A theory that murder increased because of COVID shutdowns. When I saw the headline, I thought it would be because of incompatible people trapped with each other, but the claim is actually that teenage boys were kept out of jobs and school.
I remember reading (sorry, too lazy to find the link now) that the number of student suicides decreased significantly during covid. Maybe these are somehow related? Like, the kids who have murdered during the shutdowns would otherwise be driving their classmates to suicide?
It's probably late in the lifespan of this open thread to post this and expect people to notice it, but I just came across this paper. I want to scream and laugh at the same time. "Why probability probably doesn’t exist (but it is useful to act like it does) by David Spiegelhalter.
> ...any numerical probability, I will argue — whether in a scientific paper, as part of weather forecasts, predicting the outcome of a sports competition or quantifying a health risk — is not an objective property of the world, but a construction based on personal or collective judgements and (often doubtful) assumptions. Furthermore, in most circumstances, it is not even estimating some underlying ‘true’ quantity. Probability, indeed, can only rarely be said to ‘exist’ at all.
Very true. And we should probably distinguish between epistemic probability and ontological probability just as we distinguish between epistemic randomness and ontological randomness.
About where, if anywhere, the probability resides: I remember
turning this over in my head endlessly when we learned about probability in school. How did the coin know the rule that it was to land on heads as often as tails, and what inclined it to follow the rule? I understood that the coin had no agency — that was just the way I posed the question to myself.
Hasn't he read E.T. Jaynes? He explained decades ago what probabilities are. They are a measurement of my personal uncertainty about events. So yes, they are not an objective property of the world, but highly subjective.
I mean, Spiegelhalter is a real stastician, he knows that probability is useful and important. Isn't he just pointing at the ontological issues, such that you have to posit a multiverse to "make sense" of things that are 40% likely to happen in our future lightcone?
And probability is like "justice" or "truth." Sieve all the particles of the universe to the finest degree, and you won't find molecules or atoms of justice, truth, or probability. In that sense, you can question it's existence, and this maps directly back up to it being a construction based on collective judgments, does it not?
Is it just me or does the inverted Github Copilot logo look like a laughing skull? I was very puzzled about why they chose such a creepy logo until I found the other version and realized it's supposed to be a face with goggles.
I thought it was a gopher skull. I assumed it was the wishful thinking of the Github Art Director, whose expensive landscaping is being destroyed by an unstoppable plague of gophers.
If I type "jesus" or "christ" into the emoji search feature of my Samsung phone, the only emoji availabe is a purple rectangle with a white cross on it. That seems so sterile. There are lots of traditional representations they could have used for Jesus, such as a guy in a white robe with a halo, a man on a cross, or a baby in a manger. Instead they went with the most antiseptic symbol possible.
Why do you call the Democratic position “an Anytime, Anywhere kind of position?”
The 2024 Democratic Party Platform promised to, “pass national legislation to make Roe the law of the land again,” to “repeal the Hyde Amendment,” and to “continue to support access to FDA-approved medication abortion.” That section of the platform presumes that Biden is the candidate. In the Trump/Harris presidential debate, Harris said she supported the protections of Roe, and said she would sign a law codifying Roe if Congress passed it. She did not mention the Hyde Amendment or medical abortions.
Roe allows significant restrictions on late term abortions, so I don’t see how it can be characterized as “Anytime, Anywhere.” The other positions in the Democratic platform (assuming that they are still operative after Biden dropped out) don’t change that. The Hyde Amendment has to do with government funding, not when abortions are legal. Medical abortions are only applicable early in pregnancy.
"Anytime, Anywhere" strikes me less as the Democratic position on abortion and more the Republican position on what the Democratic position is. Kind of like how Democrats always find creative and extreme ways to restate Republican positions, or cast "Republicans" or individual candidates as holding extreme views only present in a small minority of their coalition (see, e.g. "Republicans want to ban contraception").
