (or at least what you're saying it is when you're citing it against Russia or something).
It has signaling value. TACO Trump is not a good look on the world stage, and Israel basically tossed Trump a perfect alley-oop.
Does it? TACO isn't about Trump's willingness to send crazy signals, it's about his follow-through. Dropping a handful of bombs mostly just reinforces the same picture everybody already had of him: he's inconsistent, unreliable, and prone to lashing out whenever the whim strikes him.
To shake the reputation you're referring to, he'd need to commit to a policy that's politically costly for him and then stick with it despite those costs. The conflict with Iran could go that way eventually, but this right now certainly isn't that.
The US has never bombed Iran before and Trump has for a while now has rhetorically taken a non-interventionsit position, so it is an important signal. This attack wasn't a whim; it's an old plan and Trump has been coordinating with Israel in their strike plans since February while considering this potential option, according to the Israel Hayom article.
"The US has never bombed Iran before and Trump has for a while now has rhetorically taken a non-interventionsit position, so it is an important signal."
That is the kind of thing that has useful reputational effects, but not the remotely the kind you're claiming.
TACO most emphatically does *not* stand for "Trump Never Lies For Years and Then Makes Sudden Reversals On Key Stances In High Stakes Situations." That would be TLFYTMSROKSIHSSS. Which aside from not rolling off the tongue nearly as well is not a claim that anyone who's been paying any attention for the past decade would ever make. Trump is notorious for lying. He's notorious for being unpredictable. He's notorious for taking big risks that calmer heads might not take.
Doing exactly the sort of thing he's notorious for doing is going to reinforce his existing reputation, not suddenly make people re-evaluate. Now, the thing that legitimately *is* going to cause reasonable observers to update is not the strike but the ceasefire. I am quite surprised (pleasantly so) that Trump seems to have de-escalted the situation so quickly.
It's still much too early to be tallying up points for the whole affair, but if he does a good job managing the ceasefire and steers the Middle East in a more peaceful direction, it *will* significantly update his reputation in my eyes (as I suspect in the eyes of many others). It still wouldn't really touch on the TACO thing--it's not that sort of situation--but I'd consider this sort of reputation more important anyway. Being able to manage a tense situation and broker a stand-down among mortal enemies is a much more important quality in a world leader than not flinching in the face of legitimately harmful consequences (which isn't, y'know, actually a good thing).
1. This is somewhat underspecified, which has led to a lot of confusion. Is Iran physically building a nuclear bomb, in the sense of "shaping a collection of 90% enriched Uranium, explosives and various other stuff to be able to produce nuclear yield?" Probably not just yet. But they're definitely doing a lot of stuff that is clearly intended to give them the capability to do that in short order. And from a practical standpoint, we should treat this about the same as one of those people who buys a maybe-not-a-gun-kit that you need to file like three things down to turn into a working gun, even if you can technically argue that "they aren't building a bomb".
2. Yes, so far as international law is a real thing (which it mostly isn't). Iran fired several hundred missiles at Israel in April 2024. That is an act of war by any definition, and there has been no peace agreement since then.
3. What diplomatic alternatives? Iran doesn't want to give up the program, and since Trump tore up the deal in his first term, there hasn't been a good way to get them back in.
4. Desirable? Probably, because the current regime is pretty bad. Viable? Much less so. The record of doing that from the air is not particularly good.
To be clear, I'm not saying I think it was a good idea. It's put Israel in an extremely awkward position where they basically have to keep bombing Iran forever, and that's not a great place to be. The US might be able to get out of it, although that heavily depends on the next few days. But I can also understand Israel feeling like they don't have a lot of options, given Iran's continued sponsorship of programs to make them not exist. Diplomacy is often messy.
Disclaimer, I'm Israeli.
We think there's a profound asymmetry in that Iran's ruling regime is really invested in annihilating Israel (the word they often use), and we are not at all invested in somehow erasing Iran from the map. This goes way beyond posture/rhetoric, in that Iran has for the past 45 years consistently funded/trained terrorist groups operating near and inside Israel, and encouraged state actors under its partial control to more hostility against Israel, creating a metaphorical "ring of fire" around it. As a partial list, these are Hizbullah, Hamas, and the regimes of Asad in Syria and Houthis in Yemen.
Israelis sincerely believe that if Iran got a nuclear bomb, the ayatollahs might well be crazy enough to first-strike it against us, even though we'd surely nuke them back (I understand that some reasonable people disagree with this, but it's possible that they underappreciate the investment Iran's made into anti-Israeli activities over the last half-century, to no apparent benefit to their nation). I believe that no reasonable outside observers would put any non-negligible probability on Israel first-striking Iran with nuclear weapons.
So, Israelis broadly agree that Iran getting nukes is an existential threat, even though they might disagree about its size. Netanyahu in particular has made a decades-long career of distracting voters from whatever domestic problems he's mired in by talking about Iran's existential threat, and the substantial part of the population that's fed up with him is really tired of it. NEVERTHELESS, even people who suspect Netanyahu of always maneuvering political decisions towards his own benefit, are still giving the government (and the army and intelligence, especially) the benefit of the doubt, because they feel that anything we can do to prevent Iran from getting nukes is vitally important.
1. Yeah, getting 60% enriched uranium was tough and expensive and they don't need it for anything else
2. I'm not sure. I think it's not more illegal than e.g. the US attacking Iraq, and I think it's definitely less illegal than Russia attacking Ukraine (because that's a war of territorial conquest, which is much much worse than "merely" waging war, both in terms of UN statute and common sense)
3. Hard to say
4. Basically the appetite grows with the eating, somewhat cynically. First two days of the war, no one was thinking of regime change, but Israel has achieved air superiority to a much greater degree, and saw much fewer victims of Iranian missiles, than planned for. Israelis think that most Iranians are fed up with the regime and that it's autocratic, oppressive and undemocratic, but we don't have regime change as a goal in itself, only insofar as it'd (possibly, uncertainly) help remove a brutal and dangerous enemy which the current regime is. It's not clear how viable regime change is, everyone says no one's ever achieved it by bombing and they're probably right; the level of discontent within the country might help but will probably not be enough absent further dramatic developments.
As an American I’m inclined to agree with you but I note among my peers the good will Israel seemed to have once have in this way has worn thin. Do you worry that Israeli actions in the recent war have cost in terms of optics (something I thought Jews as a people were mindful of)? Or do you see that as unavoidable or down to the whims of political trends?
There always was that tradeoff around preemptive strikes. But the attacker's advantage is overwhelming - compare the 6 days war to Yom Kipur, and Israel war with Iran to the war in Gaza.
It is not good for the world's stability, but this seem to be the case
Like
1. Yeah, almost certainly. We follow the converging lines of evidence, right? Iran has had a "peaceful, civilian" nuclear program for decades and have produced:
a) three heavily-guarded and secretive enrichment facilities, one of which was dug into a mountain.
b) zero civilian nuclear reactors.
c) 400kg of 60%-enriched uranium. I.e., the sort of fissile material that no one would ever create unless they wanted to build a weapon.
Everyone agrees on all of this; everyone agrees that their breakout time has been measured in the weeks for a while; everyone knows they could put together and detonate a dirty bomb if they wanted. The only question is whether they're pushing for weaponization (i.e., building a ballistic-/hypersonic-deliverable warhead): Israeli intelligence said yes; American intelligence said no. Eli Lake & Justin Logan had a good debate on the Moynihan Report podcast (https://www.youtube.com/watch?v=4VtZy6flYJA) which went over similar grounds, and Lake made the point: look at all that the Mossad has accomplished in Iran. They set up a drone factory, called a bunch of high-level generals to an insecure meeting *against explicit orders*, they destroyed anti-air batteries on the ground—Israeli intelligence networks are very deeply entrenched in the regime. American intelligence, on the other hand, well, https://www.reuters.com/investigates/special-report/usa-spies-iran/. So I think we have good reason to trust the Israelis when they say Iran was actually breaking out, was actually pushing for a weapon.
(One more reason in this direction: the damage to the Axis of Resistance over the past months has been totally disastrous for the regime. They feel much less safe, and it's not unreasonable to think they'd make a last-ditch push for a nuclear deterrent.)
2. Israel and Iran have been in a state of war since October 7, 2023... If we want to play the "who struck first?" game, Iran doesn't win. I think it's a dumb game to play, though, and the strike can simply be justified as a preemptive, preventive measure against the Iranian nuclear threat (see [1]).
3. I do. The best Trump could've done would've been a rewrite of the JCPOA which absolutely did slow Iranian progress toward a bomb... but it did absolutely nothing about (exacerbated, in fact) Iranian support for awful, destabilizing terror networks. Iran is responsible for much of the carnage in Yemen, Lebanon, Syria, and, in fact, Iraq, not to mention Israel & Palestine. This oversight was bad; the terror networks were enabled when we lifted sanctions without sufficiently strict conditions. At worst, this war will put pressure on the Ayatollah to accede to a worse deal.
4. It's absolutely desirable—90 million people are suffering under a terrible authoritarian regime. Whether it's viable is another question: Haviv Rettig Gur (an excellent Israeli historian & journalist) made an appearance on Sam Harris' podcast the other day, and he said something like, "for 40 years, the regime has done nothing but suppress opposition groups. Their one job internally has been to make it impossible for dissidents to offer a realistic, well-organized alternative." I think there's a lot of truth in that, and it makes this very difficult. Because a ground op + occupation is not a good idea! I mean, the logistics alone... [https://deepleft.substack.com/p/against-war-with-iran]. Any sort of regime change will have to be homegrown. I think Israel and the US should make the environment as friendly to it as possible, and Reza Pahlavi should advertise himself to Iranians directly, and so on—but it's ultimately all up to the Iranian people. Israel should be sure to hit only legitimate targets (especially things like the Basij headquarters [https://www.timesofisrael.com/liveblog_entry/israel-confirms-hitting-basij-hq-evin-prison-destruction-of-israel-clock-in-tehran/], which was a very good move) to minimize rally-around-the-flag effects. But that's really all they can (should) do to help.
I think you might need to split this into two responses. One for the US, and the other for Israel. I'm not sure how you could possibly determine that Israel *doesn't* have a "legal" right to declare war against a country that's spent decades supporting multiple terrorist organizations with the primary purpose of destroying Israel, who also recently sent hundreds of missiles directly against Israel. If that's not casus belli then what possibly could be?
As for the other points, Israel couldn't do worse with a different regime, especially if it was popular with the Iranian people (the current regime is not). And diplomacy with a country willing to spend billions on a "we're going to exterminate your people" platform every year seems like a non-starter.
As for the nuclear program - it's just about incontrovertible that Iran has built the capacity to make a nuclear weapon and separately the capacity to mount a nuclear warhead on a mid range missile. They even made a point to show off the missiles and the payload size a few years ago. Whether they have or will take the more minor steps to complete a nuclear warhead and attach it to an existing missile is not known. This is in the face of multiple countries telling them to stop in very serious tones. That they are a major sponsor of international terrorism, including against the US and its allies, makes even the possibility of nuclear armament a significant concern for anyone on the potential receiving end of a nuclear strike.
The responses to you about number 2 in particular ("international law is fake", "the US should not be subject to international law") disturb me and tend to reinforce my opposition to the war if this is the sort of thinking that characterizes relatively thoughtful supporters of it.
International law is fake because there's no enforcing body. You can become an international pariah for using Big Nuclear Weapons -- and to some extent, a Little International Paraih for using chemical weapons (like Israel has). But these are schoolyard popularity contests, not the Principal. In the schoolyard, rules are enforced by "people willing to beat each other up, or, in extremis, kill the wrongdoer."
In that sense, there is international law, it's just fuzzy and ill-defined, and operates more on a "We Think That Makes The Game Unfun" (for example, assassinating world leaders makes the game unfun.) and "Can you catch us" basis.
It is a truth that economically ruining the German economy is against international law (blowing up the pipeline). That said, nobody's going to war over it.
Can you elaborate on your thinking on international law? Why are those arguments not thoughtful?
My steelman of those arguments.
Law: a binding custom or practice of a community : a rule of conduct or action prescribed or formally recognized as binding or enforced by a controlling authority [1]
Looking at the defiition I see there are two ways for a (international) law to be recognized at least de facto. Either a binding custom or practice and/or a controlling authority. As there is no world government, I think I can say there is controlling authority so international law doesn't meet that definition. I would argue that a controlling authority is needed for there to be any law, but I will ignore it so I can address the other part; a binding custom or practice.
I do not claim to know all of the treaties and agreements that the United States has signed. The United States is an UN member so has at least agreed to live by the UN Charter.[2] Article 2 of the UN Charter starts as "The Organization and its Members, in pursuit of the Purposes stated in Article 1, shall act in accordance with the following Principles." Bullets 3 and 4 are most germane to whether military force can be used.
3) All Members shall settle their international disputes by peaceful means in such a manner that international peace and security, and justice, are not endangered.
4) All Members shall refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations.
The above are supportive that the US may have acted contrary to the UN Charter. But there are two counterarguments.
a) Iran has been attacking US and Israel for the last 45 or so years through proxies. The proxies (Hamas, Hezbollah, Houthis, etc) give Iran some deniability. I guess if one does not accept those proxies are actually supported by Iran that may mean the US is acting badly. But I don't think anybody would argue that because there is so much evidence that Iran is the key (only?) backer of at least the 3 proxies I named.
So, it looks like Iran has been consistently violating 2.3 and 2.4 of the UN Charter. What should the US and Israel do to defend itself? I argue the US and Israel have legitimate reasons to attack Iran with military force.
b) Article 2 is making no claims to be a law. It itself says "principles" and not law or even rule. As the charter was likely debated for a long time, I don't think this is an oversight. So article 2 is not claiming to be law.
So, potentially article 2 can be used to justify the customs and practice part of being law. But also potentially, article 2 was written to give it the veneer of being a law, but the wording itself does not resemble a de jure law. Thus, I would say article 2 is not meant to bind any of the countries. I make this argument because I think the "customs and practices" are meant to address laws that are created by common law (or similar customs in non-Anglio phile world). So is there a custom to not militarily attack other countries? I think it is a worthwhile goal, but almost since the UN Charter's inception, countries have attacked other countries militarily. Russia attacking Czechoslovakia, North Korea attacking South Korea. US becoming militarily engaged in Vietnam. I think there is more evidence that Article 2 is a great ideal, but not really a binding custom or practice.
If the US or Israel has signed on to other treaties and agreements that limit use of their respective militaries, I would love to learn about them.
[1] https://www.merriam-webster.com/dictionary/law
[2] https://www.un.org/en/about-us/un-charter/full-text
> Iran has been attacking US and Israel for the last 45 or so years through proxies
In any case where Iranian proxies have attacked the US it’s in war situations like Yemen. In Syria the US wasn’t directly involved but the support of rebels against Assad has meant that a branch of Al Queda have come to power, about two decades after the attack on the WTC. This is an odd war on terror.
Yeah, the Middle East is where the enemy of my enemy is my friend becomes a (realpolitik) fact.
Iranian proxies have attacked US directly outside of Yemen. It is suspected that the pre-cursor of Hezbollah was responsible for the bombing of the US Marine barrack in Lebanon.
https://en.wikipedia.org/wiki/1983_Beirut_barracks_bombings
I guess the attack could be considered a war situation, but the reason I used suspected is because the group that claimed responsibility is "Islamic Jihad", not Hezbollah. But arguably, if two "militaries" are attacking each other then it is a de facto military conflict.
Yes I know about the Beirut bombings. However if an example of Hezbollah attacking the west is pre Hezbollah, not in the west, during a military conflict and from the 80s the existential threat seems weak.
Your argument seems to move the goal post. Your argument now is that the threat needs to be existential to warrant a military response. My argument was that a country can respond militarily if it is attacked militarily. Which we have lots of evidence Iran was doing to the US and Israel.
I appreciate this comment, because I've felt a bit baffled as well but in the opposite direction. It's one of those situations where you listen to someone who disagrees with you and feel lost for how to engage. But after listening to a few podcasts and articles written by people who are opposed to the war, I think I understand why the two sides seem so far apart from each other.
It really all comes down to the Obama Iran deal and whether you think it was stopping Iran from building nuclear weapons. People who support the war think that it was a terrible deal because it wouldn't stop them from making weapons, and people who oppose the war seem to think it did stop them from making weapons and would continue to stop them. I've heard several commentors say something along the lines of:
1. The Obama Iran deal stopped Iran from getting nukes.
2. Trump tore up the deal, and failed to negotiate a new one.
3. Therefore, this war is not actually about Iran getting nukes.
It makes sense if you believe premise one, but people on my side don't and didn't. I remember back when Obama made the Iran deal all the commentators I would read and listen to were in agreement that it gave Iran too much and didn't do enough to get rid of Iran's ability to make a bomb. That maybe Iran would pause it's development, but the deal would leave Iran's ability to start up weapon production again any time they wanted with little ability to stop them. If I recall correctly that was the primary reason why the Republicans in Congress refused to vote for the treaty, which is why Obama had to make the deal as a non-binding executive action (which is why Trump could just toss it when he was elected).
We know that Iran had developed the facilities needed to create nuclear weapons. There's a reason they put the centrifuges under a mountain; it was so that Israel couldn't destroy it. You don't put civilian nuclear power equipment in a bunker under a mountain! Them getting to keep all that equipment and the capability to enrich uranium was not acceptable to our side, and certainly not acceptable to Trump. We are not going to just trust that Iran isn't going to use it going forward, even if they aren't using it right now.
So to answer your point 1: Whether they were in the process of building a nuke, they had the ability to create a nuke and Trump has been very clear for years that Iran having that ability is not acceptable. Yet in our negotiations with Iran they have made it clear that they will not accept any deal that removes their ability to enrich uranium. If they're not going to agree to it, then giving Israel the go ahead to take those facilities out (and then using our bunker busters on the ones Israel can't hit) is our BATNA.
As for your other points:
2. I don't know and I don't really care. If the actions violate a treaty that the US and Israel has signed then I may care, but "International Law" in general is not something I think the US should be subject to. Besides, was Iran following international law all these years as they have funded the Houthis, Hezbollah, and Hamas?
3. Yes, I think diplomatic alternatives have been adequately explored. Iran made it clear that they would not accept any agreement that removes their ability to enrich uranium, and Trump will not accept any agreement that lets them keep that ability.
4. I think regime change could potentially be desirable. The Ayatollah is an evil man who is the head of an oppressive state that openly wishes for the destruction of the United States and many of our allies. A regime that is less evil, less oppressive, and not actively hostile to US interests would be great! The question is how viable that is, which I'm not sure of. In any case, regime change or no they now no longer have the ability to enrich uranium. If regime change is necessary to keep them from re-obtaining that ability, then regime change may be required.
Probably an unpopular opinion, but I have quite a lot of trouble caring about whether Iran might have gotten nukes. Even as much as I'm generally in favor of nuclear non-proliferation, I look around and notice that the list of people who currently have the authority to launch nukes include:
1. Vladimir Putin
2. Xi Jinping
3. Donald Trump
4. Benajamin Netanyahu
5. Kim Jong Un
6. Narendra Modi
Obviously others will disagree, but from where I sit, that list includes a shockingly high number of the most evil and irresponsible human beings on the godamned planet. While I'd certainly prefer that Ali Khameni not join them, it's really, really hard to believe that the sky will fall if he does: not with club membership already looking like that. Meanwhile, the U.S. track record of getting involved in wars in the Middle East is utterly abysmal. I have no confidence whatsoever that this will produce any sort of positive, long-term change in the region, and plenty of reason to expect that it will make everything worse. On top of that, the domestic angle should not be overlooked: in using force without any buy-in from congress, Trump is once again pushing the legal and normative boundaries and trying to expand his power. It would be concerning from any president, but from a president who has done so much of that already, and sent so many strong authoritarian signals, it is downright alarming.
I agree that a lot of evil and irresponsible person have nukes, but that's not an argument in favor of letting another evil and irresponsible person have a nuke. Especially not when the cost to prevent them from joining the list is "Let Israel off the leash and drop some big bombs, nobody (American) gets killed".
"Especially not when the cost to prevent them from joining the list is "Let Israel off the leash and drop some big bombs, nobody (American) gets killed". "
If that is, in fact, the only cost then I agree. And things are looking better than they did yesterday, for sure. But the cost--in terms of escalation risk--still looks considerable, so I remain unconvinced for the moment.
We aren't "letting" Iran have nuclear weapons, any more than Iran is "letting" us have nuclear weapons. Neither nation has the power to prevent the other from possessing a nuclear arsenal if that's what they decide to do.
Excluding the very hypothetical case where we break out a few dozen of our own nuclear missiles and pre-emptively kill ten million Iranians while destroying Iran as an industrial nation, which I *hope* we all understand isn't going to happen even though it is the only way to stop Iran from having nuclear weapons if that's what they really want.
Is it your position that we (whether "we" is US+Israel, NATO, or all other nuke-possessing nations, or something else) can't possibly simply play whack-a-mole with any and all refinement sites and/or ICBM/MRBM launch sites Iran tries to put together, faster than they can do that?
I'll admit, part of me feels like that's very possible, in light of the last few days. Another part of me grabs that part by the collar and reminds it that Iran is roughly thrice the size Afghanistan, with comparable terrain, and the US couldn't pacify Afghanistan. Then again, Afghanistan never tried to build a nuclear missile defense program, and if it tried, we probably would have been able to notice and stop it pretty easily during occupation. OTOOH, Iran has a much better economy; OTOOOH, Israel seems to have considerably more infiltration; OTGH, I think I have to say I don't know either way.
The political side also doesn't tell me anything definitive. The West seems pretty gung-ho for no nukes in Iran; Russia and China probably have the only non-West votes that could matter, and their positions seem to be somewhere between "ehh" and "abstains courteously". But I can't help but imagine they could be natural allies on this issue.
Yeah, pretty much. It's the old saw of "we have to get lucky every time, they only have to get lucky once". Iran has the state and industrial capacity to run a massively redundant, decentralized nuclear arms program, and to rebuild it again and again if need be. They've got all the technical data they need, it's just down to manufacturing now. They understand OPSEC and maskirovka, and they've got an awful lot of generic factories, warehouses, and caves to hide things in.
And they've got enough 60% enriched uranium right now to build at least 8-12 bombs, that can be turned into 90% in a few months with a few hundred centrifuges. The HEU is basically indestructible if they're not stupid about it.
Because Iran spent most of the past twenty years developing a nuclear arms program but conspicuously *not* using it to build bombs, because they instead very conspicuously made it a point of diplomatic negotiation, Israel was able to construct a fairly precise and detailed map of what they needed to hit this time. They're not going to get twenty years next time, and they probably won't get even two.
I think you left out the head of Pakistan (Shehbaz Sharif?).
I also left off Emmanuel Macron, for example and potentially several others (depending on how you count). The list wasn't supposed to be comprehensive: just the heads of nuclear-armed states that I already had a very poor opinion of. I couldn't even have told you Sharif's name without looking it up, so I can't possibly claim an opinion on his qualities as a leader or a human being.
I actually would rather any of these people have nukes than Iran. They're a religious theocracy convinced that military victory over their enemies is proof of divine favor, and who make real world massive sacrifices in furtherance of that belief. Everyone else cares more about preserving their country's existence by a much bigger margin than the regime in Iran.
I have a very poor opinion of Iran's leadership, but the "yeah, but THESE guys are too crazy to be trusted with nukes" line still feels like a very convenient excuse that gets trotted out whenever somebody wants to justify preemptive force. I certainly heard the same thing about Hussein in 2003. I didn't think any more highly of him, but I was still skeptical of that claim then, as I think I was right to be.
Wait, what?!????
Hussein is the one leader in the middle east who was most explicitly in favor of getting nukes as a means to destroy Israel
If he was strategically smarter and didn't invade Iran and then Kuwait and just went for building nukes the world would have ended up in a much worse timeline
https://www.fpri.org/article/2011/08/why-did-saddam-want-the-bomb-the-israel-factor-and-the-iraqi-nuclear-program/
"I remember back when Obama made the Iran deal all the commentators I would read and listen to were in agreement that it gave Iran too much and didn't do enough to get rid of Iran's ability to make a bomb. That maybe Iran would pause it's development, but the deal would leave Iran's ability to start up weapon production again any time they wanted with little ability to stop them. "
That is correct. Also, a lot of people really, really wanted Iran to stop sponsoring international terrorism, and to stop holding annual "Death to America" rallies. And it's quite understandable that Americans would want all of these things.
Some of us also want a trillion dollars, working FTL starships, eternal youth, harems filled with enthusiastically consenting supermodels, and other such things.
There was never any possibility of Iran agreeing to a deal where they so thoroughly divested themselves of nuclear technology that the would have no ability to restart nuclear weapons production. Given their status as charter members of the Axis of Evil, they'd have been fools to agree to any such thing.
And there was never any possibility of the Europeans or the Chinese pushing Iran to accept such a deal under threat of Maximum Sanctions. They had too much money to make trading with Iran, and their interests were satisfied with an Iran that was holding indefinitely at six months from a bomb. And sponsoring terrorism against Not Europe or China, and holding Death to Not Europe or China rallies was just fine with them.
The deal that was possible, with the United States and Europe and China and Iran all on board, was one where Iran agreed to indefinitely hold at six months from an atom bomb. That was the *very best deal available*. It was that or nothing. Obama chose that. Trump chose nothing. Well, OK, he chose to bet everything on his deal-making skills trumping Obama's, but same thing.
Posturing fools, thinking that tough talk and a pious refusal to compromise will cow their rivals into abject submission. And now thinking that a Short Victorious War will solve everything.
I'm....super happy with the JCPOA, and I'm super happy with the current (now seemingly over?) war.
Iran has been massively weakened. Going to war with them 15 years ago would have been a bad idea, today it's a great idea. Their negotiating position barely changed in between. So the desired course of action has.
Note that the deal was not for Iran to hold indefinitely at six months from an atom bomb. The centrifuge and enrichment limits in the deal were to start phasing out after 10 years and fully expire after 15 years. In other words, even if Trump had not pulled out of the JCPOA and Iran fully respected it, the limits it imposed would have started to phase out in January 2026.
Excellent comment, I agree except for regime change. I don't want us involved in any more countries' internal nonsense. If we can stomp their nuclear capacity from the sky then great, let's do that and be done with it. No more endless and expensive rehabilitation projects.
There's various ways to do regime change.
The Iraqi model where we commit to propping up the new regime against all other comers is a bad one, it easily becomes an open-ended engagement on behalf of a new regime that probably doesn't deserve it.
The minimalist model of regime change is the one where you decapitate the current regime and make it clear that the next regime had better be nice or you'll decapitate that one too.
On the other hand, we stopped doing as much propping for the Iraqi regime and they're still holding on fine. Seems like regime change in Iraq "worked", but was far more expensive in blood and treasure than expected and arguably wasn't worth the cost.
Where the "Iraqi model" failed completely was Afghanistan.
1) Yes, I'd say with ~90% certainty. Otherwise, why would Israel and the US have attacked? Public reports like the one you linked aren't worth the paper they're printed on, the actual intel is likely all classified. Any side can say whatever they want. What matters is the ACTIONS they take. The US and Israel bombing Iran, despite the significant risks, is a pretty strong signal that there was credible intel that they were building a nuclear weapon. Iran refusing to sign a new nuclear deal with the US (despite crippling sanctions) is another strong signal that they are genuinely trying to build a nuclear weapon. ~10% uncertainty because honestly nobody really knows what is actually going on behind the scenes here and there are still technically plausible (though less likely) reasons to believe otherwise; I just think they're unconvincing.
2) International law is mostly fake and doesn't really mean anything. The US has veto power (de facto and de jure, depending on the particular arena) in most matters of international policy anyways. I don't put any stock here.
3) Yes and no, yes in that I think it's possible that a diplomatic policy that conceded much much more to Iran could have possibly succeeded, no in that I don't think that we could have achieved the results we really wanted (stronger guarantees they won't ever try to make nuclear weapons) at this point.
4) Viable? Yes with low confidence, the Iranian government seems pretty competant at staying in power for now, but they're obviously incompetent in a hundred other ways that might make it eaiser. I'd guess regime change would be harder than Sadaam's Iraq, easier than doing it in e.g. China or Russia. Desireable? Medium confidence, the Iranian government is pretty awful as far as governments go so any change would have a good chance of being better in general; but it's still a roll of the dice. There are good reasons to want it to happen and there are good reasons to be more cautious and hope for slow internal reform.
Isn't your response to #1 falsified by the Iraq war?
What do you mean by "falsified" here? What is your certainty estimate that Iran was/was not aiming to develop nuclear weapons? I could be persuaded my estimate is too high, but "there are counterexamples or plausible other reasons" fits totally fine within my model; I just think those plausible other reasons are less likely. I think there are obvious ways in which Iran in 2025 differs meaningfully from Iraq in 2001.
> Otherwise, why would Israel and the US have attacked?
Not disagreeing with your certainty estimate, but regardless of the true value, I would guess that Israel attacked now because this is the best time for them to do so. Iran's proxies around Israel are the weakest they have been for a while, so it's a better time than the past; and US approval is not guaranteed past the end of Trump's presidency or even the midterm elections, so it's a better time than the future.
Given the younger generation's feelings about Gaza, Israel may be facing several *decades* without the US backing that they are used to. So now is their last best chance to secure themselves for the years ahead.
Why would Israel have attacked otherwise? Some variant of the following: Netanyahu wants/needs to stay in power, and he has figured that his best way of keeping a coalition together, and of mitigating the widespread unpopularity of himself and his government, is as a wartime leader rather than a peacetime leader. At the very least, I don't think it's safe to reason from the premise that Netanyahu is acting in Israel's best interests, rather than in his own perceived best interests.
It's worth noting that he's under no particular political threat at the moment (his coalition is pretty much in line), and starting a war now in the hopes of it helping with an election in a year and a half seems counterproductive (if that were his reasoning he'd have preferred it closer to the election).
Netanyahu is under no particular threat right now because he seems to busy trying to destroy Hamas right now, and the Israeli people will for very good and sufficient reason rally behind pretty much any leader that is trying to destroy Hamas.
But fairly soon, either Hamas is going to be destroyed, or it's going to start looking like Netanyahu is just playing a lame-ass game of whack-a-mole with the remnant of Hamas rather than systematically trying to eliminate it. And then Netanyahu's political invincibility goes away, people start asking questions like who was responsible for the massive intelligence and military failures that let 10/7 happen in the first place, and the Israelis may wind up dumping Netanyahu as fast as the British dumped Churchill after VE day.
But unlike Churchill, Netanyahu is currently facing criminal prosecution for a laundry list of felonies that could keep him in prison for the rest of his life. So the theory that he's mostly waging war with Iran to renew his "You can't throw me in jail *now*, I'm busy Defending Israel Against an Existential Threat" card, is at least plausible.
It doesn't help him at all to do this now though. He's under no threat until the election in late 2026 (his court case is likely to drag at least that long, and his coalition partners are dependent on him). And there's no plausible way to stretch a political boost from an Iran strike until then (if it went on for that long he'd long since be hated for it). If he was doing this in mid-2026 it'd be plausible, but it doesn't work now.
Israel is not the United States, holding presidential elections like clockwork every four years and at no other time ever. Israel holds snap elections every time their governing coalition falls apart, which has historically happened about every two years on a most erratic schedule. The current administraion is overdue, and the end of a war is the sort of thing that causes major political realignments, so no, Netanyahu is not safe until 2026.
As I mentioned, I think there are plausible reasons (yours is one of many I've heard), but I don't think they are as likely or convincing as the far more trivial answer, which is that Iran really is trying to build a nuclear weapon and Israel/US really are trying to stop them.
The US didn’t believe that Iran was trying to build a nuclear weapon until last week. Netanyahu has been saying Iran is a few weeks away for years.
Please reread what I have written (twice now) and let me know if you still think there is disagreement
Yeh. I’d say we disagree.
The Israeli planning and groundwork for these attacks go back almost a decade. Israel was waiting for a window where Hezbollah no longer was a constraint on preemption. The opening of Syrian airspace also greatly helped the logistics (midair refueling closer to Iran). It's not like Bibi woke up a few weeks ago, looked at his poll numbers, and decided to Leeroy Jenkins the IDF into Iran.
That's an entirely different point. It's still not safe to reason from the premise that Netanyahu is acting in Israel's best interests, rather than in his own perceived best interests. If you have other reasons, you can put them in response to the original question!
Bibi spent years saying that anyone who wanted to help out Israel would fund Hamas. Either you think that "funding Hamas" is in Israel's best interests, or Bibi is not "in general" acting in Israel's best interests.
I beleive I answered the question: Israel judged Iran was still a massive threat and had an ideal opportunity to attack before its proxies could regroup. Even some of his political opponents are cheering the outcome so far; Bibi is far from the only hawk in Israel.
Israel, were it sane, would judge a "denuclearized" Iran as more of a threat than a nuclear power Iran. The first might/would threaten its relationship with the United States, without which there is no Israel.
This might be why Israel attacked Iran only when there was a serious prospect of Trump negotiating peace.
This is all good stuff. I agree you answered the original question, but I still don't understand why it's relevant to what I said, rather than to the original question itself.
But attacking Iran without good reason would make him less popular, not more so. This smells more and more like a conspiracy theory.
"But attacking Iran without good reason would make him less popular, not more so."
I have absolutely no confidence that this is true.
I am not unambiguously in favor, but I'm on the other side of your arguments?
I think that Iran can clearly see that they will only be safe from hostile foreign action after they have nukes, so they are building nukes. We know they are working on nuclear power, they say it's for civilian purposes, I assume it's for nuclear missiles because they would need to be really stupid to trust that they would be safe without nuclear missiles - see North Korea and Ukraine for the relevant contrast.
I assume the war against Iran is illegal under international law but I don't respect international law to be sensible enough to care too much about it; I think the Iran hostage crisis forty years ago is a sufficient casus belli if the requirement is "only invade governments who act badly", but nobody is using it. We're just saying "we want you not to build nukes, so we're invading," and this seems like... honestly a better excuse than "only invade governments who act badly"?
I think we spent twenty years exploring diplomatic alternatives to war, but I also think they are pointless. Neither the Republicans nor the Democrats can be trusted to keep their side of the bargain, so obviously the Iranians won't keep a bargain they don't expect us to keep, and there's really nothing we can pay them worth more to them than being able to nuke attackers.
Regime change is viable (in that there's a loud opposition that protests the government regularly) and desirable (in that the current government seems worse than the opposition). The question is if it is worth paying the immense cost in treasure and blood that would be needed to achieve it, and if foreign intervention will make that cost "World War Three" instead of anything that could be worth paying, and those questions I don't know the answers to.
I had no thought of anything to review, and am too lazy anyway. If anything it would have been either roadside rest areas or mission-style gas stations, and that’s been done to death I’m sure.
So kudos to all for your creativity in coming up with novel subjects .
I am always tardy in knowing myself, in this case realizing that that Google doc was the thing I was born to review, and savagely.
My house burned down in the terrible Rowena fire in The Dalles, Oregon. Around 60 homes were destroyed in the wildfire, which is suspected to have been started by the cargo train that passes through the area. I was out of town for work at the time and was not able to get home to gather anything. While all my possessions were destroyed, whats more devastating is loss of my beautiful cat, Yuka. I don't know if he remembers this, but Scott played with and complimented Yuka after a meetup in Portland, I want to say in 2020 but the exact date escapes me.
I have been reading this blog for a long time, but I post rarely so I'm not sure if this is an appropriate place to share this. I was urged by many of my friends to make a GoFundMe. This probably doesn't fall under effective altruism. I'm blessed to have a good job and lots of support around me, but dealing with the insurance company is very stressful and anything donated here contributes to my peace of mind. I'm trying to hold what is here off to the side in case the insurance comes through, and if it does I'll be donating anything here to other families in my neighborhood, many of whom did not have insurance and are a lot worse off than me. (I read these open threads so often, but I'm not sure I've seen a GoFundMe so if this is a bad place for this, someone in charge let me know and I'll remove it)
https://gofund.me/0e769de1
Well shit, very sorry to hear all of that.
Thank you. All things considered, I'm doing okay mentally, but I am starting to feel the stress as I get into it with the insurance company. I'm planning on talking to a lawyer this week and that should help ease my mind a bit. From my experience talking to a lawyer can only help.
I'm shocked that my two favorites among the 28 reviews I read did not make it into the finalists/honorable mentions:
- Two years of parenthood [a father's perspective]
- The emperor of all maladies [cancer and the history of its treatments]
For those wanting to read the best reviews, I highly recommend visiting those, presumably in addition to the ones Scott mentioned.
This is the second year that The Emperor of All Maladies didn't make it. I thought it deserved to be a finalist last year, so naturally I would have liked to see it be a finalist this year as well. However, it was a book review in a "not a book review" contest so I could see it getting penalized for that.
Thanks for your support as always! To be honest I re-submitted it largely because you and a handful of other commenters said encouraging words, but I wasn’t too surprised to see it miss out. The ACX community likes hard facts and analysis, not Christian symbolism and TS Eliot references.
Ah well, c'est la vie. You should publish it somewhere, even if just on Substack. I'd let you guest post it on my Substack if you wanted, but I have less than a hundred subscribers so it wouldn't be much of a platform.
You’re more than welcome to publish it (anonymously or under your name or whatever you like.) I wrote it so that maybe people with cancer or their families would connect with it, not for glory or fame.
I know an oncologist who’s been published in newspapers and written several books so perhaps I’ll send her an email and see.
Oh no, JFK Conspiracies lost! I liked that one, it was clever and insightful.
Me too. Whoever wrote that definitely made an entertaining review.
I did too! It was easily the most fun to read...
There's something that I find really suspicious about the claims of the danger of AI that Scott makes, and some of the posts Scott makes, which I think makes claims of the danger of AI less believable.
Like in the AI 2027 scenario Scott endorsed, even the aligned scenario is pretty dystopic (a small group of people with all the power, no real guarantees of good outcomes despite the scenario trying to paint over it imagining a luxury future). Since both scenarios are pretty bad, and specially since there's a large given chance of the worst scenario, someone who believes them should be expected to oppose AI with all their power, trying to slow it down as much as possible, as early as possible, to improve the end results.
Honestly I'd expect anyone who thinks AI is very likely to go catastrophically wrong, and who cares about morality at all, to be radically militant in their opposition to AI, probably spending most of their wealth in measures against it, after all if the AI gets as powerful as it's claimed, there's no turning back, and the damage would be tremendous. So do we see this kind of opposition?
Obviously there's parts of this opposition that we can't see just looking at Scott's blog, but what I'm confused by is posts like "The Colors Of Her Coat", or the "Make A Personalized AI Kids’ Book" mentioned here (which I haven't read, but from the title can infer its bent). These kinds of posts serve as propaganda for AI, helping hype it up and promoting a positive view of it, which seems really incongruent with thinking the AI will be very likely bad. If AI is so likely to be catastrophic, why is he using his blog, which probably is a large part of the political capital he can use to exert influence, to make posts to say "ooh look how pretty the pictures the computer makes". Surely if it's a matter of such import, any influence we can wield to make the chance of bad AI go down is useful, and any influence that can increase the chance should be avoided.
If Scott thinks the chance of bad ai is so great, and such a big threat, why does he still make posts like that? even if they help shift attitudes only a smidge, given the stakes it still should be taken into account. shouldn't he avoid making them, or even retroactively erase them? even more, why doesn't he try to make posts that help build opposition to AI, like he has never really gone into the economic arguments against AI, which is very relevant for artists. All of this seems at best like a strategic blunder, and at worst, evidence of bad faith. Why should we believe the hyperbolic claims for misaligned AI, if he doesn't even shift his blog away from things that help hype up AI?
To me this just makes these claims not credible, those attitudes are not those of someone for whom AI is the greatest threat facing humanity, they're more those of someone who maybe is helping hype up AI, because maybe they have some investments that would benefit from that hype. I don't want to accuse Scott of being a bad faith actor, maybe he just follows his intellectual whims, maybe he just hasn't considered the part of strategy revolving around his blog posts (itself a bit of a strange omision), or maybe he thinks it's intellectually dishonest to post strategically? (though as long as it's not misrepresenting reality it shouldn't be) but regardless, seeing that kind of very "ai positive" post pretty strongly tells me "this guy doesn't consider this a serious threat"
I was especially turned off during his most recent back-and-forth with Tyler Cowen, in which both sides cited o3. I preferred it when Scott would simply state outright that he didn't have a source ("Epistemic status: I made it up") and frame his arguments accordingly.
By all means, use AI to find real sources--but then check those sources to make sure they're real, and that they actually say what the AI claimed they said, and then link directly to them.
I'm likewise annoyed that Scott spends a lot of time worrying about AI but then generates (almost invariably poor-quality) AI images for use on the site, and effectively *promotes* them on posts like the "Make a Personalized AI Kids' Book" (which had mostly poor-quality AI images, FWIW).
Maybe it's just a finger in the cracking dam, but let's try to *avoid* all of the AI bullshit, m'kay? The image-generating stuff is ugly and the LLM stuff is both badly styled and can't be trusted not to make stuff up, cite fake sources, or have random agendas.
It's all a waste of time.
I don't see a contradiction. AI being useful is a precondition for AI being dangerous in the sense that Scott et al are worried about. Moreover, if I understand the AI safety arguments rightly, the aim is to try to maximize humanity's chance of threading the needle of building a useful and well-aligned ASI and using it to substantially better the human condition. Butlerian Jihad scenarios where there is a concerted effort to kill and bury AI development are considered a failure mode to be avoided, especially the risk that the attempt is unsuccessful (i.e. most countries ban further advanced AI development, but a badly-aligned ASI gets developed and deployed anyway in a non-banning country or a country that fails to enforce its ban well enough), just as much as the scenarios where laissez-faire regulation permits rapid development of unfriendly AIs.
Caveat: I am explaining a position not my own. I expect with moderately high confidence that the current crop of LLMs will plateau in capability well short of ASI or even AGI. They'll prove to be useful tools once the dust settles and the hype dies down, especially for natural language processing and computer vision, and might be an important piece of the puzzle for future AGI development, but I consider Scott's 2027 scenario very unlikely. The only reason I don't dismiss that scenario completely out of hand is that smart, knowledgeable people whose judgement I respect (e.g. Scott) have spent time and effort analyzing the issue and concluded that it is likely.
I would note that "butlerian jihad scenarios" are a failure mode to people like Scott and Yud, because they actually do want to reap the potential spoils of AI down the road and are okay with elements of transhumanism. But a concerted effort to kill and bury AI development for good would absolutely not be a failure-mode to the growing group of opponents who simply do not want this ever to be created, and instead prefer to maintain human supremacy indefinitely. People like Scott probably will not like many of the motivations and arguments that those real people bring to bear against AI, but I have no doubt that messy ugly chauvinistic or fear-driven populism is a more powerful force, and I think at some point they will need to decide whether to tactically join the growing coalition that doesn't want this stuff to ever become part of human life.
I for one very much doubt that mere intelligence is sufficient to do the types of good that AI proponents are always touting, many of the more longterm futurist dreams could just by physically impossible, and I'm not even convinced that e.g. a cure for cancer is a thing that can be created no matter how smart you are. All of the destructive uses, OTOH, have proof of concept, and are much more likely. I find every possible outcome where a superintelligence exists to be worse than humans retaining primacy on the earth, and this will just be a battle we have to fight and a line we have to hold for as long as we value our species.
It's tempting to agree that LLMs are looking like they may plateau, but I think we are possibly not getting the whole story of what they may have internally (and I get that we can't trust that due to either hype or possibly downplaying some research angles), and AI2027 really only depends on them reaching better-than-human coding ability, along with some assumptions about "research taste" slowly shifting over to them. That seems like the tipping point, that it would eventually happen from there, even if the timeline is likely to be longer.
i don't think AI being useful is a precondition for being dangerous. If AI decides to... say, "make anime real" that's not actually useful (unless you care a lot about meme campaign promises). But "make anime real" costs a lot of computing time. Whether this leads to melted ice cream, and consequential health risks, or whether this costs google some fraction of search capacity, we can say that AI "wanting to do big things" is dangerous, regardless of whether the AI is useful or being useful while doing so.
The danger that Scott, EY, etc are worried about is an AI that is useful enough in the short-to-medium term that humans are tempted to trust it with real power which if misused could lead to dystopia or Armageddon. This might happen directly by using it to automate real physical stuff at large scales (like the classic fictional trope of giving and AI control of nuclear ICBMs) or indirectly by uncritically taking its advice at corporate or national policy levels (like in some of Asimov's robot stories, although there the AI takeover is intended to be benignly paternalistic rather than malignant or hostile).
Even the kind of danger you're talking about, of wasteful use of resources towards frivolous aims causing environmental damage and opportunity costs, doesn't happen unless people with control over large amounts of investment capital think it's to their advantage to spend huge piles of money developing and running that AI.
It's certainly possible for an AI to seem much more useful than it actually is. I strongly suspect this is happening now with the current crop of LLMs. But that should be self-limiting as eventually the hype cycle dies down.
If you think LLMs are currently mostly useless, a dead end in terms of capabilities, and the worst case scenario is wasted resources, then pragmatism and honestly should both incline you towards making the case that they're useless and wasteful. But if yo think they could easily become either an existential threat to humanity or the wellspring of a utopia, then Scott's approach of talking up both their utility and their danger makes sense.
It's not about AI being useful by itself, it's about reinforcing the view that it's useful in other people's eyes, thus helping drive engagement with it, helping drive hype, helping drive investment, helping drive development (is this a too weak slippery slope? I'm not sure what the weakest link is, but my point is, given the supposed stakes, gambling with the chances doesn't seem rational). The color of the dress post even speculates about post singularity AI giving us wonders, thus helping people be slightly more comfortable with the idea it'll work out alright/it'll end up aligned. But the whole point of the more AI safety concerned people is that alignment is extremely unlikely, so we should be extremely, extremely cautious about continuing development. Maybe butlerian jihad is a failure mode of that, but a big part of the point of people like Eliezer and Scott is that threading that needle is extremely unlikely, so we shouldn't risk it until we're extremely extremely sure. In that effort, and seeing how the capital of the world is prone to hype and short term gain, maximizing opposition to AI, for someone in a pretty influential position, seems to me to be most consistent with the position.
A problem with strategic lying is that people may draw conclusions different from what you intended, and then you have made the situation more difficult for yourself.
If you convince people that AI is useless (which many of them already believe, it seems to be quite popular opinion), that would imply that it is harmless. No need to worry about it, no need to regulate it...
I'm not advocating strategic lying, rather strategic selecting of topics, while remaining truthful... so not choosing to highlight how cute and pretty the computer images are, and instead choosing to focus on topics that are still true, like the economic aspects I mentioned, but which reinforce negative sentiment. I don't think that strategic selectiveness would be any more intellectually dishonest than following intellectual whims. And yeah, implying it's useless could be counterproductive, so the topic of usefulness could be avoided, and instead focus on how it's evil or bad in other ways.
The real AI safety activism is political organization. Right now we're in the "this is so cool!" stage, like Facebook or smart phones in 2009. Soon we'll be in the weaponization stage and lobbyists will go to the wall to prevent regulation or raise taxes on the AI owners to pay for UBI.
There's something that I find really suspicious about the claims of the danger of AI that Scott makes, and some of the posts Scott makes, which I think makes claims of the danger of AI less believable.
Like in the AI 2027 scenario Scott endorsed, even the aligned scenario is pretty dystopic (a small group of people with all the power, no real guarantees of good outcomes despite the scenario trying to paint over it imagining a luxury future). Since both scenarios are pretty bad, and specially since there's a large given chance of the worst scenario, someone who believes them should be expected to oppose AI with all their power, trying to slow it down as much as possible, as early as possible, to improve the end results.
Honestly I'd expect anyone who thinks AI is very likely to go catastrophically wrong, and who cares about morality at all, to be radically militant in their opposition to AI, probably spending most of their wealth in measures against it, after all if the AI gets as powerful as it's claimed, there's no turning back, and the damage would be tremendous. So do we see this kind of opposition?
Obviously there's parts of this opposition that we can't see just looking at Scott's blog, but what I'm confused by is posts like "The Colors Of Her Coat", or the "Make A Personalized AI Kids’ Book" mentioned here (which I haven't read, but from the title can infer its bent). These kinds of posts serve as propaganda for AI, helping hype it up and promoting a positive view of it, which seems really incongruent with thinking the AI will be very likely bad. If AI is so likely to be catastrophic, why is he using his blog, which probably is a large part of the political capital he can use to exert influence, to make posts to say "ooh look how pretty the pictures the computer makes". Surely if it's a matter of such import, any influence we can wield to make the chance of bad AI go down is useful, and any influence that can increase the chance should be avoided.
If Scott thinks the chance of bad ai is so great, and such a big threat, why does he still make posts like that? even if they help shift attitudes only a smidge, given the stakes it still should be taken into account. shouldn't he avoid making them, or even retroactively erase them? even more, why doesn't he try to make posts that help build opposition to AI, like he has never really gone into the economic arguments against AI, which is very relevant for artists. All of this seems at best like a strategic blunder, and at worst, evidence of bad faith. Why should we believe the hyperbolic claims for misaligned AI, if he doesn't even shift his blog away from things that help hype up AI?
To me this just makes these claims not credible, those attitudes are not those of someone for whom AI is the greatest threat facing humanity, they're more those of someone who maybe is helping hype up AI, because maybe they have some investments that would benefit from that hype. I don't want to accuse Scott of being a bad faith actor, maybe he just follows his intellectual whims, maybe he just hasn't considered the part of strategy revolving around his blog posts (itself a bit of a strange omision), or maybe he thinks it's intellectually dishonest to post strategically? (though as long as it's not misrepresenting reality it shouldn't be) but regardless, seeing that kind of very "ai positive" post pretty strongly tells me "this guy doesn't consider this a serious threat"
Scott, you listed "Men" as a finalist, but there's no review titled "Men", and two reviews which include the word "Men" in their titles. Could you please clarify which review you're referring to? Many thanks!