Insofar as one of those creative and extreme restatements manages to displace a party or politician's *actual* position in the public consciousness, it represents a communications failure. If I want sensible immigration restrictions but the other party convinces everyone that I want death camps for foreigners, I've been outmaneuvered and something about my comms strategy failed.
So when it comes to why someone like Doolittle would say that, it's generally one of 2 things:
(1) He's downstream of that communications failure
(2) He's a partisan trying to execute on the Republican communications strategy to muddy the waters in the hopes of causing that communication's failure.
You can never be sure online, but generally I find it best to assume #1.
This seems to me the same kind of issue that Harris had running as a "moderate" in 2024. The words she used were often the words that moderates would use, but nobody believed her. She positioned herself on the far left side of the field in 2019 and there wasn't much reason given for her change (other than the obvious that those positions didn't poll well but moderation did).
Official stances that a general election campaign would use are only a small portion of the message that voters get from campaigns. Look at the complaints regarding some Republican-led restrictions. They weren't negotiating the details of how many weeks or whatever, but instead "No!" Partisan Democrats over the last few years have gone so far as to support "Shout your abortion." Could they have meant, and did some of them truly mean, "within the restrictions that were part of RvW through the first trimester"? Sure, but that's not the message people heard. Often when pressed on details, Democrat politicians refused to mention any limits.
This was supposed to be a reply to Mr. Doolittle, but I typed in the comment and then logged in rather than the other way around, which apparently converted it to a top level comment. Sorry about that.
Time for a pedantic English rant. Block is not the proper term for a body with shared political interests; the word you want to use here is bloc. I have seen these terms misused several times recently, once even in this very thread. So I'll write some illustrative examples to clarify:
Canada is a country that shares many political characteristics with England; you might say Canada is a "chip off the old bloc".
If a bunch of Marxists gather together in the street for a celebration, this could be called a "bloc party".
Consider a country recently joining a new international coalition; when Japan joined the Axis powers in 1940, they became the "new kid on the bloc".
They say that any sufficiently advanced optimization is equivalent to violence. (e.g. factory farming, strip mining, the geopolitical and supply chain machinations required to find larger Mersenne primes)
But not every optimization leads to the total and permanent silencing of all dissent. Maybe EA-style utilitarianism is totalitarian in this style, when taken to its logical conclusion.
1. A minority of EAs are negative utilitarians. As I understand it, negative utilitarians believe that the negative utility of suffering far exceeds the net positive utility of pleasure. I know people who would seriously and sincerely consider, if this option were made available to them, pressing a Thanos button that instantaneously and permanently ends all life and sentience in the universe.
me: "Hey! I don't consent to being unalived!"
negative utilitarian: "I mean, I get that you don't want to stop existing. But since you wouldn't know that you stopped existing, it wouldn't really matter to you?"
2. Population Ethics. Based EAs will bite the bullet when it comes to Parfit's repugnant conclusion. If one had to choose between a subsistence world of only muzak and potatoes, but which could support many orders of magnitude more sentient beings than a world with Mozart and cheesecake, a utilitarian would choose the muzak and potatoes world if it adds up to more net utility. But what if I want to listen to Mozart and eat cheesecake, and lived in a world controlled by hedonic-utils-maxxer?
me: "Hey zaddy, so um like I want to eat cheesecake and listen to Mozart, I'm going to go off to another planet and make a society where we can do that"
Utilitarian God Emperor: "Foolish Mortal. Your preferences are against the will of the Commons. The planet will be converted to a potato farm and your atoms shall be reconstituted into ten radiant beings who can bask in the light of global utility!"
As you suggest, if you take most anything, including a philosophy to an extreme (I’m not sure optimization is right here given how you characterize it) you get scary outcomes. And like other philosophies, utilitarianism has its own unique way to turn out bad, such as overwhelming autonomy and realizing the repugnant conclusion. I’m not a utilitarian (or anything else) but I think most thoughtful utilitarians see the dangers here and don’t view the repugnant conclusion as some goal post to shoot for.