I am almost certain it's "The Men Are Not Alright," which was really excellent.
(Though I don't think I read any others with "Men" in the title, so if that one is *really* really excellent, I could be wrong...)
The other was "Shrinking: Men", which Scott wouldn't have abbreviated with "Men". So yes, it's "The Men Are Not Alright". Which I also found excellent.
I keep hearing that AI is putting Silicon Valley techies out of work, but I have yet to see any verifiable numbers and sources. If anyone can point to some relatively well-controlled studies (or at least studies with transparent methodologies), I'd appreciate it. In the meantime, it seems like a recent change to the tax code has likely been the real culprit in the current round of layoffs. Tech companies can no longer deduct the full cost of R&D at the end of the year...
https://qz.com/tech-layoffs-tax-code-trump-section-174-microsoft-meta-1851783502
Hi beowulf. When I posted a while ago that I was reading The Dark Tower series and loving it, you replied that you'd liked it a lot too, til the end when Roland reaches the dark tower. I'm piggybacking on this thread to tell you I absolutely hated the end too. Not only did it cause me pain and suffering, your honor, it also grossly violated my values regarding what constitutes fairness and logic in literary narrative.
It was obvious in the book (also from knowing a bit about King) that he was not comfortable having a hero that killed a bunch of people. It doesn't align with his politics and his values. Seems to me he just regressed to his 60's liberal-pacifist values and gave in to a craving to punish his hero.
And the ending does retroactive damage to the whole story, right? It's all just a re-run.
I am so bothered by the ending that I am thinking of writing a new one. I'd just do a new version of the very last bit -- the part that happens after Roland enters the tower. So I thought I'd ask you how you'd like it to have ended.
Call me sentimental, but I wanted an upbeat ending. I wanted to see a renewal of Gilead and the mystical rose.
I did too. But I also thought Roland had earned an outcome of that magnitude. He was born with huge talents and used up every bit of himself turning them towards saving the world -- all the worlds -- and he succeeded. Even the people who died or suffered helping him -- he helped them too, and sharing his life for a while made their lives larger.
I think "real culprit" may be a bit strong.
A lot of things happened over the past five years that showed up at about the same time:
1) "Free"money went away. It is easier to justify borrowing a lot of money to pay for things like engineers when the interest rate is near 0%. At 5% things change a lot.
2) During Covid Facebook and other 'tech' companies (not Apple) had a huge hiring surge. With no clear idea what to do with the folks hired. This created a bunch of employees with no clear financial justification and when that collided with (1) ...
3) Elon Musk fired about 75% of Twitter and things seemed ... fine? This can encourage other company execs to ask why they have the headcount they do. Especially when combined with (2)
4) The tax changes that you mentioned. I'll note that my employer is sorta tech (we build expensive semiconductor test equipment) and we haven't experienced big layoffs. But we are older and profitable so financially different from a lot of the smaller tech companies.
5) At least one fairly large tech firm, Intel, is in financial trouble. Intel had 130,000 employees in 2022, finished 2024 with about 108,000 and is expected to lay off another 10,000 soon. This isn't because of tax code changes :-)
6) AI is disrupting, but *how* is not clear.
So ... yeah, tax code changes. And maybe that was the biggest factor. But it wasn't the only one.
7) The economy is now in a state of uncertainty. Hiring more people is not a priorit.
My understanding is that the tax code changes are especially punishing to rapidly growing companies like startups. They won't matter that much to a stable company.
2) is perhaps better understood in the context of BLM, and the tech companies hiring a lot of "Good Workers" to do nothing, in order to have their stats look better.
3) Elon Musk bought a company that had no Test System, and had been doing hotfixes to Prod for 6 years. There was/is? serious doubt if X.COM goes down, it can be rebooted.
I'm possibly missing something here, but can't Scott just use a convertible promissory note that triggers on a later financing round or some other appropriate condition? Just put no interest on it until maturity, and intend to forgive the loan if the triggering event doesn't happen prior to maturity. That is a financial instrument that already exists and you don't need to reinvent the wheel here.
The problem is that when the grant is given, there is no for-profit corporation. There’s no entity that has equity to negotiate over. You can’t have a convertible note with a company that doesn’t exist yet.
I suspect the deeper problem is that when a nonprofit converts into a for-profit, in most cases the people involved simply create a new for-profit entity with no formal relation to the nonprofit. The OpenAI situation where the nonprofit clearly has a lot of valuable intellectual property is quite rare.
But it would be great to get something like this working. In general the puzzle of how to fund nonprofits in a way that makes it possible for them to grow and succeed seems unsolved.
Does that work if the counterparty is currently a nonprofit?
Sure, assuming that we mean a formally-registered not for profit. Those are corporate entities with identified officers and etc. They are governed by state incorporation laws and, just as with an LLC or whatever else, can/do enter into contracts or debts with a willing counterparty.
If the not-for-profit corporation has achieved federal tax-exempt status (a separate process conducted by the IRS), then you know that it has full legal status in its state of registry. The IRS by law can't even consider an application for 501(c) status if the applicant isn't "in good standing" as a corporation under its home-state's laws.
> Sure, assuming that we mean a formally-registered not for profit.
There isn't any other kind of nonprofit. If you're not formally registered, you're a sole proprietorship, and those aren't nonprofit.
There are plenty of local groups which identify themselves as nonprofits and behave like nonprofits but don't do the paperwork. You should not donate to any such group, ever, but they do exist.
But anyway it sounds like I misunderstood the actual question posed.
The point I was getting at was more like "what would the note be convertible to, since nonprofits can't issue equity", but Kevin made it better in the other subthread.
Ah, got it.
Inspired by Scott’s occasional forays into speculative fiction, as a way of thinking through strange ideas sideways, I’ve started a fiction blog of my own. The stories lean speculative, sometimes philosophical, sometimes more straightforward, but always written with the hope that the narrative does more than entertain. So far, I have touched on such small topics as the nature of the universe, a direction superintelligent AI can take us, and encountering truly alien lifeforms.
For those who enjoy puzzles: I’ve hidden Easter eggs throughout—some overt, others deeply buried. A few are probably too obscure (sorry), but most are findable with the right mix of attention and paranoia. I’d be curious to know what you spot.
If that sounds like your kind of thing, you’re warmly invited to explore:
https://sisyphusofmyth.substack.com
And if you do read, I’d love to hear what you think, especially if you think you’ve cracked one.
I liked the stories, but I am not in the mood for puzzles, so I didn't try to find any.
Thank you for the kind words! I would very much appreciate a shoutout, if you might be so kind! I hope you subscribe too - I have a couple more in the works now.
I'm kind of surprised none of the game/movie/TV/music reviews made the cut but I guess that's just the ACX audience. Was sort of hoping for at least one token game & TV/movie review to make it in just for variety (and also because I definitely wrote one of those).
I enjoyed tongue-in-cheek review of Detective Pikachu as a serious / deep literary work.
The game reviews mostly weren't as good. There is no universal rule that everything related to video games is low quality and low effort but gamers gonna game and do poor work.
I can accept that they mostly weren't as good--I read several of them myself--but I don't agree that none of them were worth showcasing.
Sure, there’s the audience considerations, but I honestly thought the video game reviews here were a bit weak compared to the quality of the kind you can find on other websites. Even if the ACX audience were more open to them, I still don’t think any of them would have made the cut.
The Tv Show reviews were on average much better, but because of how unconventional this year’s format is, I think adding a few token representatives would have decreased variety, ironically enough.
Ollantay was put in that category, although it's a stage play. Other than that, the media review (as opposed to something-else review) that I ranked highest was Kiki's Delivery Service, one of five that earned a 9 from me.
I really enjoyed the review of Dylan’s “Simple Twist of Fate”
I am a little surprised about that, but oh well. We like our books!
I understand that a perpetual motion machine is prohibited by classical physics. But what about at the quantum level? Since atoms are always in motion, except at absolute zero, could one construct a perpetual motion machine that would continue on until the universe is empty?
A superfluid current in a perfectly isolated toroidal container would, I believe, theoretically continue forever. This isn't really what's meant by 'perpetual motion' though, which people use to mean energy output greater than energy input.
In this house we obey the laws of thermodynamics!
Sure, under those rules the machine will have "perpetual motion," but in the classic sense of usting them to power things, as soon as you hook a belt and pulley or a turbine up to that spinning atom you have introduced friction for the purpose of removing energy, so game over!
In the sense you're describing, by the Third Law, it is in fact impossible to create a machine that ISN'T a "perpetual motion machine."
An atom can stay in motion forever, but that isn’t typically what people mean by a “perpetual motion machine”, they mean something that you can extract useful work from with no energy input, and that is still forbidden by quantum mechanics.
Classical mechanics would allow the "literally in motion forever" perpetual motion machine--say, a planet with a moon orbiting it. But that's not what people mean by the term , so....
I think there are two categories of perpetual motion machines, which then have strict & relaxed definitions.
1. The truly impossible perpetual motion machine is one that produces net energy. Lots of cranks latch onto this one because it’d be a source of “free energy”. Doesn’t work, even at the quantum level
2a. A machine that keeps running forever (where forever is defined as a reasonably long time). For this, I suspect there are quantum perpetual motion machines, but from a certain perspective most orbits are perpetual motion machines, so it’s not that special.
2b. A machine that keeps running for a truly infinite amount of time. For this I suspect it’s impossible even at the quantum level because you have to grapple with decay of subatomic particles and the heat death of the universe.
My favorite old candidate for a perpetual motion machine was a strong magnet pulling an iron ball up a ramp with a hole near the top.
The thinking was that the ball would drop onto a lower ramp and return to the starting point. An appealing idea at first glance till you think it through and realize that a magnet strong enough to pull the ball up the ramp would be stronger than the force of gravity met at the hole.
Ask anyone who has recovered a magnet from an old hard drive and loosely held it within a couple inches of their refrigerator how gravity would fare in that contest.
Heck, I bought some magnets at Michael’s once to do some cute crafting, laid everything out separately on the table, and struggled to glue before the magnets did their own cute crafting and stuck themselves together.
> but from a certain perspective most orbits are perpetual motion machines
Don't orbiting objects lose energy through gravitational waves, leading to the eventual decay of the orbit?
Yes, that's why they're not in the "truly infinite" category 2b.
You can stay in motion...as long as you don't try to extract work (ie energy) from the system. That's also true classically.
Yes, a particle in a well has non-zero kinetic energy even in the ground state. But extracting that requires being able to transition to a lower allowed state...and there aren't any below the ground state. Definitionally.
Stumbled across an article that will probably be of interest to some ACX subscribers. Posted it last week in a reply to somebody, but I'm going to repost it as a top-level comment.
India is always stereotyped as an extremely spiritual place even historically, but there was actually a philosophy called Charvaka or Lokayata which is purely materialist. Most scholars have believed that this philosophy is relatively short-lived and mostly irrelevant for most of Indian history, but new research seems to show things aren't so cut-and-dried. (Notably, the name Lokayata literally means "prevalent among the people.)
https://blog.apaonline.org/2020/06/16/the-untold-history-of-indias-vital-atheist-philosophy/
There are some flaws with this article - for instance, the quote from Maitreyi in the Upanishads is misleading, since in the original context "no awareness" meant merging with Brahman, not annihilation - but overall there's some really interesting stuff there.
I'll also add one thing not noted in the article: that the oldest strata of Tamil literature, the Sangam poetry, has no references to religion or the supernatural. Tamil culture, especially at this time, was probably descended more directly from the Indus River Valley civilization (whose people likely traveled to the southern tip of the subcontinent) than Northern Indian culture was, having been subject to serious Indo-Aryan admixture, so it's entirely plausible that the Indus River Valley Civilization sites (Harappa, Mohenjo-Daro) had no religion as we would now conceive such things.
That's very speculative, though, and Mohenjo-Daro was structured in multiple levels with a large bath at the center of the city, which would seem to indicate some notion of ritual purity, like that which upholds the caste system, had developed. And ritual purity presumably comes with some degree of religious belief.
"No big temples or ritual centres that we have identified" does not mean "no religion/belief in spirits". People might have had little shrines in their homes, or they might have practiced rituals in a small way but not like organised religion. Or if they all got together round a bonfire at a particular part of the year and did things like drive their flocks through it for good luck, is that religion or not?
https://www.rte.ie/brainstorm/2025/0623/1390519-bonfires-st-johns-eve-june-23rd-folklore-traditions-ireland-bonna-night/
Roman funeral practices included having wax masks of dead ancestors/actors wearing those masks carried in procession. Is that religion? If they did something similar in Mohenjo-Daro, what would have remained to be dug up centuries later?
https://www.penn.museum/sites/expedition/recreating-roman-wax-masks/
Doesn't the existence of this seal, https://en.wikipedia.org/wiki/Pashupati_seal, with the classic horned god, disprove that?
Yeah, the Pashupati seal is probably the biggest piece of evidence against the idea that they had no religion. It does seem to portray a god, but for what it's worth the position of Wendy Doniger (a scholar who supports the protoatheist hypothesis for the Indus River Valley Civilization) is that we have no way of knowing what the Pashupati seal is meant to be portraying.
We have no idea what the Gundestrup Cauldron is meant to be portraying, either, but it's an interesting parallel (horned male figure, seated, surrounded by animals):
https://en.wikipedia.org/wiki/Gundestrup_cauldron#Interior_plates
https://en.wikipedia.org/wiki/Gundestrup_cauldron#/media/File:Gundestrupkedlen-_00054_(cropped).jpg
I'm considering changing my job, and perhaps my career, but I feel like my skills are too eclectic at this point, and I'm pretty confused at what my options are, and I'd like to brainstorm with someone because I feel pretty lost. I'm considering hiring a career coach, if only for the sake of someone holding the other end of the brainstorming conversation, but I worry that a regular career coach would try to shoehorn me into one of the mental categories they already have even if it's a bad fit; is there some version of it that may realistically work on someone all over the place and from the general ACX-comments crowd?
UPD. I wrote up broader context in response to a question in a comment; here it is:
I studied linguistics out of passion (understand the common laws of human communication systems yadda yadda), and from there I switched to ML (because natural language processing) in like 2016, before transformers were even invented; ML seemed really cool to me, the closest thing to magic in existing technology. I studied lots of math and other technical stuff as part of my linguistics education (our program was informally known as "mathematical linguistics"), but not as deep as a real mathematicial, and my knowledge of linguistics never became as deep as a true linguist's.
Then I went to France to a PhD school, for a thesis about natural language generation (this was before RLHF); this was not a topic chosen by love, I just wanted some position and I seemed to just like NLP generally and figuring things out.
The PhD broke me due to what felt like any lack of meaning, guidance, goal-setting etc; experiments I designed were not interesting in any ways and failed miserably. I decided that I wanted to be part of a startup, where everyone is working toward a common goal; where my chaotic strengths of broad figuring-things-out and hyperfocus would be valued. Besides, leaving my PhD meant my French residence permit was not valid anymore, and God forbid I'd have to return to Russia, so I was looking for a country that wasn't trying to arrest me or kill me. I didn't get a PhD degree (I do have a Master's).
So I joined an ML startup in a small country in West Asia, and loved the atmosphere. (I'm currently the only ML engineer in this ML startup. We're a team of like 10 people.) My way of getting shit done was finally valued, and I appreciated the stock options. Besides, being startup meant trying to be lean, using smaller, economic methods, instead of any "we were dancing around for 6 months and got a 0.5% improvement in a leaderboard" stuff. In addition, I really found a use for my passion for linguistics. Since our startup is connected with machine translation, being able to read our customers' data and actually understand what goes wrong where had been a really valuable addition to my engineering competences.
But now I've been burning out. Here's why:
· Interactions with the team and especially C-level management.
— The team was initially pretty diverse with people of all backgrounds, so I didn't feel weird being a foreigner and various other labels of weirdo, but now I'm the odd one out.
— Recently, I've developed severe sensory sensitivity, and communicated to them that since they are not willing to allow remote work, I'd like to work from upstairs (we're renting a two-floor house for an office); they felt it was too distant from the team, but allowed me to work from the kitchen. Now, however, they told me that a permanent move outside of the room where everyone works is not acceptable, in any other area even if it wouldn't inconvenience others, because that may threaten the quality of communication (the theory being that communication is much easier when they can walk up to me in the same room than when they have to open a door and see me in another room). I did explain that it influences my productivity, but they said that basically it's not only my productivity that has to be managed, but other teammates' as well (I'm still doubtful other teammates' productivity would suffer from me being in another room as much as mine suffers from a constant sensory overload).
— There was a time when I showed up with my nails polished red (I'm biologically male and look male), and the CTO personally told me that I would have to wash that off, because what if some day I'm in charge of hiring people and they are uncomfortable, or what if my teammates are uncomfortable ("So if I ask them and everyone is OK with it..." — "No, what if they are uncomfortable but too nice to tell you?")
— All of that makes me feel like there is a certain amount of, uh, restriction of my personal independence / agency, because who the hell cares where I sit or how I paint my nails if I ship quality code fast? And being the only ML engineer in the team, I suspect they may not realize that my leaving would actually be an inconvenience, since there is some learning curve to our infrastructure...
· I'm tired of ML / AI as a field. While I was shipping small and cheap solutions, NLP started to suffer megalomania, and everyone hiring NLP engineers seems to have training gigantic chatgpt-like shit in mind, which is not sexy for me (and I still have no experience with it, but I could figure it out). It seems foolish not to use my tired knowledge to *some* benefit, but I would hope to apply it somewhere where I still see a sense of meaning.
In addition, in these recent years searching for meaning, I have pursued various pet projects. For example, I have created a project called theorem marketplace (link: https://theorem-marketplace.com/) and while doing it, figured out some basic web3 development (because I had to create an Ethereum contract with an oracle). Also, I have periods when I'm in love with molecular biology, so I'm considering barging into bioinformatics. (In pure AI, I find interpretability research a la anthropic very sexy, but I have no credentials to signal I would be a good fit; and maybe I wouldn't).
Have you looked at all the EA adjacent interpretability programs? (ARENA, MATS etc) Some end in research papers and are essentially tailored for professionals that want to move towards AI safety research (of which Intepretability is a relevant subfield)
Bioinformatics probably requires you doing another PhD. I would recommend you looking at specific topics you might be interested in and trying to pick specific supervisors. Scope and quality of different bioinformaticics work varies widely.
I had done a pretty drastic career shift myself (Medicine → CS). I would say the most difficult part is primarily a lack of network, so if you are able to create work in your spare time that signals for anyone interested that you are highly motivated, that can be really helpful to be seen.
What's your education/career background?
Okay, I wrote up broader context.
I studied linguistics out of passion (understand the common laws of human communication systems yadda yadda), and from there I switched to ML (because natural language processing) in like 2016, before transformers were even invented; ML seemed really cool to me, the closest thing to magic in existing technology. I studied lots of math and other technical stuff as part of my linguistics education (our program was informally known as "mathematical linguistics"), but not as deep as a real mathematicial, and my knowledge of linguistics never became as deep as a true linguist's.
Then I went to France to a PhD school, for a thesis about natural language generation (this was before RLHF); this was not a topic chosen by love, I just wanted some position and I seemed to just like NLP generally and figuring things out.
The PhD broke me due to what felt like any lack of meaning, guidance, goal-setting etc; experiments I designed were not interesting in any ways and failed miserably. I decided that I wanted to be part of a startup, where everyone is working toward a common goal; where my chaotic strengths of broad figuring-things-out and hyperfocus would be valued. Besides, leaving my PhD meant my French residence permit was not valid anymore, and God forbid I'd have to return to Russia, so I was looking for a country that wasn't trying to arrest me or kill me. I didn't get a PhD degree (I do have a Master's).
So I joined an ML startup in a small country in West Asia, and loved the atmosphere. (I'm currently the only ML engineer in this ML startup. We're a team of like 10 people.) My way of getting shit done was finally valued, and I appreciated the stock options. Besides, being startup meant trying to be lean, using smaller, economic methods, instead of any "we were dancing around for 6 months and got a 0.5% improvement in a leaderboard" stuff. In addition, I really found a use for my passion for linguistics. Since our startup is connected with machine translation, being able to read our customers' data and actually understand what goes wrong where had been a really valuable addition to my engineering competences.
But now I've been burning out. Here's why:
· Interactions with the team and especially C-level management.
— The team was initially pretty diverse with people of all backgrounds, so I didn't feel weird being a foreigner and various other labels of weirdo, but now I'm the odd one out.
— Recently, I've developed severe sensory sensitivity, and communicated to them that since they are not willing to allow remote work, I'd like to work from upstairs (we're renting a two-floor house for an office); they felt it was too distant from the team, but allowed me to work from the kitchen. Now, however, they told me that a permanent move outside of the room where everyone works is not acceptable, in any other area even if it wouldn't inconvenience others, because that may threaten the quality of communication (the theory being that communication is much easier when they can walk up to me in the same room than when they have to open a door and see me in another room). I did explain that it influences my productivity, but they said that basically it's not only my productivity that has to be managed, but other teammates' as well (I'm still doubtful other teammates' productivity would suffer from me being in another room as much as mine suffers from a constant sensory overload).
— There was a time when I showed up with my nails polished red (I'm biologically male and look male), and the CTO personally told me that I would have to wash that off, because what if some day I'm in charge of hiring people and they are uncomfortable, or what if my teammates are uncomfortable ("So if I ask them and everyone is OK with it..." — "No, what if they are uncomfortable but too nice to tell you?")
— All of that makes me feel like there is a certain amount of, uh, restriction of my personal independence / agency, because who the hell cares where I sit or how I paint my nails if I ship quality code fast? And being the only ML engineer in the team, I suspect they may not realize that my leaving would actually be an inconvenience, since there is some learning curve to our infrastructure...
· I'm tired of ML / AI as a field. While I was shipping small and cheap solutions, NLP started to suffer megalomania, and everyone hiring NLP engineers seems to have training gigantic chatgpt-like shit in mind, which is not sexy for me (and I still have no experience with it, but I could figure it out). It seems foolish not to use my tired knowledge to *some* benefit, but I would hope to apply it somewhere where I still see a sense of meaning.
In addition, in these recent years searching for meaning, I have pursued various pet projects. For example, I have created a project called theorem marketplace (link: https://theorem-marketplace.com/) and while doing it, figured out some basic web3 development (because I had to create an Ethereum contract with an oracle). Also, I have periods when I'm in love with molecular biology, so I'm considering barging into bioinformatics. (In pure AI, I find interpretability research a la anthropic very sexy, but I have no credentials to signal I would be a good fit; and maybe I wouldn't).
Hmm, have you considered embedded/system/scientific programming? Those might appeal to you if you like working on lean, efficiency-oriented code (assuming you're not concerned about software devs being replaced with AI in the coming years - personally, I'm very concerned!).
This is not magical enough to my liking, I fear.
I've done this with Claude. You might want to start there and use it with a coach.
A right career coach will help *guide* your brainstorming on what you want in your next job/career. My opinion is they should not be prescriptive. My recommendation would be to ask the career coach for examples of former clients and how they helped them. You should also be honest on what help you need (eg help with brainstorming, keeping you accountable to a timeline, providing resources).
A right career coach will help you understand your interests, skills, and experiences so you can find the right new job/career for you. In fact, you may feel it is going one step backwards (evaluating those things again) to allow you to move two steps forward (finding your new job/career). But a right career coach will work with your pace and design a program that fits you. They also should have ready access to resources that would take you more time and effort to find on your own.
I’m going through this process myself and would be happy to exchange notes. Less so a dramatic change and more or a correction, but I’ve spent way too much time contemplating it.
Go ahead and share your wisdom.
I don't have any notes on my end yet, I'm just trying to pinpoint in what general direction I should be moving, but my current state is much more awayness from an unacceptable situation than any towardness with any specific goal in mind.
Perhaps I should have just written a commend with the specifics of my situation, instead of being silly and vague. But that may take another while...
The AI safety/alignment/regulation space needs people like you. Desperately.
Would you care to elaborate?
Zvi Moskowitz' substack Don't Worry About the Vase is a good place to start. He gives comprehensive AI progress updates weekly. In them is a section called 'get involved' about open positions. Linguists have a skill set, don't think about this work as programming. Market yourself as a smart person in the language field. Stranger things have happened!
I'm actually quite a good coder (come on, I'm an ML engineer in an ML startup), and I'm sorta aware of AI safety research (I even slightly helped run an AI safety reading group in the country where I am). That just makes me another person with an unimpressive application, no?
I work on the Hill, and just flagging that the AI moratorium on states regulating AI for ten years is likely to pass Congress as part of the Republican reconciliation package. The House version includes a blanket ban. The Senate version of the legislation, which blocks states from accessing broadband funding if they pass AI regulations, just passed the Senate parliamentarian's review. This was unexpected. In the Senate, a provision must be primarily related to federal funding to pass this review, and most experts though this provision would be stripped out.
The Senate will then have to pass the bill (it almost certainly will) and then the House and Senate will have to work out the differences in their package in "conference," and then the final version has to pass both chambers. As things stand right now, this is likely (75%+) to happen.