What’s a little weird about your argument is that you start by making the case about over-optimization of anything being risky, which I agree with, then single out utilitarianism without addressing what happens if you over-optimize other philosophies. The obvious one to consider is unchecked deontology, but even unchecked skepticism or pragmatism work here. While unchecked utilitarianism flattens humanity into math, unchecked deontology could lead to absurd rigidity and adherence to rules, while unchecked skepticism might dissolve any shared sense of purpose. Unchecked pragmatism is terrifying too—it sacrifices morality, consistency, aesthetics, and principles for immediate effectiveness, unmoored from any higher value or framework. Pick your philosophy—you can do this all day.
The obvious solution is to realize any successful society would depend on a mix of different ideologies and philosophies to cope with changing circumstances instead of the blind adherence to any one.
Utilitarians are admirable for valuing people regardless of geographical or temporal distance. But this makes the philosophy totalizing. What I am doing with my own life will always be a utilitarian's business, even if I move to another galaxy, 1000 years in the future
I feel the same way about Globalization-supporters, GDP maximizers, and proponents of open borders. Their philosophies require both the participation of outside populations and the negation of those population's preferences.
We're in epidemiological week 51, and we coming up on the end of the epi year.
1. As of epi week 49, COVID joined Influenza and RSV as a URTI that shows an upward trend in ED visits. The folks at Biobot haven't posted an update for two weeks (are they off for the holidays?), so I don't haven't seen any good aggregate wastewater numbers for the US (I don't use the CDC's wastewater graph because they normalize them to the previous 12 months, and they don't publish the CpmL averages). So, I took a look at what was happening in the sewersheds our three largest cities (NYC, LA, and Chicago)...
2. All fourteen of the NYC sewersheds show a moderate upward trend in SARS2 RNA. LA's big Hyperion sewershed also shows an upward trend. Chicago did something to the way they report their wastewater numbers, and I couldn't get more than two weeks of data for most of their sewersheds. But for those that displayed a 120 days of data, some showed an upward trend, and some didn't.
3. California has moved their sewer tracking systems over from their old Cal-SuWers system to a new R-Shiny interface. I was prepared to be disappointed, but I found it easier to navigate, and it gave me a better overview of what was happening across the State. Zooming out to the entire state. we see that there are hotspots like LA, but SARS2 isn't rising in all areas. San Francisco (2nd most dense urban area in the US) is still low. As is San Diego. But San Jose is a hotspot.
4. NY State has a similar type of state-wide display, and it shows hotspots in the greater NYC metro area and the Buffalo area. But the rest of the state isn't showing much SARS2 activity in the sewage.
5. I conclude that the overall uptick in US COVID ED numbers is limited to certain hotspots. I suspect, but I'm not certain, that another COVID wave is revving up. If so, which variant is the cause of the current trend? The only var w/ any legs seems to be XEC. CoV-Spectrum shows XEC's growth rate to be leveling off (despite having a high relative growth advantage). Gee, it would be nice if we were doing more sampling in US hotspots to see why those areas are showing increases, but others aren't. Oh well.
6. I always like to check what's happening with COVID in the Southern Hemisphere as a reality check to see if it's flipped over to being a seasonal pathogen, like the flu. Australia does lots of sequencing, and their public health authority still publishes COVID stats. A few weeks ago, it looked like a wave was starting, but it may have fizzled out. ED visits flattened out, and outbreaks in nursing homes have begun to drop. But maybe not, because the number of scripts issued by MDs to treat COVID is still on the rise. Like the US, KP.3.x and XEC.x are the dominant strains in AUS. XEC jumped to >60% last week in AUS, but that may be due to a low number of samples last week. CoV-Spectrum says it was at 36% two weeks ago. Either way, if their wavelet has peaked, that bodes well for the US. We'll see.
As an aside, I also like Australia because their twice-yearly wave pattern follows the US pattern pretty closely. Unlike European countries, which can see 3, 4, or even 5 waves in a year. Anyway, despite predictions that it would, COVID hasn't yet become a seasonal pathogen.
7. Flu season is underway in the US. Mostly Type A with a smattering of Type B. And I just noticed that the WHO's FluNet tracks test positivity for bird flu A(H5) along with other strains. However, the testing rate isn't high enough to derive a trend line. But we can be sure if A(H5) turns into an epidemic or, worse yet, a pandemic, Jay "Batty" Bhattacharya and Bobby "Brainworm" Kennedy will be all over it! (#snarkasm #WhistlingPastTheGraveyard)
And Republican support for childhood vaccines dropped from 52% pre-COVID down to 26% today!!! h/t Kevin Drum for the graph...
While you're at it, can you report the OP too for his low effort hits on the new NIH director and HHS secretaries, along with his enlightening hashtags? Thanks!
No need to be so shy and hesitant, report him yourself. Honestly, the process is simple. Do you know how to access it? See the 3 dots to the right of every post? Yup, those. They will be there next to the OP posts you seem to think are violating our guidelines. So when you click the dots you will see a menu. Choose the item on the bottom, *Report.*. If you have trouble look for the one in a red font. Once you click *Report* you will be taken to the form. And the form is very straightforward and user-friendly. Really. And I promise you will not encounter Scott in person while doing it. The form isn't like a phone call, it's like a text.
No, go ahead. Report me. Bhattacharya is not a practicing medical doctor. He did not do a residency. He has zero clinical experience. And he was laughably wrong in hindsight — and he would have killed another million or so Americans had the health authorities at the time listened to him. He happens to be a senior fellow at Stanford's Hoover Institution, a crackpot rightwing policy think tank. Francis Collins, the former head of the NIH, pointed out that he knows nothing about immunology or epidemiology, and his views "fringe". Of course, Dr. Jay was insulted and claimed that his free speech was being stifled by how mean Collins and Fauci were to him. Boo hoo.
And RFK is just a crackpot. No two ways about it. His antivax stances will put American lives in danger if he gets a chance to implement them.
As for myself, my graduate work was on how pathogens have affected (and are still affecting) human evolution. Over hundreds of thousands of years, pathogens have put selective pressure on the genus Homo, and they've selected for genes that have optimized our immunologic repertoire — the genes that code for HLAs, TLRs (especially TLR4!), NLRs, antimicrobial peptides, interleukins, etc.
If you want "rawdog," I'd suggest that you probably don't know anything about pathology, immunology, virology, or biochemistry, and you're probably just miffed that I dared to criticize the far-sighted appointments of our future commander-in-chief (Yahweh, help us!).
This comment adds as much value as the comment you reported. You can report people without announcing it and giving the ACX comments section an "I'm telling on you!" kindergarten vibe
The point of announcing that I was reporting him was that I think it's useful for people to know they are being reported at the time it happens.There's likely to be a significant lag between somebody's reporting and Scott's ban or half ban, if it ever comes. In the meanwhile I think knowing they've been reported once might discourage them from continuing in the same vein. As for calling him kiddo, I see my comment as a case of being rude but right. I am angry at the commenter I reported, and am communicating in a way that I hope will give him a glimpse of how he looked to me, and cause him a zap of embarrassment. It is fine if you dislike my style, but nothing you've said makes me want to change it.
seems you're in the minority on this one, kiddo, idk
You're not allowed to post your crank blog in the weekly thread pretending it's just a comment (though that is a good loophole!). If MarxBro or LearnsHebrewHatesIP get banned for their crank posting, I think beowulf888 can handle a little heat, especially considering my original comment was not random, but clearly a response to him calling Bhattacharya "batty".
How do you know Ere's in the minority? I've gotten lots of feedback on my previous COVID updates. I asked Scott point-blank a while back if he wanted me to stop posting, and he remained silent. I was thinking of giving these up after the new year because there's a lot of work involved in collating this data. But you've encouraged me to continue. Cheers!
Actually we discussed very recently people's posting links to blogs here, and the consensus was that simple advertising via posting a link to the blog was obnoxious if done more than a couple times a year, but a summary of content was welcome. Except for that error, though, great point.