The time to try to get this provision removed - probably in conference - is now (if you oppose this, I do), as there is some Republican dissent on this provision.
Thank you. I agree this is terrible law, and I don't want to wait around for the courts to have their say. I will ask my rep (Tenney) to fight for states rights on this.
This is something that would make sense to be federal. Any state limiting AI use can cause problems for everyone. But if there was ever one time in the history of everything that it's worth using every tool at your disposal to get what you want, that time is now.
The problem is that it purports to prohibit any state law that regulates AI, but there are a number of legal fields where the power typically belongs to the states and not the feds and in those areas it will be necessary for states to do so. Any state licensing board from doctors and lawyers down to trades might potentially need to regulate AI uses. States will need to create liability schemes for misuse of AI, or products liability schemes for the developers themselves, as without a federal law both the state courts AND the federal courts (when diversity jurisdiction applied) would have to rely on state law for any such civil action. How about a revision to jury instructions that prohibits jurors from using AI during deliberations, is that prohibited here too?
I understand that pro-AI people are afraid of a few states gumming up the works for everybody (and I'd be happy about that as an anti-AI person) but there are a lot of actual practical problems that will require regulation that an impartial legislator would have to concede are state law powers.
>This is something that would make sense to be federal. Any state limiting AI use can cause problems for everyone.
Agreed. I _really_ don't want to see regulation that clobbers American AI like the regulations that clobbered civilian nuclear power. Still, relatively sane regulations that just impose reasonable transparency requirements, like the New York RAISE Act (covered in detail in https://thezvi.substack.com/p/rtfb-the-raise-act ) look like they would be reasonable laws to enact at the federal level.
Thank you for the insight. I'll see if I can contact my reps in time.
To be clear, the thing likely to pass is a moratorium on states regulating AI. So an AI free-for-all is winning, which is bad. We want to call and say we want states to be able to regulate AI, so we want our reps to vote against the moratorium on AI regulation. Correct?
Yes, exactly. This would technically NOT force states to not regulate AI, but if they want to get access to broadband federal funding, they cannot essentially regulate AI for ten years. Since all states want this funding, it's a lot of pressure.
Calling/emailing never hurts, though requesting virtual meetings if you live in a red state/district or your friends do would be the most helpful. And then pulling any connections that you have.
Thanks!
I just called my congresswoman (D), who had an aide answer, and my senators (R), who had voicemail answer. I left the following message, which anyone reading this can copy and optionally add some nuance regarding broadband funding:
Hi, my name is [name] and I'm a constituent from [town, state]. My phone number is [number]. I'm calling about the Republican reconciliation package. Believe it or not, it includes a moratorium on states' ability to regulate AI. Now, I oppose AI because I think it can become dangerous. I want to preserve states’ ability to regulate AI. So, because the Republican reconciliation package includes a moratorium on states’ ability to regulate AI, I'm calling to ask [representative’s title and name] to either vote against it, or to work hard to remove that part. Thank you very much.
That's perfect! Offices keep track of how many calls they receive in support/opposition to issues.
TBH it's the first time I've called my reps in probably a decade. I'm trying to get better about that, and this is an important issue.
Wohoo congrats! Never a bad thing. Good time to get more involved in issues you care about.
I don't really understand who besides Marc Andreessen thinks this is a good idea? Where is this push to preemptively keep away all AI regulation coming from? Even Altman seems interested in baseline regulations, and certainly Musk and Dario do as well.
The military…just guessing.
Why would you guess that? State regulations certainly wouldn’t apply to the military…
The military is getting more and more input from new companies like PLTR, Anduril and others. AI is important to them. It’s not a hill I will die on…that was the thinking that drove my impulsive guess.
I think they are saying different things publicly vs privately. Altman especially has been having a lot of conversations with the admin
Big if true. I might have to reconsider my “Altman means well and is not actually a super villain” stance.
Did you miss The OpenAI Files? Good summary place to get caught up to speed on the "sama isn't a straight shooter" position.
David Sacks and Jensen Huang would be another two pushing the memento moratorium, at least. I don't think Zvi has a specific standalone post addressing the issue, just regular weekly references to its status, else I'd point there.
Oh yeah forgot about Sacks. What a nightmare this whole thing is
It's very telling that a good portion of the people who have worked with him completely distrust him
Musk is genuinely opposed to the ban, but his advocacy has been carried out in an ineffective manner
Nah, the time to oppose this was before. At this point it's likely too late. I mentioned a while ago during the Biden administration the AI pause people were making pretty obvious mistakes and it would bite them if Republicans got back into power. Now Republicans are back into power and it's biting them. This is a common issue in politics: if you wait until you're in a position of weakness it is almost always too late.
If it makes you feel any better, it's only a ban on AI specific regulations. You can still pass laws that affect AI but they have to apply to AI and non-AI things equally.
Well hopefully if this sneaks through we can get creative and single it out in ways that either include some other software that wouldn't matter, or which could fail to apply to theoretical AI frameworks that aren't neural nets and which nobody is using. "All software distributors must be able to document how their product actually works" , "software containing non-algorithmic outputs in user interfaces shall create strict liability upon the developer for damages proximately caused by the output".
I still think you'll end up wrong about the political valence of this, the populist RW isn't having any of this AI stuff and the Republicans have to care about more than the Chamber of Commerce these days.
I'm sure this will be adjudicated but disparate impact is well understood. What would work is if they're willing to burden the entire software industry generally. For example, if they're willing to force Google search to exist under the same restrictions at ChatGPT.
If you think the Chamber of Commerce is the ones pushing for less AI regulation then I'm sure you don't have a handle on the political economy of this. It's not the populist RW either but they're also DEEPLY suspicious of regulation of AI since they believe it's being weaponized against them. Which was another thing I thought was a long term mistake by the mostly left leaning AI regulation crowd.
I'm seeing both my state's GOP senators criticize it (Hawley strongly, Schmitt somewhat obliquely as he seems to have some broader goal on AI regulation), as well as MTG and Luna coming out against it from the House. I was surprised the parliamentarian wasn't quietly encouraged to just toss it out to avoid any fight when it gets back to the House, as they have a razor thin margin there and the RW populist opposition is sufficiently eccentric and out on their own island that they may be willing to tank the bill over it.
there is a lot of private concern/dissent... but Republicans tend to fall in line to just pass the package. Still a decent chance to get it stripped in conference
Thanks for the hopeful news!
Participating in a review contest is kind of rough: there are a lot more good entries than finalist spots, or even finalists + honorable mentions. Lots of people write something interesting and worthwhile and get nothing for all their work, sometimes not even feedback. To help with that I'm going to award my special Tasty's Consolations Prize, for all the reviews that I liked that didn't make it into the finals (or got honorably mentioned). The winners get nothing except for dubious bragging rights and a warm feeling of being appreciated. I read "only" 50% of all the reviews, so I obviously must have missed lots of worthy entries.
* 11 Poetic Forms, Reviewed in Verse
https://docs.google.com/document/d/1d0vRSj1E93joWWvbUen2XGuDjN_mM94ybMIAADzM2fo/edit?tab=t.0#heading=h.bdyc6ymrmrt2
It's clever and it showed me new things and it featured Actual Poetry (the kind that adheres to some predetermined structure of rhymes or meter), what more can I ask.
Enjoyed the sestina so much, I got close to memorizing it.
I should say though: I think the review doesn't do the villanelle justice - the example villanelle really does seem rigid and repetitive and one couldn't blame you for reading and concluding that it's a very restrictive and stilted form based on that example. But how different it can sound in the hands of a great poet! https://allpoetry.com/If-I-Could-Tell-You And of course, the rage-rage-yadda-yadda-something-something-light, is also an example of a villanelle. Maybe it's unfair to unfavorably compare the Reviewer to W. H. Auden, but someone has to defend the honor of the villanelle.
* “As Little as Possible”: Slave Morality in Chinatown and Spec Ops: The Line https://docs.google.com/document/d/1d0vRSj1E93joWWvbUen2XGuDjN_mM94ybMIAADzM2fo/edit?tab=t.0#heading=h.ylmh0fxq3gss
Two celebrated pieces of media, that the reviewer likes on the artistic level both seem to come close to saying the same thing the reviewers pushes back against: "trying to make things better will always inevitably backfire and make everything worse". I'm a little hesitant to praise this review, because it kind of says nothing new that I didn't already agree with.
* Earth
https://docs.google.com/document/d/1d0vRSj1E93joWWvbUen2XGuDjN_mM94ybMIAADzM2fo/edit?tab=t.0#heading=h.9b5iqpz1vxbb
It's a fun sci-fi story parodying YouTube/gaming culture and video game auteur culture about present day Earth being horrible.
* Love
https://docs.google.com/document/d/1a3q0Z2tuPLLbDeg5-pfEffkajGjrfPDwE7ZMs7uaWQs/edit?tab=t.0#heading=h.i8v5hwoui0lf
This is a story of the Reviewers tragic romance. It's exactly the same as a million other such stories, but it doesn't make it less painful to the person it's happening to. Reviewer, I'm sorry it happened to you. I wish you well.
* My Imagination
https://docs.google.com/document/d/1a3q0Z2tuPLLbDeg5-pfEffkajGjrfPDwE7ZMs7uaWQs/edit?tab=t.0#heading=h.5p8ou8bafkgh
Since he was a kid, the Reviewer has been spinning in his mind an elaborate, anime-like/Marvel-like fantasy world with an extremely convoluted story and worldbuilding, and kept spinning it into adulthood. It's not that this is great art that is going to become next Lord of the Rings or Star Wars, but it seems like there's something worthwhile in anything a person put so much thought into.
As a kid, I had an imaginary world like that too. It seems like this is fairly common. When someone grows up to become a writer, it makes for a good back-story - "at the age of 5, so-and-so was already writing epic fantasy", except that this isn't unique to writers - seemingly lots of people did. Then later we all thought there's no point to it if you're not going to write it down, and writing down is a ton of work, so almost everyone gave up on it, while the Reviewer kept going on and on. Is there a point? Some part of me wants to say this kind of escapism must be unhealthy somehow, or at lest a waste of time, while another part is delighted end sort of envious.
* Rubbermaid products
https://docs.google.com/document/d/1a3q0Z2tuPLLbDeg5-pfEffkajGjrfPDwE7ZMs7uaWQs/edit?tab=t.0#heading=h.7c5deelkzhqz
A paean to the brand of cleaning utensils and more, and a manual on how to clean a toilet. Very Some Guy-core.
* The World as a Whole
https://docs.google.com/document/d/1jYVJFIz5-aMi0LCgsC9AN6BncJDNVGaMU37QmwZ1vzA/edit?tab=t.0#heading=h.hw9u0dhx3ksg
I don't have much to say about this one. It is another bleak one capturing real pain and horror at the badness of the world.
* GETTING OVER IT WITH BENNETT FODDY: a perfect review in which I fail https://docs.google.com/document/d/1B3YxYxLVFjKGicaEkqvt353kt0uimn0PSUQv1PyaHuI/edit?tab=t.0#heading=h.bfsgvhmf2607
A review of a famously difficult and frustration-inducing game, which is known for constantly destroying your progress and bringing you back to square 1. It is a gimmicky review, but a good one. I'm a little surprised this one didn't make it into the finals. It is also half-autobiography? Yes. It made me want to get Getting Over it and play it solely as a masochistic character building exercise. So far I haven't.
* Deathbed Ballads
https://docs.google.com/document/d/1hjIUcKi-vIM9RGRZDWPRyUtzZLlVrpaY6Jy81iZVYi4/edit?tab=t.0#heading=h.mnx8wqk31rum
So this is about a clade of songs sharing the same pattern: a dying man tells the narrator a story of how he ended up that way. The pattern changes, mutates, develops pseudopods, plays as a serious tragedy here, as a joke there. The review is a pretty serious analysis, and long. Perhaps it suffers somewhat from not having any kind of grand unifying idea at the end, and it also suffers a little from not including youtube links to the actual songs, for the copyright reasons, I imagine. I suspect lots of readers didn't finish this one.
I enjoyed it for an opportunity to see song evolution in progress and for all the interesting music I got to listen. My favourite of the bunch was The Unfortunate Rake https://www.youtube.com/watch?v=1aQd99zfIRo, an entirely serious and heart-breakingly sad ballad, about the hero dying from syphilis.
* The Tale Of The Princess Kaguya
https://docs.google.com/document/d/17WAcFrTOExgendrk2lGTxRIR4LVC3ZGbiU5Xe3kF4mg/edit?tab=t.0#heading=h.uzxd2fc95slp
A review of a Ghibli non-Miyazaki cartoon, adapting a Japanese folk tale. A tale is simple, but serious and sad, and it should provide some food for thought. The Reviewer gets a very gloomy message out of it, that for everyone there's a thing that would be good to do, that would be right and beneficial to do, but that goes so much against our nature, that we'll probably never get ourselves to go ahead and do it. Unfortunately, this rings true to me, and truer than I would have liked. Possibly the review would have done better if the plot summary section had been shorter. On the one hand, the plot summary was essential for the point the Reviewer wanted to make, on the other hand I could imagine readers getting bored there. But it is a review that stuck with me all the same.
* ******* ******
https://docs.google.com/document/d/17WAcFrTOExgendrk2lGTxRIR4LVC3ZGbiU5Xe3kF4mg/edit?tab=t.0#heading=h.l0j92yt4nqxc
Thank you for delving into a dark and terrible subject in the name of anthropology. I don't have you courage and won't be watching the series any time soon, but this makes your investigation all the more valuable. You are the true mythic hero for going where others dare not go, for staring into the abyss, for descending into the cave that is also the underworld somehow. I now understand modern kids slightly better thanks to your efforts.
Hehe, you know you've got a good hobby when it makes other people simultaneously delighted and worried about your health. I wrote the Imagination review, glad you found it worth engaging with! If it eases your worries somewhat, it's not like my life revolves around this fictional universe, it just sort of sits in the background most of the time, and I'll occasionally add a few details when one of life's natural lulls comes my way, or my creative juices are flowing. I suppose the point is to do something that is intellectually stimulating, and creatively rewarding in the long term. Storycrafting is a surprisingly entertaining way to pass the time, I definitely recommend it.
In other news, kudos for making it through ~50% of the reviews, that is far above what I was able to achieve. Definite agree on the Rubbermaid review, there's nothing quite as entertaining as a person's hot blooded passion for mundane objects. Really captured the spirit of the contest in a way that few other reviews managed.
Thanks for your kind words re: Skibidi Toilet! I don't think it was THAT brave of me - like I wrote, it got surprisingly engrossing quickly - but I always appreciate being called a true mythic hero so don't let me stop you. dop dop dop yes yes
I just looked at the Rubbermaid review, prompted by your mention. I hope the reviewer is old enough to have experienced the Before - not just Before Amazon, but Before the grocery store had a large display of brushes, mops, and brooms - when the Fuller Brush Man (happy day) came to your door, and gave you a little “extra” if you made a purchase.
I wrote the Deathbed Ballads review. I'm glad you liked it, and I agree with most of your criticisms of it. The length and complexity snuck up on me while writing it. I was about a third of the way through before I realized how much serious scholarship there was on the topic (especially for the Unfortunate Lad and St. James Infirmary) that I needed to engage with in order to treat the topic fairly. I was particularly handicapped by two sources I really wanted to dig into being unavailable on any reasonable terms: one was an academic paper that was behind an extremely steep paywall and the other was a then-temporarily-out-of-print book that I didn't have time to buy and read in time to use by the time I realized I needed it.
If I'd had another couple weeks to work on it, I would have gone back, strengthened the conclusion, and revised the introduction and body of the review to work towards that. And I would have needed to do some additional background research to support the conclusion. The conclusion I probably would have worked towards was further emphasis on the importance of syphilis to the song clade and the significance of the disease to the times, places, and communities where the core elements of the song cropped up: first in big metropolitan commerce hubs like Victorian London, and again in the American black community during the interwar years. In the latter case, there was a big and now-little-remembered epidemic of syphilis that probably had its roots in the American expeditionary force to France in World War One, and the epidemic persisted and spread among a discriminated-against minority community due in large part to marginalization, poverty, lack of access to diagnosis, neglect by public health authorities, and lack of good treatments even for people with access to medical care. I would have liked to have explored that in more detail and connected it to St. James Infirmary, and also explored parallels between the interwar syphilis epidemic and the AIDS epidemic in the gay community in the late 20th century. I only touched on pieces of this, partly because I was out of time and would have needed to do a lot more research to treat it fairly, and partly because the review was already too long.
Another piece I started doing but abandoned was hypothetical cladeogram of the song genre. I gave up on it because it wasn't coming together visually, I was out of time, and it was extremely speculative to begin with and would have needed a ton of explication and qualification to not be misleading.
Part of my problem, timing-wise, was that I started out intending my topic to be an analysis of the lyrical content of the music of the American Civil War as a corpus of data on cultural attitudes about the war's conduct and causes. I abandoned that and switched to what I hoped would be a lighter topic once I found that very quickly becoming too research-heavy. When my "lighter" topic also proved to be too heavy-weight, I had even less time to finish it.
I went back and forth on the question of including youtube links at the top of each section, and might have made the wrong call there. The advantage of including them is obvious. I decided against links or embeds (except for a couple of footnotes where youtube was the best source for an obscure-but-important variant) because of a combination of copyright concerns, clutter concerns (especially for St. James Infirmary, where there are so many major variations), and worries about how the formats of both google docs and substack would handle embedded youtube videos.
I wrote the Getting Over It review--thanks for the kind words! Nice to see a few comments about it here and there to make the hours put into it feel worthwhile. Hope you have a good time (or enlighteningly bad time) trying to get over it ;)
This was one of my favorites too :)
Hopefully this isn't against the spirit because finalists have already been chosen, but since you opened a discussion here, I'll just follow up that I really disliked the review of The Tale of Princess Kaguya. I think the reviewer misreads the story so badly (possibly due to missing cultural context, possibly due to the source material itself, i.e. the original folk tale, being somewhat inconsistent so that the movie makers' efforts to harmonize the themes made it quite subtle and complex) that they had to project an entirely orthogonal point onto it, and I have half a mind to write a counter-review.
(Interestingly I just checked my voting records and I actually gave it a 7, which was above average on my scale. Thinking back on it, I probably still thought that the essay was well-written and its points are well-made, despite missing the mark as a review of the movie per se.)
I wrote it. Open to discussing this, if you're interested. If you feel a need to correct me -- which I definitely get, I feel the need to correct people if they give interpretations of art I don't like -- well I promise I'll read your take.
As a preliminary note, I don't have much background knowledge, neither about the time period nor about anime in general. In my mind this doesn't matter because I've made very few claims about what the movie is trying to do, and more just described what I got out of it. (With some exceptions.) I would be extremely surprised if Takahata had thought about anything that I'm talking about in section IX, for example, but I don't think that matters.
Edit: there was some discussion under the Voting post about what to do with reviews that aren't really reviews, whether to penalize (and I somewhat self-servingly responded that I'd vote based on how much I liked them, with no penalty). The truth is that I immediately thought of my essay as in that category. I never approached it as, "let me tell you what this movie is about", at all. But as was discussed before, this was a "non-book review" contest, rather than a "non-[book-review]" contest, so it's totally legitimate to take issue with that. It just didn't occur to me that anyone would care about it being a "proper" review until after the deadline. (I also titled my essay "Little Floating Bamboo" rather than "Review of the Tale of the Princess Kaguya", but Scott changed all the essays to just have the name of the thing they're reviewing, which, in retrospect, was a pretty obvious move to make things less confusing.)
Hello! I'd love to discuss this, thanks for following up! Reading back my comment now, I realize it probably read a bit dismissive. I apologize for that.
In the end, I largely agree with your self-assessment. I think your review is a good essay, in that it makes a point (about doing Genuinely Hard Things) relevant to the reader and defends it clearly. I even agree that "I would be extremely surprised if Takahata had thought about anything that I'm talking about in section IX" and that those points deserve to be made. But I think it does matter that Takahata had so little to do with it, and that is the main reason why I "really disliked the review".
Takahata clearly made the film with his own intended themes, and I would have liked to see a review that engaged with that more, especially because the film made a valiant effort to wrangle the source material into a coherent film for a modern audience, but made it just a bit too subtle and complicated, and in spite of that I really love it.
Allow me some time to compose a full response, because as it turns out I have a lot to say.
Thank you for your review/essay (I thought it fit comfortably in the "review" category, which is broad)! First, for encouraging me to watch the movie before reading your review; I'd seen most Ghibli output, but not that one, and it was time well spent. Second, because it was an entertaining read. Third, for including the source material (Japanese); although I don't read Japanese, doing so spoke to your commitment to the original rather than to a translator's gloss.
As with C.A., what I took as the theme of the movie was completely different from what you took from it. Which in and of itself is fine; great art ought to speak to different people in different ways. But the review left me unfulfilled because I didn't see your message in the film and your review didn't make a strong enough case to convince me that the message was there.
I have no idea whether what I got from the film was the (or an) intended message either. For me, it was a meditation on happiness.
So many people wanted to make the Princess happy, or claimed to want to make her happy. This starts with the bamboo cutter, then the bamboo cutter's wife nearly rips the baby out of his hands because (as she says) only she can make the baby happy. Furthermore, they didn't want to make her happy because happiness was a good thing, they wanted to make her happy because making her happy would in turn make them happy. And making her happy was a form of control, of possession.
Is trying to make people happy a bad thing? Yes, the movie said to me, and the problem is not the "happy" part but the "make" part.
The Princess, in turn, does not find happiness in anything others do that's supposed to make her happy. Her friend growing up in the village didn't consciously try, and she does find happiness there with him. Her flight with him at the end is her imagining happiness with him, but to do so she abducts him away from his wife and child, so the happiness she envisions together is no better for him than the happiness others imposed on her.
Finally, the arrival of the Moon people is accompanied by visions of effortless happiness. And while she returns to that, she is giving up the life on Earth where happiness is so hard to come by but so much more precious because of it.
Huh.
Well -- I said in the review that she would've been happy in the forest village but not in the capitol, and that her father never succeeded to make her happy.
I didn't talk about intentionality, and it is true that Sutemaru is the only one who didn't try. But that's not the reason he succeeded. He succeeded because he didn't put her on a pedestal, whereas everyone else did (which I did talk about!). You could absolutely make her happy by choice, you'd just have to understand what she wants. (Like you, idk what Takahata thought, but if he came out in an interview and said, the point of the movie is that making a girl (or people in general?) happy is only possible if you don't try, then I'd just say, well, he's wrong ¯\_(ツ)_/¯)
I don't agree on the 'kaguya abducts sutemaru from his family' point. She technically does for this one sequence. But she wants to live in the village with him. I don't think there's any indication that he'd have to leave his family for her. I mentioned in the review how their relationship is explicitly suggested to be non-romantic, so it doesn't conflict with him having a wife and kids.
(And her mother I think did succeed. She didn't succeed all the way because she can't because she's a mother, but she made her life much more pleasant than it would have otherwise been. She did about all she could.)
A simple proposal to make the world (well the US) a slightly better place:
Congress passes a law, imposing a tax & paperwork/compliance requirements on any American company, who serves GPDR/cookie warnings to an American IP address. If you fall into that category, you have to pay some sort of annual tax, and there's Deliberately Annoying Paperwork that you have to file with the government. The point is to harass these companies and make them stop serving cookie warning nag screens to Americans. My personal view is that the GPDR has absolutely ruined the Internet and made the world a worse place.
The law does not prevent companies outside the US from having cookie warnings. If a German website feels like it needs to comply with the GPDR when Americans visit, OK no problem, you're still free to do that. It does not prevent American companies from complying with the GPDR and having cookie warnings when European IP addresses visit their websites. If Disney needs to comply with the GPDR in Europe, again, no problem.
The idea behind the law is to counteract the Brussels effect as it relates to purely internal US web traffic. Millions of American websites harass US citizens with cookie nag screens, even though the GPDR obviously does not legally apply to us. I'm guessing everyone's in-house counsel just told them it was easier to comply. Now, we can change the incentive structure and make it easier for them *not* to comply with Europe's insane laws. Constitutionally the US cannot forbid American websites from having cookie nag screens, but we can certainly make it difficult..... American sovereignty will be restored, and all of us will stop being annoyed by obnoxious European laws :)
Imma make a killing selling a special device that cuts the top of bottle caps off again. (Has that EUism reached burger land or is it just us who have to suffer it?)
Not yet - it usually starts in California
BTW, re
Beyond Bottles:
The directive also addresses other single-use plastics, with measures like bans on plastic cutlery and increased recycled content requirements for plastic bottles.
Glad I'm in the USA. I often find it convenient to use plastic cutlery to eat off non-stick surfaces without scratching them.
I've heard that one problem with cookie warnings is that they (well, in many implementations) make the internet harder to use for blind and otherwise vision-impaired folks who use screen readers because they steal focus and make sites harder to navigate (and are frequently hard to close from a screen reader, especially if there is a gazillion options). I wish someone had sufficient political weight to argue cookie warnings out of existence because they destroy accessibility.
I'd just make it so cookie warnings have to follow some kind of standard that makes it trivial for your browser to deal with them.
Cookie warnings are malicious compliance. GDPR does not require anything like what has been implemented. Everyone should pitch in to punish the perpetrators, not just Americans.
First of all, you're a freak and a weirdo (unlike everyone else on this website! Especially me!) if you don't have cookies generally blocked and use per-site permissions in a browser which implements these to strictly regulate what sites can set cookies at all on an opt-in basis.
Secondly, any decent adblocker will have an annoyances filter which strips the cookie warnings. It takes seconds to add this filter and you should be using an adblocker anyway if you like to browse a usable internet.
I use Adblock to nuke nag screens regularly, but:
1. Some websites (thankfully a minority) are completely locked until you agree to their terms & conditions which include cookies. Adblock can't fix that
2. I use two adblockers, the more powerful/useful one is Ublock Origin, which Google is in the process of removing. I already lost it last week when one of my computers updated, and I only have it on this current browser because I've blocked updates so far
"Some websites (thankfully a minority) are completely locked until you agree to their terms & conditions which include cookies. Adblock can't fix that"
I've never encountered this personally. What websites?
"Ublock Origin, which Google is in the process of removing"
Wait, what? Filthy, if true. Move to Firefox, is my answer to that.
Have you tried the Lite version of the extension? What features do you find lacking? I understand there are some, but I don't seem to suffer from their absence.
> First of all, you're a freak and a weirdo (unlike everyone else on this website! Especially me!)
Yeah come for Scott’s writing and stay to hang out with fellow weirdos.
You have now made me stand up and cheer for GDPR. Oh gosh, the poor Americans have to click through a warning screen? Welcome to the rest of the world, which gets the Americanised version of everything shoved down our necks. Sorry, the calendar is set to American date format, you'll have to change it. Paper sizes? American. Spelling? Three guesses, and the first two don't count. Default language? English - US of course!
Turn about is fair play, let you Yanks suffer a taste of what we routinely go through.
I think there's a significant difference between "we happen to have most businesses based in our country, due to the strength of our economy, and they naturally reflect our preferences ", and "we're throwing tantrems and the threat of government coercion to try to force businesses *not* based in our country to adopt our preferences" actually.
"we're throwing tantrems and the threat of government coercion to try to force businesses *not* based in our country to adopt our preferences"
Yeah, but that's how America always does it, so what can you do? 🤷♀️
Can you walk me through the paper size frustrations? Do you have a large stockpile of metric unit paper bought for a metric unit printer misguidedly built for the European market that is now incompatible once that effort was abandoned? Is there a need for your notebooks to conform to A4. Does A4 not make good paper airplanes because your 30cm straight edge doesnt line up with it correctly? I am struggling to see the use case where its inconvenient for you.
I’m largely pro European but the EU does itself no favours with GDPR. I feel if you got some monkeys together, smashed them over the head with a hammer a few dozen times, threw them off a cliff, and finished up by getting them drunk on fermented banana juice before giving them all a few typewriters to play with you might get this kind of output.
I found out a few days ago I may have to appoint, or appoint myself, as a data controller because my doorbell camera might see a small part of the local village, although I’d be well within my rights to take pictures while in the village.
That kind of nonsense is bad law, as is any kind of law that can be interpreted in a fashion the writer didn’t conceive of.
No. We have 300 million people and more money than God so we get to set the rules. No fucking cookies.
I mean, I'm as annoyed about American Defaultism as the next European, but I also find cookie warnings incredibly annoying and pointless and would be in favour of any efforts that might make them less ubiquitous.
Get a browser extension. I use "I still don't care about cookies."
Is there any equivalent of this for mobile browsers?
Depends on what browser and OS you're using, but if it supports extensions at all, it probably lets you use this one.
I use a Firefox fork (Iceraven) on Android, and I think it actually suggested I use this over the older version "I don't care about cookies." If you prefer a Chromium-based one, Kiwi Browser has been recently discontinued, but Edge Canary should work.
I don't want my country- the world hegemon- to bow down to privacy laws implemented in Europe. The EU is free to regulate the EU, but they can't regulate America. I find the whole idea ridiculous. It's a basic sovereignty issue
I wonder how many people in the US agree with your sentiment? Any estimates?
Oh, I agree completely, and think the US should go much further and threaten to not honor NATO defense agreements as a way to reshape EU laws and regulations particularly repellant to core "American values."
But making "the world (well the US) a slightly better place" is a longer-term thing. You can fix your specific cookie issue in less than a minute.
And that *wouldn't* count as imposing regulations on another country, "a basic sovereignty issue", as the comment you completely agree with puts it?
Of course it would. My proposed policy is that America extend its practice of infringing on the sovereignty of lesser territories to our enemies in Europe, the greatest threat to our freedoms.
Greater hostility between Europe and the US is definitely desirable.
Am I the only one who thinks being able to reject non-essential cookies is actually a useful feature?
(obviously the UX should be better though, ideally a choice you make once in your browser instead of separately on every website; something like https://en.wikipedia.org/wiki/Global_Privacy_Control)
"Am I the only one who thinks being able to reject non-essential cookies is actually a useful feature?"
You are not. But getting a *bunch* of dialogs to do this can be really annoying. And you get a bunch of dialogs even if you check "yeah, sure, whatever." Is the annoyance worth the usefulness? Not clear ... especially if the servers just get more sophisticated in how they track you (which is my guess ...)
The thing is, not only GDPR does *not* require the bunch of dialogs, but displaying those dialogs is actually *against* GDPR, which says that denying all cookies should be equally easy as accepting all cookies.
So what you see is not even malicious compliance, but more like malicious pretense of compliance. And your reaction is a proof that this type of corporate propaganda works perfectly.
It is also not clear to me whether the annoyance is worth it, but it very well might? I'm honestly not sure.
It's not just that we can reject cookies now, there's also the side effect that websites are less likely to implement unnecessary tracking cookies in the first place since it now involves a lot more effort.
Well, YOU could try Consent-O-Matic.
Good idea, I just installed it.
That said, like the OP, for the sake of this discussion I'm more intersted in discussing what the EU and US laws should be, not so much whether the laws can be worked around by savvy users.
What is your estimate of the cost of Israel’s two-week campaign against Iran?
According to The Economist, from October 7, 2023, up until just before the attack on Iran, Israel had spent approximately $85 billion, or roughly $150 million per day. I estimate that during the campaign against Iran, this daily rate increased to $285 million, due to the use of more sophisticated munitions, advanced military planning, logistical demands, covert ground operations within Iran, and costlier manned and unmanned aerial missions.
Based on that, my ballpark estimate for a two-week campaign likely cost close to $4 billion focused on Iran alone. Would be great to hear your thoughts.
I think this is extremely hard even to estimate it, since damages from Iran's missile strikes are top secret. Or, if you prefer it this way, Israel denies all hits except those it cannot plus those which have some PR advantage.
That seems reasonable.
I want to acknowledge the review of "The World as a Whole" - it was my favorite one this year.
The skibidi toilet review deserved to be on the finalists list smh 😔
I genuinely liked it.
Me too. I know it was silly source material, but the point of the contest is that you can review any kind of silly thing you want to. The quality of the review was pretty high, imo.
Skibidi
Scooby Doo
Scott, thanks for giving us extra time to read the entrants. I am so mad at myself that I fell Ill and couldn't make use of the extra time. I found it much harder to rate these than the book reviews.
What were people's strategies for rating things that were clearly very well written but not your interest ? Or things that were to your interest that you really enjoyed but were way shallower than other entries ?
Just wanna know if someone else figured out a good rating algorithm. I have been self-conscious about this after realising I was basically the only person who was rating Kindle books on a bell curve and inadvertently ruining the life of every author I was rating 3* (which was the majority).
Pure vibes! Rubrics are for try-hards! It feels like an 8 to me!
Here was the system I ended up using.
First I sorted all the reviews in order of most to least favourite. (I compared two reviews side-by-side, and decided which one I liked more, and then repeated that. Specifically I did it by emulating the Selection Sort Algorithm)
Afterwards, I read through my list from top to bottom and asked myself whether the reviews were similar in quality or if the latest review I was reading had a noticeable drop in quality relative to the last one I was reading. If there was, I marked this drop with a dash.
All reviews in between two dashes form what I call a “grouping”. All reviews within a grouping ended up being given similar numerical rating to one another.
My system worked out in such a way that I ended up producing 8 different groupings, which corresponded pretty closely to the 10 point scale. All ratings that were 8 or higher were confined solely within the first/highest grouping.
In terms of metrics, the only one I considered was how much I personally liked the review (relative to the others). I feel like it’s a review’s job to make me interested in its subject. If it doesn’t do so, then it is a point against it.
I know this is biased toward poorly written reviews of interesting subjects scoring better than well-written reviews of banal ones, but let’s face it, banal subjects have no chance in this contest (RIP Skibidi toilet). The only things that are going to make it to finalist anyway are well-written reviews of interesting subjects.
At least that’s how I justified it when I was designing this system. The fact that the mashed potatoes review somehow ended up as a finalist makes me very interested in learning what the author did to convince people to give it such high ratings. I await the official ACX post of it with interest.
I love reading how people take their voting very seriously and have all kinds of systems to ensure fairness and consistency. God bless all your honourable little hearts!
I just go "I liked this, eight out of ten!" Anything more sophisticated I might engage in is akin to the Eurovision voting (i.e. "this topic is one I dislike so good luck getting anything higher than a three, mate" or "ooh this is one of our friends and allies and neighbours, douze points!")
https://www.maynoothuniversity.ie/research/spotlight-research/all-you-need-know-about-eurovision-voting-patterns
For the sake of calibration, I didn't submit my ratings until I'd done 42 of them (I ultimately did 74 in all). I wound up assigning qualitative descriptions to each rating value; 1 was "I award you no points and may God have mercy on your soul", 6 was "replacement level", 10 was "deserves to win it all". My final distribution was (from 1 to 10) 1, 4, 6, 13, 8, 18, 11, 7, 5, 1.
I think I was a pretty rough rater: normed to a bell curve (ask me about my awesome spreadsheet) and penalized the ones I wasn’t interested in.
(I mean, I figured that interest has to matter, right? We want to read reviews that we want to read, not just ones that are written nicely. So if everyone rates opinionatedly, then the most generally interesting ones will do well and we’ll all be happier.)
I definitely punished shallowness (and shortness) in general: I like reviews that take up half my day to read and think about; ones that tell me a new story and teach me something about it. In the end, I think this mostly meant poorer ratings for video game, music, and tv/movie reviews, which I’m not sure how to feel about. On the one hand, it might’ve been good to see more content-type diversity among finalists; on the other, again, if we all prefer lengthy reviews of weird quirky things, why waste time and space?
We should try sortition. (link: https://archive.is/ztIRY) In the current US system, there's no way to get the best/smartest/most virtuous/etc. people on the ballot, only those with cash and connections, and how they got those usually makes them least fit to serve/lead. True representatives should be closer to average or median. We have eg Congressional Research Service that can assist. Less corruption, less extremism, less Peter principle. Maintain voting for initiatives/referendums/etc that are more important and less manipulable (as far as, ideas aren't people, so less distortion from social psychology). Biggest problem is getting it past the current system. Something something trading freedom of "we the people" for security of "politicians". More links: https://www.reddit.com/r/todayilearned/comments/1j91das/comment/mhamcg1/)
Very unpopular opinion:
Universal suffrage was a mistake; most people are too stupid to be allowed to influence political decisions.
Sortition would turn this up to 11.
Hard pass
This problem was already solved a hundred years ago in The Napoleon of Notting Hill. You simply choose the new absolute monarch via random ballot of all adult men.
EDIT: I see Deiseach has predictably beat me to the punch! Damn you, potato-American!
Great minds think alike...
.... fools seldom differ.
I'm 100% potato, no American adulteration at all!
Hmm... Any comments on the Peruvian portion of Pope Leo's career, given where the potato was first domesticated? :-)
Are we entirely sure he has *no* Irish ancestry at all? The man was born in Chicago 😁
Good question! I don't know. Many Thanks!
Why do you think it is that we have not really had a sortition system of governance (outside of juries etc) since the 19th century? Why did the places that used to use sortition (such as some parts of Switzerland) cease to do so?
I'm a fan of sortition, but I'll offer a discussion of flaws that I see in it, especially as applied to a modern large liberal democracy like the US.
High variance is a big issue. A 500-person legislature (a bit larger than the current US House of Representatives, a bit smaller than the British House of Commons, and exactly the size of the Boule of Classical Athens) chosen by sortition from any nontrivial eligible population would represent the preferences of the population as a whole with a margin of error of +/-4% if chosen at large. That's probably okay if most important issues are 60/40 or worse, especially in situations (like Athens and most other Mediterranean classical democracies) where there's a rule or norm that major or contentious issues get referred to direct-democracy institutions for ratification. But for anything close to 50/50 the margin of error is going to be uncomfortably significant, especially if the legislature is the final word on the subject.
Athens also used a system where the citizens were divided into ten phyles (sometime translated as "tribes"), each of which got fifty seats in the legislature by sortition. I'd initially assumed that the phyles were class-based like the tribes of the Roman Republic, but what little I can find about them suggests they were more in the nature of geographical districts. Either way, a lot of issues would likely see each phyle having relatively homogenous interests and policy preferences, so on those issues at least the apportionment by phyle would probably make the composition of the legislature more representative and stable.
The modern US would have trouble copying this in a robust way, since we're big enough and non-homogenous enough that our equivalent of phyles would be either too numerous or too internally heterogenous. The obvious counterpart to geographical phyles would be states, which are both, especially if you try to apportion seats by population. One representative chosen at random from Montana would be a terrible sample, while 52 represenatives chosen at random from California could skew significantly depending on how many happened to be chosen from the Bay Area, or the greater LA area, or the Central Valley, or from several other regions with distinct interests and political distributions. My understanding is that most medium states have significant regional and urban/rural divides which would produce very unstable results among the dozen or so representatives those states have.
There are other problems with scaling sortition to a larger population. Classical Athens used sortition to select a 500-person legislature out of maybe 50,000 male citizens. The eligible population was probably quite a bit less than that, since I'm pretty sure the 50k figure includes children and young adults below the minimum age of 30. So 1-2% of the population was in the legislature in a given year. Combined with short terms (one year), this means that somewhere between 30-60% of citizen men who don't die young would probably be chosen at least once in their lifetimes. A bit more or less for any given citizen, depending on how malapportioned the tribes were. And the phyles would have averaged about 5k male citizens each, small enough that you're fairly likely to have at least one current member of your legislature in you immediate social circle and virtually certain to have a friend or relative as a member of the legislature at some point over a period of several years. In the US, a 500-person legislature would be 1/500,000 of the adult citizen population, so your chance of ever serving in Congress in your lifetime (assuming Athens-like annual terms) would be about 0.0112% and your odds of knowing a member of Congress personally wouldn't be much better.
There's also the problem of legibility and trust, which is compounded when the fraction of the population being chosen is so much smaller. For a population of maybe 3k eligible candidates per phyle, it's pretty easy to design a procedure where a large fraction of them or even all of them can observe or even participate in the drawing. I don't know how they did it historically, but having eligible candidates line up and draw colored balls from a bag, or write their own names on something (not sure what: I know Athenians used potshards as ballots in other contexts, but they don't seem mechanically suitable here) and toss it into a pot that would be stirred and drawn from publicly by a group of respectable local citizens. In a population the size of the US, you could do it fairly with any of a number of lottery systems, but it would be a lot harder for individual citizens to be able to observe that closely enough to overcome suspicion that a sinister cabal was fixing the drawing.
The last problem with sortition is that elected representatives are supposed to be better than the average citizen at legislating. The ideal espoused in early modern England was "Not everyone in Parliament should be there, but almost everyone who should be there is". And in the early US, both parties expected this to happen in some form: Federalists expected voters to choose existing established pillars of the community (usually major landowners, successful businessmen, or respected civic leaders), while Jeffersonian Republicans hoped that voters would learn to discern a "natural aristocracy" of people with the requisite talents and virtues. Most Americans, if asked for a gut feel on the subject, would be highly skeptical of this in general but moderately likely to have a higher opinion of their own current representative. The latter is probably more significant, since if some random guy is preferable to your own representative, there's usually at least one or two reasonable approximations thereof on at least the primary election ballot.
Just found a possible solution to the lottery problem: Have 100 different entities in the country, preferable universities with large engineering departments each make a mechanical random number generator. Have them publish articles about hiw they work. On the day of the draw, all 100 machines will generate a random number at exactly the same time, and immediatly post it publicly. All these numbers are fed as input to a big, unpredictable, but completely open source function, which gives the final number, determining the selected person. The function is made in a way so even if you controlled 99 of the inputs the final output would still be unpredictable.
This system would be much safer than normal voting. The only way to hack it would be to have 100 groups of nerdy engineers from all over the country sabotage their beautiful random number generators at the same time.
Realistically if you had sortition then the next thing that would happen is a permanent political class of "assistants" who show up to help our naive temporary representatives with their duties. Some of the assistants will be Republicans, some of them will be Democrats, and there'll be a bunch who are "independents" (and just happen to be on the payroll of Facebook or Chevron or whatever). They'll be just itching to take all the hard work of legislating off your hands, and they totally definitely promise to let you make all your own decisions on anything important. Or you could be an assistantless representative, which will go just about as well as being a lawyerless client.
I'm speculating, but in a sortition system there will surely arise some kind of class of people who manage to become the true power behind the throne.
As I understand it, this is almost exactly how most lobbying works under the current system. Members of Congress hire their own staffs, but the budgets are relatively small, the pay scales stingy, and the staff also needs to handle a ton of constituent services and media relations work in addition to legislative analysis. Lobbyists show up offering their analysis of pending legislation and helpful suggestions for revising it, which just so happens to be to the benefit of their clients (those clients might be businesses, nonprofits, advocacy groups, labor unions, foreign governments, etc), and the workload facing the legislators and their staffs is high enough that despite the obvious conflicts of interests, they tend to rely on input from lobbyists the same way and for the same reasons that judges rely on input from the lawyers in the cases before them.
This can work out okay if there are good lobbyists representing sides whose positions bracket the truth on any given issue, and if the legislators and their staffers are reasonably competent and diligent at identifying which lobbyists are moderate enough in their dishonesty to be worth listening to. How well it works in practice probably varies quite a bit, especially to the extent that there are no doubt groups that don't have good (or any) lobbyists and whose interests thus don't get recommended to the considerations of the legislators and their staffs.
Anything close to 50/50 doesn't matter anyway. Either way, about half the population is for it and half is against it. We have much bigger problems from things like some people not bothering to vote, or in the US, all the electoral college nonsense.
Permit me to quote "The Napoleon of Notting Hill":
"The old gentleman opened his eyes with some surprise.
"Are you, then," he said, "no longer a democracy in England?"
Barker laughed.
"The situation invites paradox," he said. "We are, in a sense, the purest democracy. We have become a despotism. Have you not noticed how continually in history democracy becomes despotism? People call it the decay of democracy. It is simply its fulfilment. Why take the trouble to number and register and enfranchise all the innumerable John Robinsons, when you can take one John Robinson with the same intellect or lack of intellect as all the rest, and have done with it? The old idealistic republicans used to found democracy on the idea that all men were equally intelligent. Believe me, the sane and enduring democracy is founded on the fact that all men are equally idiotic. Why should we not choose out of them one as much as another. All that we want for Government is a man not criminal and insane, who can rapidly look over some petitions and sign some proclamations. To think what time was wasted in arguing about the House of Lords, Tories saying it ought to be preserved because it was clever, and Radicals saying it ought to be destroyed because it was stupid, and all the time no one saw that it was right because it was stupid, because that chance mob of ordinary men thrown there by accident of blood, were a great democratic protest against the Lower House, against the eternal insolence of the aristocracy of talents. We have established now in England, the thing towards which all systems have dimly groped, the dull popular despotism without illusions. We want one man at the head of our State, not because he is brilliant or virtuous, but because he is one man and not a chattering crowd. To avoid the possible chance of hereditary diseases or such things, we have abandoned hereditary monarchy. The King of England is chosen like a juryman upon an official rotation list. Beyond that the whole system is quietly despotic, and we have not found it raise a murmur."
Hard agree. My only modification would be that I think we need some degree of selection, so I would preselect for each seat say 8 people (could be any number) with sortition, then have a first vote two weeks later. The people in question can also excuse themselves if eg they are planning to leave the country, are unwell, are facing trial, or (and I hate this category) are indispensable at work (small businesses might disproportionately benefit from this which would be really unfortunate).
After the first vote, are retained those candidates who received from than 20% of the vote, but at least 3.
Those 3/4 candidates then have two more weeks in which to make their case (there could be pre-booked public or even in some cases local tv/radio events for this) and there is a final vote.
My toy idea would be to impanel some three-ish juries per district. Each jury would be empowered to interview candidates; subpoena records, witnesses, and expert testimony; and report out a nomination of maybe 1-3 candidates who should appear on the general election ballot. Or zero, if the jury fails to reach a supermajority to report out a slate of nominees. The general election ballot would then be based on candidates who are nominated by at least one jury, with some kind of runoff or ranked-preference voting to accommodate cases where the juries report out larger slates with little or no overlap.
I think that is much too complicated and loses the core advantage of sortition?
Might be a stupid question but: why do we need bunker busters for underground facilities, instead of just bombing all the exits? Is a destroyed evevator shaft much easier to repair than I would imagine?
Yes, an elevator shaft or tunnel entrance is far, far easier to repair than an enrichment centerfuge cascade. If the strikes actually managed to directly hit the enrichment hall and most of the uranium Iran had was in those centerfuges, then they have likely suffered a multi-year or even multi-decade setback. Whether or not that actually happened remains to be seen.
"Is a destroyed evevator shaft much easier to repair than I would imagine?"
It is easier and cheaper to excavate a new hole/entrance than it is to repair a mostly blown up facility (especially if the facility has a lot of expensive equipment).
Or to put it another way, Iran's nuclear budget wasn't mostly going towards elevator shafts.
The things we're looking to destroy are extremely valuable and hard to build; it would be extremely worth it to re-excavate a nuclear enrichment facility.
The act of bombing another country is generally something you don't want to have to do more than once.
I was thinking there would be an obvious threat of "if you try to dig another entrance we'll bomb it immediately", but you're probably right, for various reasons.
I understand why you thought that because it's something I thought of as well, but imagine Iran unearths the site in like 5 years when the geopolitical situation has changed and now the US is not capable of destroying it.
"...and now the US is not capable of destroying it"
Or even just unwilling.
I expect that part of Trump's calculation on what to do is considering what will happen in the next US president has a very different view of how dangerous Iran is. This would lead to choices that do NOT require follow through by future presidents.
Can a person be wrong about what their motivation or beliefs are? I don’t mean intentionally deceptive - I mean, actually being wrong in their own self-assessments. Why or why not? What evidence can a person give that their motivation or beliefs are what they think they are?
My hunch would be that anyone who says they are only motivated by pursuit of the truth is lying to themselves without realizing it, because doing so gives them social rewards and confidence, whereas truth on something you can’t directly act on has almost zero instrumental effect on pursuit of survival and thriving goals, which are terminal and therefore higher priority than satisfying curiosity.
https://graymirror.substack.com/p/the-origin-of-woke-a-george-mason
> As a proud person of TESCREAL, I couldn’t help but notice Cofnas’s model conflicts with one of the stylized facts of our traditional ways of sociology: that people generally form their beliefs not from an analytical weighing of relevant issues but based on what will help them as social creatures. I won't defend this with serious social science here because, one, I am a high theorist and it wouldn't help me as a social creature, and two, Robin Hanson and Kevin Simler do that in their magisterial 2017 classic, The Elephant in the Brain.
> Following the traditional formula of the George Mason School, The Elephant in the Brain takes a well-established mainstream concept—in this case Robert Trivers-style self-deception—and takes it to its radical but patently correct conclusions. In many ways the most ambitious of the Masonian canon, The Elephant in the Brain has such radical implications for so many areas of human life that, in less scattered times, one could imagine it spawning an era-defining psychology metanarrative like The Interpretation of Dreams or at least Thinking, Fast and Slow. The last few chapters probe self-deception’s implications for sociology, and I’d aspirationally frame this essay as a coda to those, which extends the framework to the phenomenon of wokeness.
----
> whereas truth on something you can’t directly act on has almost zero instrumental effect on pursuit of survival and thriving goals, which are terminal and therefore higher priority than satisfying curiosity.
I'm not sure I agree with you, here. But I'm having trouble putting my reasoning into words. <mumble mumble> Karl Friston and Symmetry Theory of Valence, <mumble mumble> Thompson Sampling. Also, do you remember that one Lesswrong essay about how wasps just do the same shit over and over, because the behavior is hard-coded?
Yes, easily. It comes from making simple choices where the perceived intent and the actual intent are not contradictory. The whole idea behind thought experiments is to probe at those weak points where the ideas conflict, and see if what you think is actually what you think you think.