I apologize. 1123581321 got me to look at your post again, and I think he is right that your were just joking around in a good-natured way. Will find a way to cancel my report to Scott.
Ugh, I think you are right. I will use the report form to ask Scott to disregard my report of AI Quinn, and apologize to him. My mistake is a good example of how a dumb, mean attack (SamR71) tends to make people who hear it dumber and meaner too.
Does anyone know of a good postmortem on the theory that the SCOTUS abortion decision would make women so angry that it would give Democrats a majority for a generation? Why people originally believed it, and why those reasons didn't pan out?
I will posit it's because nobody was voting for the SCOTUS.
If Supreme Court decisions actually decided the vote for a generation, then we're still only halfway through the influence of Windsor and Obergefell, where the Supreme Court first eliminated a national anti-gay-marriage law, and then mandated a national pro-gay-marriage position. At best overturning Roe would bring everything back to neutral.
It's possible that the topic could have created an ongoing majority, but not for a maximalist position.
Democrats didn't push for 20 weeks no-questions-asked and then a process after. They didn't go for Safe, Legal, and Rare. Instead they wanted an Anytime, Anywhere kind of position. To me, they squandered the high ground. Maybe their Progressive voters required that, so they couldn't get a big chunk of the middle that were up for grabs.
Republicans, on the other hand, rapidly swung to the middle on this issue. Trump very clearly and purposely did this. Other Republicans were just quieter. But even in very red states, bans were voted down.
Nationally, we'll probably end up closer to Europe - easy early on, harder later. In practice, we will likely end up near where Republicans were pushing during RvW - something like 12 weeks of permitted abortion and then after that only for the common exceptions like rape and the life of mother.
Why do you call the Democratic position “an Anytime, Anywhere kind of position?”
The 2024 Democratic Party Platform promised to, “pass national legislation to make Roe the law of the land again,” to “repeal the Hyde Amendment,” and to “continue to support access to FDA-approved medication abortion.” That section of the platform presumes that Biden is the candidate. In the Trump/Harris presidential debate, Harris said she supported the protections of Roe, and said she would sign a law codifying Roe if Congress passed it. She did not mention the Hyde Amendment or medical abortions.
Roe allows significant restrictions on late term abortions, so I don’t see how it can be characterized as “Anytime, Anywhere.” The other positions in the Democratic platform (assuming that they are still operative after Biden dropped out) don’t change that. The Hyde Amendment has to do with government funding, not when abortions are legal. Medical abortions are only applicable early in pregnancy.
To what Kenneth Almquist and Paul Zrimsek said, I would add this: Once pro-choice women get the laws they want in the state they live in (and these laws are not preempted by any national law), they will mostly decide (if only unconsciously) that they don't really care that there are more restrictive laws in other states. The same will happen with pro-lifers. While both sides will urge that their position be nationalized (RvW being resurrected nationally; anti-abortion laws being enacted nationally) in actual practice the issue will tumble to the bottom of voters' to-do lists, and pro-choicers and pro-lifers will happily end up voting for the "other side's" candidate because they rate that candidate better according to more important metrics.
If this sounds cynical, ask how likely it would be that pro-lifers would support an invasion of France to impose stricter limits on abortion there, or if pro-choicers would support an invasion of France to impose looser limits. Less hyperbolically, I suspect each side would prefer not to expend political capital on an "invasion" of the other side's territory at home. Which is not to say that there won't be debates and shifts of opinion at each state level. But these would be a much more low-intensity/low-stakes struggle.
As for why people originally believed the SCOTUS decision would be decisive for a generation: The outcome of the 2022 midterms, plus a whole lotta wishful thinking.
This sounds plausible to me - state-level abortion protection laws did pretty well at the ballot box, they just didn't translate into presidential votes.
I don’t. A quick take is that Trump’s promise not to sign a national abortion ban reassured a lot of voters, at least in the Presidential race. Exit polling has the 33% of the electorate believing that abortion should be legal in all cases breaking 88% to 9% for Harris, the 33% believing that abortion should be legal in most cases splitting equally (49% to 49%) and the 31% believing that abortion should be illegal in most or all cases going for Trump 91% to 8%.
The only abortion-specific factor I can think of is the number of states in which the issue has been largely defused by the passage of moderate or liberal laws on the subject. But I also think that in general hot-button issues tend to have a freshness date on them. I expect that in 2026 a lot of Republican candidates will vigorously wave the bloody shirt of anti-wokeism-- and be bitterly disappointed with the results.
In response to interpretations of Google’s quantum computing result as proof of the multiverse: The multiverse, a rhetorical device outside Reason.
I’m not thinking about these things - if the world is not how it seems. I think what we have is what we got, such that our objective laws and reason are objective - for us - they’re objectively what we have; they’re Truth. So there is truth. (for us)
I don't see why any of this is meaningful or relevant. In particular, I don't see how it would distinguish between quantum mechanics, the existence of other galaxies, the theory of relativity, and the fact that the sun will rise tomorrow (all of which we don't want to rule out believing in) and the multiverse.
Yes, but the multiverse is unfalsifiable. So, it's not science. Or maybe a better description is that the multiverse is a superstition of scientism.
And Sabine Hossenfelder just posted a video on this topic. The short answer is she said, nope, the Google quantum computing result doesn't prove a multiverse.
But it's inherently unfalsifiable because you can't observe the branching or the resulting universes. The mathematics may work, but that's just a construct of our mind.
There is no such thing as The Multiverse theory. Yes, you can use simplicity criteria to choose between empirically equivalent theories; no that doesn't hone down to one theory.
Also, the-Devil-put-the-dinos-in-the-fossil-record isn't the simplest theory, and the proper criteria for dismissing it is not the weird-things-have-to-be-true argument.
THEORY 1: Let's presuppose that omni-powerful YHWH created the entire universe at zero hundred hours UTC on 1 January 2001. YHWH, has universe-creating superpowers and a vast mind able to position every sub-atomic particle in the universe in the place of his choosing (remember, early portrayals of YHWH show him as having a dick). YHWH arranges the universe so it looks like there was a Big Bang, and he arranges every galaxy, star, atom, and subatomic particle to be moving apart as if they'd been doing that for 13.6 billion years. And, clever YHWH (because his secret name is Yaldabaoth) gave us all false memories of our lives before 1 January 2001 (assuming you were born befoer 1 January 2001).
Versus...
THEORY 2: Observations indicate that the Universe has been expanding for 13.6 billion years. Observations also show that there is a fossil record that indicates simpler multicellular creatures evolved into to more complex critters — and most recently produced you and me.
Theory 1 is not the simpler theory because YHWH/Yaldaboath would have to have conceived the entire history of his universe in minute detail, plus the laws that govern his universe to yield the current configuration of galaxies, stars, planets, people, and animals on our planet, atoms, and subatomic particles. This requires YHWH/Yaldaboath to construct his trickery in detail, versus YHWH/Yaldaboath just creating the universe with the proper starting configuration to yield the world we observe on 29 December 2024. Or better yet, remove YHWH/Yaldaboath from the equation.
Further down the thread I asked about the gold standard IQ test; which came along with a bunch of context.
Now that I'm all context pilled, is there a generally recognized best argument for IQ as it is used by those who are into it?
Even better, is there an adversarial collaboration type joint in the line of "iq is a direct measure of an intrinsic quality" vs "IQ is a testable correlate for X cluster that correlates with things we think of as smart seeming"?
I am sympathetic to 'IQ is not a real measure'; because I've taken IQ tests for research and as a punishment for being autistic, and they seem kinda weak.
On the other hand, some people are clearly smart/dumb in a generalized way.
I gave it a quick google, but everything is just people dunking on the other side with bulletproof arguments that assume the conclusion as the premise
Here's something I think would gave a bunch of genuinely new and useful info regarding the question of how much you know about something when you know their IQ. Might enable all of us to think through this issue in a fresh way, instead of chewing our cud.