(personal example; I used to think I wasn't picky about clothes, until someone took me to a shop containing nothing but shirts with large phrases written on the front in bold, and I realized that actually I have a strong preference for expressing as little as possible with my clothes.)
Yes. A lot of my personal growth has come from realizing that the reasons I think I do certain things (like fight with my wife, or avoid making new friends, or play video games) are not the actual reasons. As an example, this morning I started to get snippy with my wife when she said "Trust me, that was the wrong was to do it." If you had asked me five years ago why I was mad at my wife I would have said that she was demeaning me, or putting me down. This morning I defused it when I realized that I was getting upset because I was feeling incompetent and incapable and the words "trust me" seemed to imply that she thought I didn't trust her and was screwing up the task out of defiance. It's more complicated than that, but the incompetence thing is a real emotional trigger for me, and I figured out several years ago that my fear of my own incompetence was the real reason I did a lot of things that before I would have chocked up to my sense of justice, or needing to defend myself, or wanting to set the record straight, etc. I was able to turn my passive aggressive comments into direct communication to her asking her if she really thought I didn't trust her or didn't know that what she was saying was true, and I found out that she wasn't even aware that she had used the words "trust me" and was just trying to help.
A lot of philosophy and religion is tied up in the idea that we don't understand ourselves and why we really do the things we do or feel the emotions we feel, etc.
Yes, absolutely. Ask anyone who has been a parent or other caretaker of young children. Kids are demonstrably and trivially wrong with extremely high confidence about their own motivations and beliefs all the time.
Yeah, until I had kids I was unaware how much of introspection is actually a learned skill.
Small kids literally do not realize that they are hungry or tired... they just get increasingly angry, and they have no idea why. Like, a hungry child will get angry, and will then throw food on the floor because of the anger, which goes completely against logic. It takes a few years until children learn to recognize "I am hungry now" and "I am tired now" as emotions.
Actually, I suspect that even adult people eating food at midnight is often not caused by actual hunger, but by misinterpreting signals of tiredness for hunger. Both of them are "low on energy", but for different reasons.
So if kids need to learn all of this, it makes sense to suspect that maybe there are things that even many adults do not know. Signals that mean something specific, but we can't decipher them, and it becomes just a generic anger.
I don't think a person can be "right" about their motivations or beliefs.
Step 1 in thinking about your motivations and beliefs is: condense/compress/collapse the totality of your motivations and beliefs into something that fits into a Conscious Thought™.
Then if you're trying to communicate your motivations and beliefs to someone else, step 2 is: condense/compress/collapse the totality of your conscious thoughts about your motivations and beliefs into something that fits into some sentences or paragraphs of whichever language you're using (plus whatever grunts, hand gestures, poetry verses, oil paintings, vlog posts, equations, etc that you have time produce and that you think might help.)
The amount of simplification required for each of these steps is so extreme that the product of each step is more "wrong" than than it is "right."
On the one hand, yes, i love it.
But on the other hand, if I think "maybe my keys are on the desk" and then I walk to the desk to get the keys, I think i'm right that my intention in going to the desk was to get the keys.
I suppose, to your point, if walking to the desk brings me by the pantry, which is open, and hey, macademia nuts - i can't know for sure whether or not the keys were instrumental to the nuts.
Is there some way to have this cake and eat it, too? Can we evolve it to, "knowledge of your own motivations or beliefs is likely only possible in small scale, limited contexts and even there is questionable?"
For sure.
On a small scale [why did he walk to the desk?] you're probably a lot more "right" than "wrong" when you think and speak about the belief [he believed the keys were on the desk] and motivation [he wanted to start the car].
But we rarely think/talk about beliefs and motivations at that small scale. Usually we're thinking about something interesting/uncertain/complex/controversial. And yet when thinking/talking about these larger scale, broader context issues we, unavoidably, perform the step 1 and step 2 dramatic oversimplifying processes mentioned in my previous comment *without noticing* that we're doing it.
Or to put it another way, I think it's a mistake (that we all make, almost all the time, without being at all aware of it) to think the "knowability" of a thing like [why did he walk to the desk?] is in the same category as the "knowability" of a thing like [is he motivated by pursuit of the truth?].
Epistemic status: Neither immaculate nor defiled ;)
"Can a person be wrong about what their motivation or beliefs are? I don’t mean intentionally deceptive - I mean, actually being wrong in their own self-assessments. Why or why not? What evidence can a person give that their motivation or beliefs are what they think they are? "
If a person states (and seems to be sincere about) that they value "X" and they keep making choices that are anti-X at some point I feel one must conclude that they are either:
a) Very wrong about their valuing X, or
b) Terrible at understanding how the world works (in at least one area of concern).
I'm eliminating "they are hypocrites" here because of the "seems sincere" bit, but if we don't have that then them talking the talk but then doing what they really want also comes into play. But that isn't your question.
I think that (b) is VERY common in other areas (e.g. most people seem to have cause-and-effect for athlete salaries and event ticket prices backwards...) so I don't see why it would be less common for motivations and beliefs.
What is probably also very common is people not wanting to make tradeoffs.
> e.g. most people seem to have cause-and-effect for athlete salaries and event ticket prices backwards..
can you say more about this?
Sure.
A common belief I see on various message boards is something like, "Great, now that player X has a new $YYY contract ticket and food prices will be going up again."
Almost certainly this is not how things work.
Consider the ticket (and food and parking) prices and player salaries at some time (like today). Say tickets average $100/ticket. And the stadium sells our 85% or whatever.
Imagine, next, that the entire league takes a 25% pay cut. Maybe there was a short strike and the players lost or the owners staged a short lockout. Whatever. There is a drop in pay, but during an off-season so the fans aren't upset because a season got disrupted. But now we have a new set of contracts and the same players (more or less) show up to play.
The fans have already established that they will pay $100/ticket to watch the players play and the same players are playing. Do we expect the owners to drop the ticket prices to $75/ticket or keep the tickets at $100/ticket and pocket the extra money?
The folks who believe that player salaries drive ticket prices expect that the owners will drop ticket prices. Economists, who expect the owners to be trying to maximize revenue, expect that the ticket prices will remain the same because the customers have established that they will pay $100/ticket so why drop prices?
The same thing plays out in the other direction. If players get more do we expect ticket prices to go up? If so, why weren't the ticket prices at that place already?
The *other* view is that the owners charge what they can for tickets and parking and food and the players get some cut of that (about 50% for the NFL). So if the fans stop showing up then ticket prices drop and eventually player salaries drop, too. If fans keep showing up even if/when ticket prices rise then the players' salaries go up because the owners have more money to spend on player salaries and many of the owners want to win (also, many leagues have salary caps and floors that are based on league revenue).
Economists generally believe that rising ticket prices drive rising player salaries rather than the other way around. Lots of fans think the causality runs the other way.
Yeh and I think this matters in the arguments that minimum wage increases will be inflationary. Only if the businesses can get away with it.
I would have thought "high athlete salaries = higher event prices to cover the costs" but is it in truth "high event prices = can then pay higher salaries to the stars"?
Yes, "high event prices = can then pay higher salaries to the stars" is more accurate. I elaborated a bit more on why/how in a response to Mark Neyer.
I sort of agree with your conclusion, but not with your argument for it.
We absolutely do not live in a world where people only act on things that have instrumental effect on pursuit of survival and thriving, and there are plenty of examples of people choosing to satisfy curiosity over survival (functionally, anyway, usually they are not aware of just how badly their curiosity will backfire).
> usually they are not aware of just how badly their curiosity will backfire
Yes but i think this is the key: in these situations, they didn't understand their risks. I think people understand 'risk of upsetting my group' FAR more intuitively than they understand other risks. I don't think people risk tribal membership ejection nearly as readily as they risk e.g. bodily harm, because you can heal a broken limb far more easily and reliably than a broken reputation.
This question is tough for me to get a handle on. In part, I think that's because the line between belief and unbelief isn't well-defined. In part, that's because belief has an important role in group membership. Robin Hanson's idea that it's easier to talk about group values by gesturing at something sacred rather than just being explicit about those values is relevant. In that way, belief is a kind of social ritual, where we don't just privately believe and suspect that other people privately believe, but we see each other making professions of belief.
I think too about how important COVID denial and COVID vaccine skepticism are as markers of membership in certain groups. People literally died for those beliefs. It's hard to think of a stronger indicator of belief than this. At the same time, I'd say that they engaged in some serious motivated reasoning to arrive at those views. That points to some shakiness of belief. They made a much bigger bet on those beliefs than any bet I'd make that they didn't really believe those things, though.
> People literally died for those beliefs
I don't think this is true. They may have died _as a result of those beliefs_ - but dying _for_ a belief means taking a risk you believe to be fatal, due to the belief. If you think COVID was all exaggerated, you aren't taking what _you_ believe to be a fatal risk, regardless of what it is.
From their perspectives, people who took the covid vaccines themselves were taking on risk for their beliefs: they took a mass-manufactured chemical product made by corporations that lie, in order to avoid something with a low fatality rate. They'd say _you_ engaged in motivated reasoning.
It seems to me, then, that whenever we look at reasoning of people with different value structures, we think _all_ of their reasoning is motivated, whereas our own is pure and based on the truth. To me it seems most likely, then, that it's always subservient to group membership.
Sometimes the part of your brain that rationalises decisions is not aware of the part driving the decisions (hunger, and so forth), so I would say not only is it possible but relatively common.
I’d go farther and say that low level self awareness is something of a rarity.
I think people can and often are wrong about these things. Another thing they're often very wrong about is the origin of their beliefs and motivation.
That said, I wouldn't automatically discount as false a claim about being interested in finding the truth. I know I am personally interested in discovering the truth even when there are negative social consequences for doing so.
Of course, I could be wrong about my own motivation, but if you want to place this in the context of the surviving and thriving framework, curiosity often leads to unexpected discoveries that have a huge survival advantage but it is not possible to predict initially which discoveries that would be and how to even begin to go about making them. All you can do is try lots and lots of things and try to make connections between those things and notice patterns etc. So it seems to me that an innate drive towards general curiosity without any underlying social motivation is something evolution very well could have produced.
I agree that this is true - curiosity is totally a thing. But I think people will generally avoid being curious about things if that curiosity itself poses an existential risk to them. Given how socially oriented we are, being curious about, e.g. the relationship between genetics and iq can pose an existential risk.
Any folks close to the topic have thoughts on the FDA commissioner’s plans to cut drug approval time by months? Are they doing it “right?” Estimated lives / QALYs saved?
https://www.washingtonpost.com/opinions/2025/06/19/faster-fda-drug-approval/
It's more talk than concrete action at this point. The FDA unveiled their new AI chatbot "Elsa" recently to help cut down review times, and despite the hype marketing from the commissioner, the insider sentiment was far less bullish. That said, there is no reason why AI should not significantly reduce review times if implemented properly, but the FDA is more in the vision / intention setting stage at this point. And most of their accelerated review pathways they have created give them all sorts of outs if they, in fact, cannot get it done.
Prediction Market for the winner when? My money is on The ACX Commentariat
What do y'all think about this foreign policy for the Middle East from the perspective of the USA?
1. We must achieve complete independence from reliance on Middle Eastern oil by any means necessary--offshore drilling, fracking, investing in nuclear power, solar, wind, whatever.
2. Then we get out. Completely. Iran closes the Strait of Hormuz? Fine, we don't need the oil from the Persian Gulf. Houthis shooting ships in the Red Sea? That's not our main shipping lane anyway. Someone attacks Israel? I'm fine with selling weapons to the Israelis, but they've done a good job of defending themselves over the past 60 years. They can survive without our help. Palestinian statehood? Let someone else figure it out.
Pivoting away from the Middle East seems like a good idea for the USA and many Presidents have endorsed it, with little success.
That said, the USA has strong opinions about all parts of the world from Outer Mongolia to the Horn of Africa, so 100% disengagement is impossible.
As others have noted, the US has already pretty much stopped using Middle Eastern oil, and is now competing with the Middle Eastern customers in the oil-export business. 1973 was a *long* time ago.
But the US economy is absolutely dependent on a great deal of continued trade with other nations which are dependent on Middle Eastern oil. And that's not going to change any time soon. Trying to turn the United States into a North Korea style autarky what doesn't need nothin' from nobody, without in the process impoverishing Americans to the point where a populist landslide will bring in a President with a mandate to reintroduce globalism, would be the work of decades. And we're almost certainly not going to stay focused on the project that long.
For the foreseeable future, Americans will suffer significant economic hardship (and the politicians responsible will risk electoral ruin), if Middle Eastern oil stops flowing.
The US is already independent of oil from the Middle East, but US politicians are not independent of donations or votes.
1. Oil is a commodity. "Independence" is impossible unless you're willing to set up a system where the price of oil is different from the rest of the world. Which basically means complete autarky and is impractical. If you simply want the US to not be directly reliant on it then the US already has that and always has. I think the Middle East is like 5% of our oil.
2. The US is highly integrated into East Asia and Europe which DO rely on Middle Eastern oil. They also rely on those trade routes. Unless you can get the EU, China, Japan, etc to sign on to energy independence and avoid their biggest trade routes you'd still have to care. Now, you might think "if they're primarily the ones benefitting why don't they help" and in that case welcome to the American diplomatic position since like the 1980s.
3. The Middle East is important strategically if you want to defend Europe or North Africa and it gives us a strong ability to influence South and East Asia in one shared defense perimeter that's continuous with Europe and North Africa.
The folks pushing energy independence often want to see Europe and Japan [maybe not China!] pay "their fair share" for the naval power to keep sea lanes (including the Persian Gulf) open. This tends to get lost, but should be kept in mind while thinking about this.
"Why is the US paying our Navy keeping the Persian Gulf oil flowing to China?" probably captures the spirit of this (whether the facts are correct or not).
Re #1, we are already close enough to independence. https://www.eia.gov/energyexplained/oil-and-petroleum-products/imports-and-exports.php
The main benefit of middle east right now is a source of investment, and a lax regulatory regime. We can build all the AI plants we want in Saudi Arabia.
The cost is jihadi fundamentalism willing to blow stuff up to gain god's favor.
If we withdraw China will swoop in and replace us.
We stopped trying for boots on the ground regime change, and effectively stopped mediating in Israel/Palestine except insofar as it improves regional stability (i.e. Abraham accords great justice for Palestinians who cares)
Couldn't be more satisfied with the current approach, actually
>If we withdraw China will swoop in and replace us
But why does that matter, given OP's implied premise?
I believe the argument is that the US gets to stay the world hegemon (also) by keeping others down. They can't afford having a competitor that has access to both advanced technology and plentiful resources.
Yes, that is obviously the argument. But, again, why is it important to be the world hegemon? If I read OP correctly, they don't think that is intrinsically valuable. And after all, it is safe to say that, of all the people in the world who are living happy lives, most do not reside in the world hegemony. So, again, why does it matter?
I guess the elites who profit from controlling a country want it to be strong so it can extract more wealth for them? Also not being the hegemon does not mean you can stay comfortably on the sidelines as number two or three or four. Germany seemed to be doing fine as not-the-hegemon until they got their energy supply cut off by the nord stream 2 pipeline sabotage. Now they have no recourse.
Leaving aside that "being the hegemon" does not mean that you control other countries, I am not asking why elites might care; i am asking why any of us, such as the OP, should care.
I don't understand. The Gulf states offer the US stuff it wants: a lot of space and lax regulatory regime and energy that they are willing to trade in exchange for military protection. If we don't do it China will offer it instead and we would be relatively weaker.
Having power and a higher share of global economy/effective resource control without giving up nice things like being able to stop government from bulldozing my house with eminent domain to build an AI chip fan seems good for me/US foreign policy
>a lot of space and lax regulatory regime
I don't understand why you think the US wants those things. Those sound like things that US companies doing business overseas might want, but not the US govt per se. Nor do I understand how "China swooping in" would change anything in that regard.
>energy that they are willing to trade in exchange for military protection.
But that is not how it works, is it? The Gulf states sell their oil on the open market' they don't trade it for military protection. And the US import very little oil from the Persian Gulf https://www.eia.gov/dnav/pet/pet_move_impcus_a2_nus_epc0_im0_mbblpd_a.htm
Who cares about oil US is self sufficient. I was talking about industrial and land capacity.
I'm saying Saudi Arabia is willing to build giant AI chip and energy facilities that the US can't for environmental regulatory reasons.
The US government of course cares about maximizing economic and strategic leverage of US aggregate entities.
So, your claim is that, if the US conducts foreign policy in the Middle East that is to Saudi Arabia's liking, it will let US companies build AI chip and energy facilities, but if not, then China will "swoop in" and replace the US as Saudi Arabia's security guarantor, and Saudi Arabia will instead let Chinese companies build those facilities? Seems like you are positing a force projection capacity that China does not currently seem to have.
And, why not build those factories in Mexico or elsewhere in Latin America?
How hard is it to build AI plants in rural Arizona or Texas? I have my share of problems with using eminent domain, but there's a lot of empty land in the US.
Impossible. We wasted billions and billions of dollars trying to build high speed rail, etc.
It's not about 'empty land' it's about bypassing environmental regulations.
That is a not a very apt comparison. High speed rail is built with government funds and hence is subject to political considerations (eg, in California, the requirement that it actually serve the people who need it most, rather than bypassing them). And both Arizona and Texas have grown rapidly; there seem to be few constraints on economic development.
"Impossible," they say, while OpenAI currently builds the massive StarGate plant in Texas: https://www.youtube.com/watch?v=GhIJs4zbH0o
This seems out of date and probably bad.
Due to advancements in energy production (something something fracking, ask somebody actually in the energy sector) the US is a net energy exporter. We probably still want access to the Middle East but, honestly, if the Middle East shut down it would probably make our own oil and energy reserves far more valuable.
The issue is that, well, the rest of the world still needs cheap oil, which means the Middle East. A world in which cheap oil keeps flowing to Europe and developing countries is a better world because developing countries deserve a decent standard of living and because we like Europe, it's full of the history and funny accents :).
But yeah, fundamentally, the US no longer needs the Middle East for its energy needs and is, at best, providing a free security good to the rest of the world and, at worst, meddling in a volatile region out of habit and the influence of varying lobbying groups (Israel, yes, but also the Saudis and Emirates and their ilk).
Wow, I think that, of the finalists and honorable mentions I only read the Watergate one (it was really good!). Good to know, there is a lot of good content to be read on the coming weeks then.
Shout-out to the review on human sexuality. Thought it was well written and had an original deep insight, even if it seems a bit flawed/incomplete. Would really have liked it to see a discussion here of it
Hey, author here; thanks for the shout out. Definitely ran out of time on this/was a little rusty at essay writing, I don't think I quite did a good enough job conveying the original insight, which is a shame. But it was still really fun to have written it, to have put something out there that was read by probably hundreds of people.
Maybe there can be a bit of a discussion here; the essay is hosted at https://cosmiczenithcanon.substack.com/p/acx-not-a-book-review-entrant-review.
The short summary/TL;DR is that human sexuality is fundamentally tied into our status detecting systems, and this explains most of the weirdness around sex, such as people being into being cucked or the prevalence of violent fantasies.
What parts of it seemed incomplete/ in need of elaboration to you?
Im going to comment both here and on the post. Who knows, maybe the algorithm takes this and signal boosts your post
I really enjoyed the status approach to sexuality but I still think what you call low-status hotness (LSH) is kinda mysterious
While I buy the idea that submitting to a high status partner and surrendering responsibility may be hot or attractive to some people, I think there is more at play in some of these fetishes. It seems like the very act of lowering one's status (either physically by pain or socially by humiliation) is the goal of the drive here, not the second-hand consequence of submitting.
This part, where pain or humiliation turns into pleasure feels like a gap not only in your system but in most explanations of human sexuality imo. I think without explaining this gap some extreme fetishes like straight up CBT or SPH remain unaccounted.
I dont know, maybe by taking pain or humiliation you increase the "status-delta" between you and your partner and it makes them hotter? This could be a viable explanation but still seems to ignore the fact that the pain is the goal for some people
Yeah, I would probably phrase the idea of low-status hotness as being about wanting that "status-delta" to be as large as possible, of wanting to be around an incredibly high status partner in a sexual context. This can be done by either having the high status partner act even higher status, by inflicting pain or punishments or giving affirmation or restraining, etc, or by having the low status partner become even lower, via receiving physical pain, social humiliations, being restrained, etc.
These often blend together; when a Dom is tying up a sub, they're increasing in status in the scene, while the sub is being reduced. I perhaps didn't clarify in the essay properly, but yeah, I think goal of the drive is to have that status differential, not the second hand consequences.
As for pain being the goal for some people... the people who I know, who have masochistic tendencies, are usually only into pain in certain specific contexts, which almost always involve a specific person hurting them. They're into being spanked, perhaps brutally hard, but they're not into stubbing their toe on a door jam repeatedly. That makes me think that pain is downstream of the status differential, not the end goal in itself.
Surprised that the Testosterone review didn't make it and kinda disappointed (1). I thought it was well argued and potentially worth making the lifestyle change and I wanted to see the ACX commentariat tear into it and see how it fared.
https://docs.google.com/document/d/1jYVJFIz5-aMi0LCgsC9AN6BncJDNVGaMU37QmwZ1vzA/edit?tab=t.0#heading=h.c6lh7sev0o6s
(1) I say that and then I'm like "Oh yeah, I didn't vote".
It was one of the good ones. I did have an issue with it: I didn't know whether I should trust the author to be unbiased enough. The author was certainly very convinced by the conclusion. They did discuss some drawbacks, but I didn't know enough to say whether it was a fair treatment. They were very dismissive of the downsides, and it sounded very surprising to me that this should be all that there is.
I did give it a good grade. I would have liked the review to reach the final, so that I learn more about it from the comments.
> They did discuss some drawbacks, but I didn't know enough to say whether it was a fair treatment. They were very dismissive of the downsides, and it sounded very surprising to me that this should be all that there is.
Author here. I probably didn't do a good enough job, I wasn't trying to dismiss the downsides, I was more trying to show that they exist, but mainly affect people who do a ton of it over time.
I basically tried to show that even at supraphysiological doses, there are essentially no side effects in all the studies we have. Now crank that down to a TRT dose, and the risk is significantly lower again.
Best of all, the downsides are all measurable and mitigable. So it should be quite safe in expectation to start, and as long as you're measuring the potential downsides (blood pressure, cholesterol, etc) on a decent cadence, it's essentially a no-to-very-low risk to try intervention for a potentially large quality of life bump.
No, my formulation was not good, you did express that. It's more that I would have like to hear a second opinion on the matter. Especially because it's in the general field of medical/nutritional intervention where there is a lot of biased advise. And it's a bit of an extraordinary claim: the positive effects are strong and very notable, and the negative effects are pretty negligible. That doesn't mean that it's wrong, and it makes it all the more important if correct.
It was a really interesting review. I won't act on it right now because I suspect that my testosterone level is still alright. But I will certainly not forget it, and may act on it in the future.
Protip: Ask your doctor to check your levels now, so that you've got a baseline for when you get older.
I skimmed it, got really interested and promised myself that I would go back, read in detail and rate it. Oh well. At least you reminded me to read it fully, I really started considering the lifestyle change as well
Ah, ACX commetariat, I am disappointed that "11 Poetic Forms, Reviewed in Verse" wasn't a finalist. It was so fun! I didn't expect it to win but I really hoped it would at least place.
Agree! I don't even like poetry very much, but I quite enjoyed that one. I really thought it deserved to be a finalist.
Completely agree. It was one of my two 10s. Hats off to the writer.
The concept was good, and probably if Scott had written it it would have been great, but the author doesn't have quite that level of verbal cleverness and so it didn't make me smile as much as I'd have liked. This kept me from rating it more highly.
I'm not a lawyer but what's wrong with the YC SAFE for #1?
Doesn't work if the recipient is currently a nonprofit, I think.
My friend is a journalist and he has been writing about the US Aid cuts in Southern Africa. He has mainly been focusing on South Africa, but recently went to Mozambique to see the effects there.
It seems to be nothing short of a total catastrophe. As in, tons of children just dying of HIV levels of catastrophe. The country's health systems seems to have been entirely propped up by USAID funding.
There are certainly debates you could have as to whether or not it is good for a country to be so reliant on USAID, but unfortunately it was and with the cuts, we are now seeing dire consequences. I've linked the article below if you want to read more. Was hoping some EA type people could see this.
https://groundup.org.za/article/mozambique-these-children-usaid-left-to-die-part-1-the-abandonment-orphans-with-hiv/
So the NGOs could not bear even a slight disruption in funding ? Perhaps they prioritize CEO salary over the dying children?
It was quite amazing that the sob stories began to appear right after the aid cut. It didn't even take a week for people to start dying the world over.
The Copenhagen Interpretation of Ethics strikes again.
https://web.archive.org/web/20220705105128/https://blog.jaibot.com/the-copenhagen-interpretation-of-ethics/
China puts hundreds of billions of dollars into neocolonial resource-extraction "development projects" in Africa, but somehow doesn't notice all the HIV. I guess the AIDS crowd don't work as hard in the mines. The European Union, richer than the United States and as they incessantly, sanctimoniously remind us *far* more morally righteous, pay a pittance to AIDS relief in Africa but really that's beneath their notice as well.
The United States, alone, sinks a hundred billion dollars into beating AIDS into submission across the African continent, and when it finally says "enough of this, time for the rest of you lot to take up the burden for a few years", *that's* the morally outrageous part.
The only winning move is not to play.
Either they're a dependent satrapy or they're not. They can't have it both ways. Well they can, but then hundreds of thousands of their people will die. And they will blame the US for it. And most people will believe them.
Without forcing people to virtue signal, I am amazed at the types of comments a post like this gets.
It reads like intellectualizing. You even explicitly say "There are certainly debates you could have as to whether or not it is good for a country to be so reliant on USAID, but unfortunately it was and with the cuts, we are now seeing dire consequences."....and the commenters jump exactly on that part.
Kids with AIDs are dying. Skilled forecasters estimate that the PEPFAR stop due to claimed "fraud, abuse and waste in USAID" will directly cost the lives of ~half a million people per year.
https://asteriskmag.substack.com/p/forecast-the-impact-of-usaid-hivaids
Deaths after one year: [257,500 - 515,000 - 772,500]
Deaths after four years: [1,015,000 - 1,998,750 - 2,953,750]
I will repeat- half a million lives per year. Even if you disagree with USAID, or think there is actual abuse, or there is nothing good about foreign aid, or it can make countries dependent, or it can foster some corruption in countries, I posit that half a million dying is an moral outrage. I would urge any commenter to recalibrate their moral compass if this is one of the first responses you have to reading the associated article. Hope indeed some EA funders will see this.
a. Are there any circumstances in which, having started this program, the US would be morally allowed to end it?
b. Are there any other countries that are obliged to pick the program up now that the US has ceased funding it?
I understand the walking-by-the-drowning-child argument for why we are obliged to keep doing this program. But that seems like it applies as much to every other country rich enough to fund it as it does to us.
And if we're not doing the walking-by-the-drowning-child argument, then we seem to be, as John pointed out above, firmly into Copenhagen ethics--we are obliged to keep helping because we started helping, the UK, EU, China, Saudi Arabia, Qatar, Japan, etc. aren't obliged because they never started.
Completely agree. It also always surprises me seeing these kinds of responses on ACX, given Scott's strong views on things like pepfar and EA
Are you familiar with the term "hostage puppy"?
> I posit that half a million dying is an moral outrage
I think the actual moral outrage here is that these people are being born in the first place when there is simply no need or future for them. These people are a cultural and genetic dead-end. They do not need to keep existing. How long are you going to keep throwing money into this black hole?
These people have more genetic diversity than the rest of the world combined. that's hardly a "dead-end" like cheetahs, which bottlenecked quite severely....
> These people have more genetic diversity than the rest of the world combined
And do you not see how that might be a problem in an environment where large, practically eusocial societies have a massive advantage over smaller ones? Diversity is not strength, it's a cancer that eats away at itself.
Of course, none of that matters if the quality of the people themselves are lacking.
Sigh. As noted earlier, "large societies" existed in Africa. It wasn't all tribes, like North America pretty much was.
North America, prior to Western Conquest, consisted of tribes slowly devolving out of being human (losing words, in particular).
Diversity in terms of genes insulates you against disease, and can provide "rare but useful" mutations, which are useful when a "big change" comes and makes most of your population "less apt to survive."
Folks that come from mountains have a very different psychology than most. This diversity is something you'd probably celebrate, as it's adaptive (and helpful for when we colonize Antarctica).
Half a million dying from an intentional, created illness is a moral outrage. Half a million dying from influenza is just an ordinary decade (and, judging by the vaccine's effectiveness, not preventable at any rate).
Africa Wins Again is a perfect slogan for the deadliness of Africa-in-general. Lions, snakes, sleeping sickness, monkeypox, ebola...
I will not be outraged because of something that I did not do. Passively letting someone die, who I could save, is not a moral outrage. It is a moral lacuna. It is not deliberately performing genocide.
Let me be perfectly clear: The perfectly preventable deaths of Ukrainians from 2022 to 2025 are on OUR HEADS. We voted for them, we got them -- blood and gore. Even the soldiers infected with HIV, deliberately, are on our heads.
I will NOT be shamed for letting someone die from their own actions. Shame on America, shame on US, sure, but shame us for what we've done. It's far more shameful than our inaction.
I'm confused. Why are the deaths of Ukrainians on our (US?) heads? We... didn't kill them? You mention that those deaths were "preventable", but isn't that exactly what OP is saying about the deaths in Africa.
Deliberate infliction of casualties with poor tourniquet usage, among other things. Deliberately giving them HIV-contaminated blood... etc.
https://pubmed.ncbi.nlm.nih.gov/38242075/
Also, when we drive the Ukrainians towards our objectives (Sevastopol), at the cost of 90,000+ dead (surprisingly, those are the Russian numbers, and they're somehow lowball, I do not understand this -- surely someone else is giving lower numbers than the lying Russians??!) -- the counter-offensive was essentially a suicide attack.
Which part of this is deliberate?
Giving them HIV-contaminated blood was deliberate. So is the training given, and more importantly the lack of reassessment and retraining as the number of amputees rose. So too is demanding that the Ukrainians do a suicide attack... (and completely blowing up the peace talks, because of overconfidence that we really could get Sevastopol).
> Skilled forecasters estimate that the PEPFAR stop due to claimed "fraud, abuse and waste in USAID" will directly cost the lives of ~half a million people per year.
The article you link to doesn't say that. It estimates what will happen if the waiver PEPFAR has been granted that allows it to continue operations doesn't work and PEPFAR is effectively cancelled. But PEPFAR is not currently stopped, though there were disruptions to its operations that are still being worked out.
Correct, I should have made that explicit. From the sources I have read, and the above-linked article, the fact that a waiver exists actually does not mean PEPFAR is continuing. Supply chain projects are not being picked up.
https://retrovirology.biomedcentral.com/articles/10.1186/s12977-025-00657-2
https://www.bushcenter.org/publications/two-minute-take-pepfar-update
https://asteriskmag.com/issues/10/how-to-triage-billion-in-aid-cuts (slightly more hopeful, people are stepping in)
Edit - for more finegrained levels of cuts:
https://www.cgdev.org/blog/new-estimates-usaid-cuts
https://www.cgdev.org/blog/usaid-cuts-new-estimates-country-level
At different points in this century, the UN has projected a 2100 population for Africa at 2 billion and 4 billion. I don’t know what they’re thinking now or if there’s any reason to pay attention to them.
But offering such numbers in this context, is not necessarily bound to shut down discussion of the future. Especially when tagged with “how could we be missing 5 million people?” etc.
A country that cannot fund a bookmobile to drop off in rural areas the ARVs it is sitting on, is unlikely to know how many people live in it, nor what they die of when they die.
Reading that, one has to wonder what the hell the government of Mozambique is doing. If I look up Wikipedia, it is a poor, underdeveloped and indebted country. On the other hand, apparently with development policies and loosening up regulation, the economy grew very fast. One more case of "the GDP is soaring but people think they're not doing well"?
So if the economy *is* growing, where's the money going? Corruption, it would seem, as well as the lingering effects of the civil war and new uprisings in the north. (Thank you Google AI for the précis):
"Mozambique's economy in 2025 is experiencing a mixed outlook. While facing contraction in the first quarter, there are projections for a gradual recovery, with expectations of 2.7% to 3.0% real GDP growth by the end of the year. The recovery is expected to be driven by the extractive sector, particularly gas production, and agriculture. However, challenges remain, including inflation, fiscal deficits, and the ongoing impact of the Islamist insurgency in the north."
Apparently it is also a popular tourist destination. So there is *some* money there, but who holds the purse strings or where it's going is the question.
https://www.afdb.org/en/countries/southern-africa/mozambique/mozambique-economic-outlook
"Mozambique’s economic structure is determined largely by the extractives sector, with its large gas reserves of 180 million cubic feet, the third largest in Africa. There has been only limited structural transformation of the economy, with a slight shift from agriculture to services. In 2001, the services sector was the main contributor to GDP growth, accounting for 50%, followed by agriculture at 32.2%, industry (including extractives) at 18%, and manufacturing at 12.2%. Two decades later, economic activity continues to be driven by the services sector, which accounted for 51.7% of GDP in 2021, while agriculture’s share had fallen to 28%. Mozambique’s structural transformation is hampered by skills shortages and a high illiteracy rate among 15- to 34-year-olds (31.4%), skills mismatches, lack of infrastructure, high public debt, and a low level of industrialization."
According to the linked story, once the USAID money stopped flowing, that was it. Nothing happened. Nobody else stepped in. No home-grown solutions or attempts by the government to intervene, no other agencies or charities. If the government is this bad, then the country might as well become a satrapy of the USA
Cogently written as usual. And the medicine is still there …
More specifically what happened is the government basically borrowed a bunch of Chinese money because they didn't like that the US wanted things like transparency or anti-corruption. They even cut services and expected the US to pick up the slack. They were betting the US would keep funding them because of stuff like... well, this story. Basically that no matter how much they favored China or cut their own public services the US would continue to pay because otherwise ordinary people would suffer.
But then China pulled back on funding as it became increasingly clear Mozambique was not going to be able to repay them or develop. And then the US cut aid. At first a little to signal displeasure then when the US did zero based budgeting it couldn't justify continuing in Mozambique generally. And now Mozambique is left in a deep crisis. It's unfortunate for the average person but it's the result of extremely bad local leadership.
I imagine China would have been interested in things like the natural gas reserves? When you've managed to frick up your government to the point that it's not even worth it for a colonising power to come along to scoop up your natural assets, that's pretty bad.
I realise they're still struggling with an underdeveloped country and a past civil war plus new insurgency in the north of the country, but it does sound unhappily like much of post-independence Africa where the new rulers decided they'd divvy up the spoils amongst themselves, and to hell with the citizens, they're not even all our tribesmen.
Mozambique does have natural gas but not a ton of it. It's ranked about 90th if I recall correctly, something like .03% of global production. That places it alongside such energy superpowers as Latvia. Most of Mozambique's exports are agricultural products in a broad sense. Stuff like nuts but also wood. And all exports are only about $12 billion and exports to China are only about $750 million. Mostly niche agricultural products and high quality lumber.
China's plan for Mozambique was, fairly transparently, "We come in and run the Chinese economic playbook. You undergo rapid industrialization to middle income and reorient your trade to focus on China." Not a bad idea overall except the part where none of that actually happened for a variety of reasons. But mostly, in my opinion, corruption and more general dysfunction.
I will say Mozambique is uniquely bad even among dysfunctional African states. It's undoubtedly in the top three and has an okay shot at being the worst. Though that honor probably goes to Republic of the Congo (not the DRC, the other one). But it's close.
There does come the point where you have to ask if USAID and the likes are doing more harm than good; if the attitudes of the rulers are "why should we even attempt to set up a functioning country, instead of just robbing the treasury blind? the bleeding-heart Westerners will take care of our people for us, as well as shovelling aid money in that we can keep creaming off", then is it in fact doing more harm than good?
Once again, I find myself in the very discombobulating position of agreeing with a Shoe0nHead video:
https://www.youtube.com/watch?v=gUOwPcfc1MM
Mozambique is a particularly egregious case and that period was relatively brief (about 5 years) before they got cut off. Also they didn't directly steal western money. They also wanted to take money from the Chinese. And in the end both pulled out. But yeah, that was more or less their plan.
This is why USAID is so big on democratization. They believe, rightly or wrongly, that this kind of corruption stems from authoritarian governments (which Mozambique has). And to be fair to that point of view, more democratic countries mostly are much better about this kind of stuff. Kenya, for example. Though also authoritarian Rwanda.
On the one hand, there is significant human suffering. These individuals are suffering and that is obviously bad.
On the other hand, this is a predictable consequence of Mozambique's actions which have repeatedly taken the US for granted and prioritized China. The reason the US funds Mozambique's healthcare system is they cut domestic funding in order to repay China and expected the US to pick up the slack. It was supremely foolish of Mozambique's leadership to expect US taxpayers to (indirectly) finance Chinese debt repayments.
China owns about 7.5% of Mozambique’s debt (or about 14% of the Mozambique’s external debt). Government spending on healthcare increasing from 2016 though 2022, going from 29% to 36% of total healthcare spending. The World Bank doesn’t have numbers more recent than 2022, but I haven’t seen any reports of subsequent cuts.
Eliminating healthcare aid to Mozambique in the hopes that will respond by defaulting on its debt strikes me as crazy on multiple levels. For one thing, Mozambique is currently running a deficit, so I doubt that defaulting on their debt would allow the government to increase healthcare spending at all, much less increase it enough to replace the money from USAID. For another, a default will harm all creditors, not just China.
Mozambique has been hiding how much debt it owes. Thus the hidden debt crisis. Anyone who says they know what percentage of Mozambique's debt is owed to China is lying. Further, total healthcare spending includes foreign aid, so that number does not indicate THE GOVERNMENT increased healthcare spending. So that statistic is misleading.
No one wants Mozambique to default on its debts. Including the US. Unless by default you mean the negotiated debt restructuring process Mozambique has asked for.
However, as FLWAB says, if you side with US rivals and cut spending to pay back US rivals (effectively transferring in net American money to China) and lie about how much debt you have and turn down transparency/anti-corruption requests then the result is unsurprising. And, to be honest, your fault.
I don't think the hope is that they'll default on their debt, and I think the idea is that if you prioritize US rivals over the US, it's not surprising if the US stops paying for your medical bills.
What are the long term consequences of developed nations propping up an entire continent which seems wholly incapable of caring for itself? Is it even desirable in utilitarian terms over the long term? What could go wrong?
1. If you have an argument that the long term costs outweigh the long term benefits, then please present it.
2. I could be wrong, but it seems that you are treating "countries " as the unit of concern, rather than human beings.
Re #2, see https://www.britannica.com/topic/human-security
What I am bringing up is that the key to an effective future eventually depends upon the concerned individual or group taking responsibility for themselves. My concern is that we are treating the people of Africa like children or livestock rather than as responsible human adults that need to take control of their own destiny.
IOW I worry we are promoting dysfunctionality and dependency. Of course, I am often wrong.
That is way too general an argument. Why is the specific aid in question an example of the phenomenon you are referring to? We are talking about medicine for sick people; we aren't talking about propping up core government functions.
I was indeed generalizing the argument. Trying to take it up a level of abstraction. Sorry.
Delivering drugs to sick people seems like commendable work to me. My question is whether the larger pattern of sending in solutions to their problems is the long term best solution. It seems like we are addressing the symptoms and by doing so interfering with their ability to become self sufficient.
Delivering drugs to sick people is fine, if they're reasonably cheap and don't lead to antbiotic resistance. Ivermectin is fine, I'll walk on hot coals for that. I won't do the same for HIV drugs -- they're expensive, and this whole "we let orphans live in the middle of the boonies" so that they are impossible to supervise without costing millions of unnecessary dollars... that's just ridiculous.
One orphanage. We have the plans from covid19. Build it there.
>My question is whether the larger pattern of sending in solutions to their problems is the long term best solution.
Yes, that is very clear. But the issue is that you can raise that question about any aid. Lend-Lease, for example: "Delivering weapons to people resisting Nazi aggression seems like commendable work to me. My question is whether the larger pattern of sending in solutions to their problems is the long term best solution. It seems like we are addressing the symptoms and by doing so interfering with their ability to become self sufficient." So, at some point, you need to make a claim that THIS PARTICULAR aid program is counterproductive in that manner.
Depends. We can get a lot out of certain "charity work" (that's called R&D for when we finish depleting our own aquifers). Ivermectin is cheap and safe, it's not a drug that we spend oodles on to make in quantity (likewise antibiotics).
HIV drugs... that's expensive.
All the t-shirts they've got, they are castoffs from America, where people won't buy a t-shirt if they don't like the logo. That's cheap writeoffs.
I'm not sure if Africa is wholly incapable of caring for itself...
Why aren’t they more self sufficient?* If the rest of the world didn’t exist at all, what would be happening there? How would they respond to HIV and countless other diseases and ailments? How would their governments and major institutions work?
By providing aid and relief are we interfering with the natural process of them evolving respectable institutions and social fabric?
Just asking the question.
I guess HIV running uncontrolled would eventually leadto a culture with obligate monogamy emerge and defeat all others.
Africa is not self-sufficient because people never needed to look to the future there. Food is too easy to grow, and they don't have winters. Future-oriented thought is rare there, and not very encouraged by the climate, or by the populace.
If there weren't the rest of the world, they'd simply let the disease kill people and breed all over again.
I'm not sure you HAVE a natural process of them evolving respectable institutions and social fabric. What you have is a tremendous reservoir of biodiversity, precisely because selection pressure is next-to-nonexistent.
Places lacking winters may have dry seasons.
In Israel, the dry season is the "dew" season. It's not like they don't still harvest food during the dry season -- it's just different food. Winter is harsh, and cold, and cruel in a way that a dry season just...isn't. You have to make a bed, early, before the rushes get too stiff, and can't be woven right. You'll die if you don't have enough wood, and a solid enough home (and a bed).
In some "warm" places, they have to put signs up warning people not to sleep on roads. (Yes, these are people of notably low intelligence).
You'd probably consider me to be a meany if I ask you to provide references that back up your thesis, but if you consider yourself to be a rationalist, you should be prepared to defend your positions.
You are faking rationalism here. You are following circular logic where poor outcomes (low IQs/poverty/whatever) of blacks is result of racism and simultaneously same poor outcomes is a proof of racism existence.
You poke holes in opponents' theories yet never prove your own (maybe because it's circular so it doesn't need explanation)
If we experimentally tested IQs of mulltiple diff pops in controlled settings and found zero diff, it would be rational to set zero as a prior and demand proof for any claims about pops we did not study yet, like we have extensively tested thousands of potential medicines, and most don't work, but testing people is unethical, and there is no ground to claim for zero prior.
Any references to evolutionary psychology ought to be thrown out on general principle. If you must have a critique on them, in general:
1) "It was effective" -- once, while we were back in Africa (nutrient consumption have changed)
2) "It's not evolutionary in nature -- merely tied to something that is."
3) It works out enough of the time that it's still in the gene pool. It may be disappearing.
This is why we don't explain "manic depression" as something that is "evolutionarily good" because it's still in the gene pool. There are good evolutionary psychology explanations for it, but you can't disprove the three ideas above, so you're just spitballing and that's not science.
Taking the whole branch of evolutionary psychology with a grain of salt:
https://pmc.ncbi.nlm.nih.gov/articles/PMC4067985/
This is a quick reference to the tremendous genetic diversity of Africa.
I can back up other bits if you want, but I'm not minded to defend terribly hard the evolutionary psychology, as it's a flawed area of study.
That's nice hyperbole disguised as rationality, but the entire continent is not "wholly incapable of caring for itself", some countries are just not capable of caring for all HIV patients. Hopefully, with new medical treatments and economic growth it will be less of a problem in the future.
HIV is just one of countless problems. Setting aside any hyperbole, the question is are we making them better through our aid or are we fostering dependency, free riding and general institutional dysfunction?
What is your take on this question?
Aid is bad when it's tied up with conditions that give the aid-giver some kind of hostile soft power over you, dependency structures are part of this problem. But It's good when it prevents millions from dying in between now and when a vaccine is being developed. In general, Africans would prefer mutual economic development to aid, but HIV is one example where immediate medium-term aid is needed until advanced medical infrastructure that physically cannot exist in Africa yet has time to develop. It would be incredibly hard to argue that PEPFAR saving something like 20 million lives is a worse outcome than simply being free of dependency.
So I can agree that mutual economic growth is preferable to aid, maybe even in general, but some cases are just... good.
I'm generally in favor of PEPFAR based on priors, but I want to push back on your definition of when aid is bad. Specifically, the USA subsidizes overproduction of crops, and then gives that excess away in the developing world, undercutting their markets and reducing their ability to grow their own economies.
That's broadly the position of this report by the Cato Institute (I'm probably a pretty typical neoliberal, but I feel anyone with the Economist logo ought to be at least open to a Cato analysis)
https://www.cato.org/blog/us-food-aid-poor-countries
Do you believe that Omicron was a netgood to the world? If so, would you support more revocation of HIV/AIDS funding? If not, how do you explain the creation of Omicron?
???
Was that question for me? No, any pandemic or epidemic is necessarily a bad thing. A pandemic like SARS-CoV-2 stressed the healthcare systems of well-developed economies, and the healthcare system of countries like India totally collapsed (China implemented a really strict lockdown, so they were able to prevent the collapse of their healthcare system—which is pretty good in their cities). Therefore, pandemics have negative downstream effects on demographics and the economy. Luckily, the world was starting to roll out vaccines as Omicron started ramping up (As of a couple of years ago, there were 23 different vaccines of various effectiveness that had been developed to combat SARS2). Just eyeballing it, it appears that Omicron and its descendants accounted for ~20% of the COVID-19 deaths. It would have been much higher without vaccines.
Yeah, even if we want longer term sustainable change, sometimes immediate aid is the best answer.
I think it's legitimately hard to figure out what kinds of aid will be helpful and what kinds won't be, for long-term development. My not-that-informed impression is that many attempts at making the world better with foreign aid have ended up basically wasting a lot of money, and some have been actively harmful.
Though providing antiretrovirals to HIV+ poor people in countries that have zero hope of affording those drugs themselves seems pretty positive--it's not like we're strangling a nascent pharmaceutical industry in the crib by mistake or something.
I know that a couple of African economists have affirmed their belief in the latter.
Yes. Many of the rationalists on this substack seem to lack any historical context and also seem to be woefully unable to think outside their tightly constrained intellectual boxes. Of course, most of the youngsters on this stack weren't alive during the initial phase of the US AIDS epidemic in the early 1980s. They might think of HIV/AIDS as a treatable disease. But NY City hospitals were overwhelmed with AIDS patients who required months of care and who inevitably died. Initially, the mode of transmission was unknown, and medical staff were accidentally infected with patients' body fluids. It was a real shitshow. ICUs were overwhelmed, and they had trouble finding space for non-AIDS patients who needed intensive critical care. So the surge in AIDS cases affected the non-infected population as well.
Now that we know hella lot more about HIV, we've discovered that immunocompromised people are mutational petri dishes for other pathogens. Luckily, most of those mutations can't overcome healthy immune systems, but occasionally they might. For instance, the Omicron variant of SARS-CoV-2 probably got its start in an immunocompromised person with HIV in central Africa.
Being laissez-faire is all very well and good, until it comes back to bite you on the ass.
Given the rise of Omicron, would you support defunding HIV-maintenance drugs in the 3rd world, as a way to encourage "less lethal mutations"? I think you could argue that Omicron, itself, is worth the cost in "deaths of HIV positive people."
Wow. That's cold. Of course, you assume you'll be one of the elect who survive next pandemic.
But an easier solution would be to remove the patents from HIV drugs—the most effective ones are still under patents because they were developed later, and can cost between US$50k and US$150K/year. South Africa has a perfectly functional pharmaceutical industry. And they manufacture the older HIV drugs that are no longer patent-protected, which they sell to other African countries and India. Of course, removing the patents from these modern drugs would hit the bottom line of investors and threaten the very foundations of our capitalist system of medicine. So f**k those Third Worlders if they can't take a joke. <#snarkasm>
Still the person who believes that it is okay to perform abortions up until 2 years of age, if it comes to that (which we all pray it doesn't). After that? Humanity can die, before it's okay to kill the kids. Call me cold all you like. : - )
Nobody has a perfectly functional pharma industry outside of Israel (India, maybe?) and China (national-security wise). You do make a good point, though.
AIDS, and more tightly speaking, gays committed genocide of the hemophiliac population. I'm not going to say this was intentional, mind, but I'm just bringing it up, in a subthread where folks are trying to shame Americans for inaction. This was not inaction on the part of gays, they were trying to make money by selling their blood.
It took about 4 years after the start of the AIDS epidemic, and about 2 years after the HIV virus was identified as the cause of AIDS, before the Red Cross and other blood banks started screening for it. Granted, it took about a year to develop an ELISA test for HIV, but IIRC there was a lot of resistance to the idea that HIV was the cause of AIDS, and then there was another round of resistance to screening (because of cost). I just checked with ChatGPT, and it says that ~6,000–10,000 hemophiliacs and ~4,000–6,000 transfusion recipients died from HIV contaminated blood products in the US. I'm surprised it was that low.
But remember, it wasn't just gays who caught HIV. I asked ChatGPT and says that ~23% of AIDS cases were women who caught it from heterosexual sex. Also, IV drug users were at higher risk. Right now in sub-Saharan Africa, more than half the cases are women. And the infections happen predominantly through heterosexual sex.
As a historically-challenged person stuck in his intellectual box, my question wasn’t really about HIV. My question was about whether we are on net creating dependency and fostering negative sociology-cultural responses to the myriad challenges facing the billions of people on this continent. Feel free to disagree. I have an open mind.
To put it in historical context, the concept of a "culture of dependence" dates back to at least the 1840s. Regarding the Irish Potato Famine in which millions died, Charles Trevelyan opined in 1846...
“The only way to prevent the [Irish] people from becoming habitually dependent on Government is to bring the relief operations to a close.”
Of course, the Irish were considered irremediably backward, with low intelligence and bad moral character by many in the United Kingdom, much like modern conservatives regard sub-Saharan Africans (without ever visiting there).
Trevelyn again: “The great evil with which we have to contend is not the physical evil of the famine, but the moral evil of the selfish, perverse and turbulent character of the people.”
So has Western Civilization been creating a culture of dependency for the past 180 years? Of course, there were bleeding heart liberals as far back as the 1700s. Jonathan Swift's "A Modest Proposal for Preventing the Children of Poor People From Being a Burden to Their Parents or Country, and for Making Them Beneficial to the Publick" comes immediately to mind.
Indeed, it is the case that an argument shaped like Swami's was wrong at least once. But I'm not sure that proves all that much....
So am I to take it that you are arguing for continued aid, and that, based upon historical precedents in Ireland, you are not worried about dependency?
In theory USAID is supposed to be helping countries get off aid. In practice I'm guessing they often fail at that part, and it sounds like they failed on Mozambique. That said the stakes are high enough that I'd prefer a "keep trying to reform until you get something that works" approach (even though it'll be slow, error-prone and relatively expensive) over a "sharp cuts and millions die" approach.
I was prepared to feel whatever feelings that article might call forth, and I know the only socially acceptable ones will be pity and recrimination against the correct quarter.
But I defy anyone to read it and not have *any* qualms about the sustainability of this … system. And I’m not sure the remote HIV-positive children and their grandmothers aren’t right in feeling that their own countrymen in the form of their “activists” abandoned them.
Is this the future people really want for the world?
I don't see what the activists are supposed to do. They have no income, now they have to spend their time in surviving instead of delivering medicine and caring for the children.
Perhaps a jaded view but the absolute abruptness with which they broke off contact, inevitably suggested to me that they were gatekeepers of the medicine, taking their cut of aid funds, more than dedicated medical personnel.
It would’ve interesting to know what they are doing now to survive.
The article said now they're doing subsistence farming and peddling wares in street markets.
I wouldn't phrase it as "taking their cut of aid funds", it sounds like distributing medicine was a full time job. Nobody would be doing that for free in a developed country either.
In a developed country, like, say, Mexico, they'd be in an orphanage. Or doing productive work, so that their HIV drugs could be paid for.
Actually, these two can be combined, in a developed country. It wouldn't take that much work for Mozambique to "hire apprentices" for skilled labor (that's an orphanage, with one nurse to deliver the HIV drugs).
Their concern is touching, then, and no doubt just such an excess of concern, for the least among them and for the future, is what got them into this pickle in the first place. We may wish to organize along similar lines, for our moral betterment.
This is true. It sounds like most of this could be fixed with "one orphanage" which centralizes the children and makes them close enough that one nurse could care for all of them.
I feel like despite the good intentions of individual actors, this is one of those modern situations that both kind of needs to work, and also must never *really work*.
Wuhan was working on a bunch of Actual Fixes for this particular situation.... Up to and including "murdering them all." I agree that charities are unlikely to fix this "issue" (as HIV treatment doesn't actually fix the disease, and seems to just make it more likely to spread... Take the 12-18 year olds in Mozambique. Sex trade is low-skill and high pay. Voila! More HIV. more need for charity.).
I've heard that modern treatment, if used according to guidelines, can make it all but impossible to spread HIV through sex. Is that not correct? Or was treatment in Africa not actually used according to guidelines?
Does the Longer term sustainable solution involve them building a culture and institutions which fosters monogamy and a revulsion toward prostitution and anal sex? Does aid interfere with this cultural development?
South Africa's still the place where squatters were using a multi-story elevator shaft as a toilet, right? I'm not exactly surprised that a country that throws squatters out of skyscrapers has "health system problems" -- there's a clear disregard for human life in South Africa.
https://mainichi.jp/english/articles/20180729/p2a/00m/0na/011000c
A note that this was seven years ago. In 2021, folks were rioting until they ran out of food (in America, the riots ended when the nikes/legos/etc were gone) -- they literally destroyed the infrastructure necessary to eat.
Not very knowledgeable about Mozambique...
I’m not sure what you are trying to say, but the linked article was very interesting and the picture especially so.
That picture is what America’s public lands are going to look like when Mike Lee and Matt Yglesias get through with them.
That’s the dream, when lunatic left and lunatic libertarian join hands and shout down what the internet so cutely terms “normies” but The Onion typically used to refer to as The Last American Adult.
Can you point to any place where, say, Yglesias opposes all building codes?
I don’t have a subscription to Matt Y’s writing and I can’t imagine getting one just to chase down an irrelevancy like that.
Then too, I don’t live in a beautiful (DC?) neighborhood.
Matt might as well be on another planet.
Your claims about his positions seem entirely baseless to me. If you've got some citation for him advocating a sprawl of barely-intact shacks, I'm willing to change my mind.
Otherwise, it seems to me that you might as well be on another planet from him.
He’s the one with the dream of 1 billion Americans, by which, of course, given birth rates, he doesn’t really mean Americans.
I think the onus is on him/his fellow dreamers of the beautiful dream to demonstrate how on earth that’s supposed to look like America as built prior to the second world war or heck I’ll be generous and give him America up to say 1970.
And not: like the rest of the world, and the worst parts of America which I doubt he ever interacts with.
Talk of Matt Y and building codes has no bearing on anything beyond his own streets. This is Not Even Wrong territory.
I'm glad Alpha School made it to the finalists. I meant to rate it highly after reading it, but I forgot.
That's my prediction for this year's winner.
I'm impatiently waiting for the post to arrive to share with some education-adjacent people I know.
Alpha School was also my favorite of all. I was a bit worried this preference might be idiosyncratic, so I'm also very glad it made it into the finalists.
For those of you interested in Effective Altruism and Neocolonialism, I wrote an overview of the arguments and counterarguments of whether EA is neocolonial: https://bobjacobs.substack.com/p/is-effective-altruism-neocolonial
I can think of steelman arguments that neocolonialism is a coherent concept, but I think you actually have to define it in a way that people in the EA community will take seriously, and see as something clearly negative.
For example, the post links to this definition (not your direct quote, but the same piece):
"Neocolonialism can be described as the subtle propagation of socio-economic and political activity by former colonial rulers aimed at reinforcing capitalism, neo-liberal globalization, and cultural subjugation of their former colonies."
If, like many EAs, you believe that capitalism and neoliberal globalization are very good things for lower income countries, and that Western/Rich country culture tends to be better (e.g. more open and tolerant, better institutions) than the culture in most poor countries, then you'll have to also make the argument that neocolonialism may be a bad thing!
There's a reason I did not use that quote, and did use the Stanford Encyclopedia of Philosophy and the Encyclopedia Britannica.
Fair - I was unfairly connecting your argument with something you don't necessarily agree with.
I think the critique of the quote that you did use: "the actions and effects of certain remnant features and agents of the colonial era in a given society” would be similar, though. These actions and effects of these features and agents could clearly be positive, and you may need to make the case that they aren't.
You also connect Africa's continued poverty to colonialism and neo-colonialism, but I don't think this is justified. While there are clearly some direct negative downstream effects of certain colonial and post-colonial policies, I don't see a clear counterfactual where the rest of Africa was not colonised by western powers, and became more successful and prosperous.
Why would I "need" to? The purpose of the blogpost wasn't to say whether neocolonialism is good or bad, it was to see whether it applies to EA. People can make whatever normative judgement they want from that analysis. If you want a discussion on normativity go read the comments, where you'll find things like:
We do have more academic ways to study the effects of colonialism on Africa. For example, we can compare groups that were ethnically partitioned by the colonizers vs those that weren't. When we do so we find (https://www.nber.org/system/files/working_papers/w17620/w17620.pdf):
> we find that the incidence, severity and duration of violence are higher in the historical homelands of partitioned groups. Third, we shed some light on the mechanisms showing that military interventions from neighboring countries are much more likely in the homelands of split groups. Fourth, our exploration of the status of ethnic groups in the political arena reveals that partitioned ethnicities are systematically discriminated from the national government and are more likely to participate in ethnic civil wars. Fifth, using individual-level data we document that respondents identifying with split groups have lower access to public goods and worse educational outcomes. The uncovered evidence brings in the foreground the detrimental repercussions of ethnic partitioning.
We can do similar studies showing that how colonizers drew the borders had negative effects: https://books.google.be/books?id=KpUqnwEACAAJ&redir_esc=y
We also have historical records of what colonizers did in the colonies, e.g. king leopold's "congo free state" was responsible for the deaths of *millions* of people [~half the population], through brutal forced labor practices, decreased agricultural productivity and the destruction of many local economies: https://www.amazon.com/King-Leopolds-Ghost-Heroism-Colonial/dp/0618001905 (the 10 million cited in this book is disputed by some scholars, but all scholars agree that it's in the millions) And it's not like Belgian meddling in Congo is all in the distant past, as recently as 1961 did they assassinate Patrice Lumumba, the first democratically elected prime minister of the DRC: Neocolonialism is ongoing. I mean, basically all scholars of colonialism agree that it was bad for the colonies, maybe scroll through the wikipedia page if you want to see some further research: https://en.wikipedia.org/wiki/Analysis_of_European_colonialism_and_colonization#Colonial_actions_and_their_impacts
[...]
Before the european colonizers the difference between Hutis and Tutsis was mainly a fluid socioeconomic one, not a rigid ethnic one. Think: Hutu as word for the stereotypical farmer (class) and Tutsi as a word for stereotypical pastoralist (class). But then the Belgians came and not only transformed it into a racial division, they also invented a story that Tutsi's came from the north and were closer to europeans and therefore the superior race. They backed this up with pseudo-scientific measurement (think skull stuff) to give it legitimacy. Then they gave Tutsis exclusive access to education, government jobs, and power, while Hutus were relegated to manual labor. To lock it in place they mandated ethnic ID cards, permanently locking people into Hutu or Tutsi categories based on ancestry, eliminating fluidity (previously, it was more like a class so Hutus could become Tutsis through things like cattle ownership or marriage).
If you think the Belgians couldn't have done more to guarantee an ethnic conflict, you don't know the Belgians: as decolonization neared they abruptly shifted support to the Hutu majority, fearing Tutsi-led independence movements. Then they armed Hutu extremists and allowed anti-Tutsi propaganda. This lead to mass killings, which were reciprocated with more massacres on and on and on, culminating in the Rwandan genocide (where ~75% of the Tutsi population was exterminated).
Again, if we treat this like a courtroom, we can't have a quantitative mathematical proof that Belgium created the massacres (though some quantitative studies come about as close as you possibly can with a historical case without a control group). And we don't have a direct decree for the massacres from Belgium (There is no secret tape of a Belgian saying "execute order 66, or something). But if instead we look at it like researchers it's pretty clear to say that Belgium created it. You know who also believes that? The Belgians, not only the contemporary Belgians (e.g. historian Filip Reyntjens wrote that the ID cards made the genocide "logistically possible") but even the colonizers themselves (e.g. Colonial administrator Pierre Ryckmans admitted: "We created a monster")
Haha, I'm not making this point to defend Leopold's slave colony of the Belgian Congo, or the random border drawing in the Scramble for Africa period. But I'd also not be keen on defending Menelik II; one of the most brutal and genocidal dictators of the 19th century, or the ritual decapitation of hundreds of slaves before the King of Dahomey when the British arrived in modern Nigeria at the turn of the 20th century.
The Rwandan case is messy, and Belgium shouldn't be absolved of any blame, but if your argument is, “Colonialism is bad because 30 years after independence, they organised their own homegrown genocide,” you might want to workshop that a bit...
The way I see it, the question "would Africa be poor today in the absence of colonialism?" is just really difficult. Obviously we don't have success stories like Singapore, Hong Kong or the settler colonies to point to in Africa, and we know that some colonial institutions (particularly in the early days of the Belgian Congo) were really bad and harmful, but we just don't have a good counter-factual.
If you want to make the case that domestic rule would have been worse, you focus on how dysfunctional Ethiopia was throughout the 20th century, how almost all pre-modern African states were particularly brutal, and how many countries seemed to fare significantly worse after the colonial adminstrators had left than they had during the colonial period. If you want to make the case that domestic rule would have been better, you point to you the handful of positive trends before and after colonial rule; places like Asante, Buganda, or Botswana which were consolidating and building institutions before European intervention, and identifying concrete mechanisms of disruption caused by colonial rule.
My guess is that, if European nations had not colonised Africa, we'd have seen indigenous empires with varied levels of warfare, oppression and slavery, but it's impossible to say whether they would have been better or worse.
A good reason that the "Analysis of European colonialism" wiki page generally suggests that colonialism made Africa worse off may be connected to the fact that journal that publish pro-colonialism pieces tend to provoke mass hysteria and serious threats of violence. (https://www.nas.org/academic-questions/35/1/the-case-for-colonialism-a-response-to-my-critics).
As for "neocolonialism" (the original point), I mistakenly thought you were attributing some negative valence to the term, suggesting that it was a significant reason Africa remains poor and dependent. But if your article's point is that neocolonialism isn't necessarily a bad thing, and can describe an asymmetrical, mutually beneficial relationship between countries, then that's reasonable.
> We do have more academic ways to study the effects of colonialism on Africa. For example, we can compare groups that were ethnically partitioned by the colonizers vs those that weren't.
But how are you going to disentangle the causation? Nobody goes to the trouble of partitioning ethnic groups just for shits and giggles, it happens when those ethnic groups are having trouble coexisting.
Borders and partitioning of ethnic groups was sometimes so arbitrary that it's actually quite a good natural experiment.
The linked article claims that "The econometric evidence
suggests that, on average, there are no systematic differences between partitioned and non-split ethnic homelands across observable characteristics that may independently affect conflict", which seems reasonable from my understanding of the process.
> it happens when those ethnic groups are having trouble coexisting.
Euhm, no. Read the second part of the comment.
While I will always appreciate calls for more clarifications, you could also try to understand the arguments against your priors using your own skills and knowledge.
I understand some arguments against my priors here. My call for clarification was more about working out if I actually disagree with Bob's claims.
I usually disagree with neocolonialist critiques because they're insufficiently utilitarian. But if it's defined in a utilitarian way: "x is only neocolonial if it actually causes harm", I may agree with this critique.
And/or engage with what I actually wrote (which was based on renowned encyclopedias, and not one person's viewpoint)
The term "neocolonialism" strikes me as poorly defined.
In my post or in online discourse?
Online and offline discource.
Yeah most philosophical/social science concepts are used much more loosely by laypeople.
By "offline" I was including academics.
Like, all of them? Even the historians, political theorists, etc, that study this for a living?
If self promotion is allowed here, I'd love to plug a recent essay on having a baby and infant suffering.
My new baby brought loads of joy into our little world, but also really unexpected levels of suffering (given that babies are conscious in any relevant way), which made me tentatively update my views slightly away from the pro-natalism that tends to be popular in these circles, and towards the universe being net-negative.
https://torchestogether.substack.com/p/most-of-the-world-is-an-adorably?r=1j27ud
That is a much better and more moving essay than I expected, thank you for writing and sharing it. Sorry I don't have anything substantive to say about it (it's past midnight here), but I did share it, and may revisit it later this week for a more proper reread.
That is a much better and more moving essay than I expected, thank you for writing and sharing it. Sorry I don't have anything substantive to say about it (it's past midnight here), but I did share it, and may revisit it later this week for a more proper reread.
In AI news, the FutureHouse platform is really impressive for scouring biology literature. It's really good making literature summaries and searching out instances of things in the literature, including documents like unpublished Ph.D. thesis. It is at least as good at ChatGPT at finding stuff and is much better at sourcing and not making stuff up. Sometimes it doesn't quite understand my question and answers a slightly different question, but it doesn't lie.
I'm intrigued by the idea that this bunch no one has heard of has a better research tool than ChatGPT. The only thing I can think is that ChatGPT is too much of a generalist. Sometimes its job is to make stuff up, sometimes it's a research tool, sometimes it just vibes with and affirms you. Likewise, it's a bad therapy AI because it's too agreeable, but it wouldn't be that hard to constrain it some so that it stops telling you that you that with the week you've had, you deserve some meth.
On a personal note I am miserable that a lot of my science professor skills are being taken over by AI but I recognize this as a positive for humanity.
re: "this bunch no one has heard of has a better research tool than ChatGPT" maybe it's because of https://sarahconstantin.substack.com/p/ai-deep-research-tools-reviewed? Although I am keen to try out otto-SR from https://www.medrxiv.org/content/10.1101/2025.06.13.25329541v1.full-text if it ever becomes available.
There's a new type of blood! Biology is wild.
https://www.cbsnews.com/news/french-scientists-discover-new-blood-type-guadeloupe-woman/
Important to remember that there are many more blood types than most people realize. 47 total fully established, but probably many more in the “wild” and undiscovered, considering how massive, widespread, and often isolated human populations are.
Most of those don't seem to really matter for transfusions.
Scott -- I represent VC investors, including one fairly high-profile investor/donor that makes grants. Happy to discuss this with you. Shoot me an email to set up a time if you wish. www.hilarymiller.tel
What would you recommend:
- Scott's approach of "something like a grant which is convertible to equity if the grantee becomes a startup", or
- The approach of donating to something like OpenAI (e.g. a nonprofit which controla a for-profit startup)
?
Hi, Jesus it depends on what your objectives are. Some people have purely donative intent in supporting certain kinds of activities. There are advantages and disadvantages to each approach.
I wrote an essay that might interest this crowd: “How to Stare Into the Sun.” (though I admit it’s more lyrical and poetic than some here may prefer)
It’s about chronic avoidance, emotional pain, and the strategies we use (consciously or not) to protect ourselves from feeling. I draw on behavioral psychology, and personal introspection to explore how avoidance loops are formed, and how to break them.
If you’ve ever intellectualised your emotions to the point of dissociation, or used hyper-rationality as a defense against vulnerability, this might resonate. There are references to Faye Webster, fight-or-flight responses, and why fear of love is often more dangerous than heartbreak itself.
Would love to hear thoughts from anyone who reads it. https://velvetnoise.substack.com/p/how-to-stare-into-the-sun-and-dive
About 15 years ago I noticed that I was being avoidant in certain situations. Not major stuff but, like, being nervous asking a Target employee where I could find detergent.. I call this "micro fear."
I bought a baseball pitch counter and carried it around for a week. Every time I caught myself being avoidant, I clicked the pitch counter. It ended up being a few dozen.
After that I decided to try reversing my behavior. I would go to Target and make it a goal to ask where I could find things, even if I already knew where they were.
Within a week I felt my background anxiety falling through the floor. I'm still avoidant sometimes, but only like 10% of what it used to be.
Note: I was never diagnosed with an anxiety disorder, and I don't think I had one. I'm just an introvert who got too comfortable staying in my technology bubble.
That’s quite powerful and beautiful, good for you. Always remarkable what works here
> a word, a homophone, that differs in meaning, but simultaneously captures another important meaning
a pun?
> "How much of it is the creative vision of the company founder and how much is due to the cultural norms of the society in which the startup was born 20 years ago?"
As someone who began going into American sex shop 28 years ago (mostly to giggle with my friends and make the browsing solo dudes intensely uncomfortable, but still!), let me assure you: creatively-shaped sex toys which don't mimic real anatomy are in no way a recent novelty for women *or* men.
I will of course acknowledge that there has been a truly *enormous* amount of innovation in sex toys, but Japanese are far from the only people in that space. Some of the really cutting-edge stuff (remote control a device (most of which aren't anatomically correct) with your phone from anywhere in the world, etc!) by Lovesense and We-Vibe came out of Singapore and Canada, respectively.
@Deiseach gave you a hard time with her comment, but...well...I agree with her that it's a real reach to assign Japanese culture itself so much of the credit for a particular toy.
Tadeusz, you have no idea how much I want to slap you right now, and if you want to put that reaction down to Meiji era anti-obscenity laws, go right ahead.
Woo-hoo, guys like to get their rocks off. So do girls. There are plenty of women-reviewing-sex-toys sites out there, so one by a guy is novel, I'll grant you that. But before you start a paean to "the Japanese are so liberal, you guys, they don't have all our hang-ups around sex", remember that Japan also has some very weird attitudes by Western standards (e.g. the idea of brother-sister incest or romantic attraction, now maybe I can trace that back to foundational myths about Izanagi and Izanami, if we're going to be doing Foucaultian deep dives?) and a culture of "men get to have sex as and when they like and want, women are still stuck on the 'virgin or whore' spectrum, hence why prostitution is legal but prostitutes and the water trade in general are low-status" may be great for the guys but not so much the gals in Japan.
Your startling and novel insights seem to boil down to:
(1) Cultures have different standards around sexuality and sexual expression
(2) A manufacturer of sex toys makes really good stuff that I enjoy
Great, thanks for your service, you have enriched our cultural understanding. But maybe throw in the oul' disclaimer about not receiving payments or other emoluments for writing the review, because right now you sound like a Tupperware party hostess?
Yes, I realise I sound very bitchy here. I think I'm too old and too female to be enthused by young men thinking orgasm is the greatest thing ever and we should all be so liberated that orgasm-seeking of all kinds is the foremost thing on our minds.
EDIT: Okay, cooling down after a minute, I should apologise to you, Tadeusz. It's not sex toys in particular, it's the "Japan is so advanced in sexual matters compared to us" attitude that grigs me.
>(e.g. the idea of brother-sister incest or romantic attraction, now maybe I can trace that back to foundational myths about Izanagi and Izanami, if we're going to be doing Foucaultian deep dives?)
But, but, but - isn't the _key_ question here whether the Japanese royal family should have sued Wagner for IP infringement in putting Sieglinde and Siegmund in his Ring?
Good on you for apologising, but you shouldn't have worded it like that to start with, especially with language that violent, and you did not bother to even adjust the first sentence in your edit. Do better.
Nowhere in his comment does Tadeusz claim that Japan is at all generally "more advanced" than Western countries, I have no idea where you even pulled that from. How much clearer does he have to state that there's a lot of nuance and contradiction, and that he isn't qualified to talk about cultural history and so didn't put it in the review?
If you do not want to read about male masturbation devices, then don't read the review, it's as easy as that. It's not written for you. If you don't want to read musings about Japanese history in a comment section, just skip this one. Sometimes it's just better to shut up and move on.
Tadeusz is my friend, and I admire the personal journey he has taken to heal from religious trauma and now sharing his perspective as someone coming from a repressive Catholic culture and upbringing. I don't have much interest in male masturbation devices myself, being a woman, but like, we can only benefit from a healthy attitude towards self-care by all genders. I don't approve of some random woman with anger management issues swooping in to make him feel bad about stuff he hasn't even said.
"some random woman with anger management issues"
Yes, you got that part right. So you should not be surprised by the following.
"Tadeusz is my friend, and I admire the personal journey he has taken to heal from religious trauma and now sharing his perspective as someone coming from a repressive Catholic culture and upbringing."
Hey, he's a big boy now, he can even buy his own Japanese fake pussies and everything. Does he really need his little friend to hold his hand and give out about bullies?
"You should marry and integrate sex as part of its intended telos with another person in its unitive and procreative senses instead of jerking off into fake plastic devices" - wow, much trauma, such repression!
For whatever it's worth, I'm not really seeing the "Japan is so advanced in sexual matters" message in his post. There's the Polish stuff, I guess, but beyond that he points to his lack of credentials and to how American protestant values affected Japan.
I don't think pointing to these contrasts automatically points to Japanese superiority on any of these matters.
Let him describe how American Protestant values affected Japan, then, if we're going to have a cultural dissection as to why the Japanese make good sex toys.
I could counter-posit "Japanese sex toy manufacturers do great masturbation aids for men because Japanese men have no idea how to get a girlfriend, are not interested in getting one even if they could, and hence all the effort into helping men have a wonderful solo experience. Is it lifelike? Well the guy would need to have sex with a real woman to compare, and since that's not going to happen, then the sensation can be as artificial as we can make it".
You're just being facetious now
>I think I'm too old and too female to be enthused by young men thinking orgasm is the greatest thing ever
Relatedly, the idea that the point of sex is to achieve orgasm seems to be far too prevalent (at least in the West. I can't speak of Japan).
In Japan, the point of sex is to achieve women rolling back their eyes and sticking their tongues out (or other silliness). In a country that values "visual self-control" over one's emotions, making someone else lose control (make unintended noises, make silly visual cues) is significant. (and the person losing control is quite likely to be embarrassed).
See also: “Permitted and Prohibited Desires” by Anne Allison.
So, sure, there’s something an anthropologist could write here.
Thanks for the reference, I'm adding this to my reading list.
Coming a bit late (still haven't read all reviews), but FWIW, I did appreciate the review and I think it speaks well of the sheer breadth of interests and topics that this space can cover. For me, part of the fun of the reviews is reading about a dozen topics I'd have never imagined caring about (some of which, to be sure, I keep having no interest in; though I can't imagine why that would ever lead me to yell at the author in the comments).