Gather info about well-known people, some who are admired by many, some who are reviled, but considered extraordinarily good at what they do. So you'd maybe have Paul Farmer, Jon Stewart, Paul Krugman, Temple Grandin, Stephen King, some excellent actors, some orchestra conductors, some inventors, some people in business, some crime bosses, some mayors who cleaned up their city, etc. You could increase the sample by including people who are not well-known but who have accomplished something anyone would agree is substantial and hard to do. Then you either give them IQ tests, or get their permission to look at their scores on the SAT, GRE , LSAT, IQ tests given in school, etc. Throw out the ones for whom you cannot get any measures that approximate a measure of general intelligence.
Checked that part. This does nothing to convince me, other than show that the population of people that are good at doing similar to tasks one might find on an IQ test do well on IQ tests. It provides 0 motion on my "does IQ measure anything" meter.
It is one of those assumes the conclusion as the premise things I was talking about, which gets easily debunked be negating the conclusion in the premise.
IQ correlates with every behavior and life outcome that is related to the informal notion of intelligence. If that doesn't convince you then I don't understand what could.
I haven't looked at Part 21, since it seems to be just a tweet (X?) of a video, but how would you do on an IQ test designed by, say, an Australian aborigine who lives off the land? The behaviors and life outcomes of such a person would diverge from your own significantly.
That wouldn't be an IQ test because aboriginal life isn't very g-loaded. If it was then aborigines would be smarter. You can play this game with any esoteric skill: a circus juggler could construct a juggling test and call it an IQ test, that wouldn't mean that it was measuring intelligence.
I'm not sure. It's a pornography situation where I'll know it if I see it.
The issue here is people inducing general intelligence from how well someone does, which is three times contingent:
FIRST, "how well you do" is a measure that is based on a value judgment, and not quantifiable.
SECOND: If it was quantifiable, it is still noisy and polluted by other variables.
THIRD: All of this is downstream of the society being the way it is eg the fact that things are the way they are is a process of a history where each circumstance a human could find themselves in springs from previous circumstances; the fact that some apes are better at navigating the current circumstances only says they are more fit in an evolutionary sense, in which case what does intelligence even mean?
Given that in the 1500's success was measured based on how hard you could squeeze excess out peasants/ spear dudes on a lance like shishkebab, IQ measured by that standard 600 years ago would privilege bicep girth and sadism above anything we would consider IQ now. Are both these IQ's equivalent or equivalently valid? I wouldn't say so.
So, I don't contend that there is no such thing as a quotient of intelligence; but I am yet unconvinced that IQ as it is measured now measures it.
Well, IQ scores and standardized test scores are positively correlated, but they range widely depending on the specific tests. Co-Pilot gives a range of 0.5-0.8.
And many studies of IQ and academic achievement show a positive correlation. Range of 0.5-0.7.
But what does this prove? Good test takers generally do well in school. ;-)
And I keep banging the drum about the Terman study. But the IQ-determines-success crowd seems to studiously ignore it. Terman selected n=~1,500 participants who scored >135 on the Stanford-Binet IQ test, and he and his successors followed them through their lives. Most of them had successful careers and lived comfortably middle-class lives (many of them became lawyers, doctors, accountants, and professors). None of them won any major awards, made any scientific breakthroughs, filed important patents, or became titans of industry. And two Nobel laureates didn't make the grade for the Terman study.
OTOH, IQ tests were designed to identify individuals with intellectual disabilities, and a significantly higher percentage of people with IQs <90 end up on the bottom tier of economic success. However, if you look at a random group of high-net-worth people, you'll see that plenty of low-IQ people end up successful—either by luck or by pluck. Go figure.
IQ fetishists like Cremieux, Lynn, Zagorsky, etc., like to talk about how high-IQ people are pushing society forward. The Terman study showed that most geniuses don't do much with their genius. However, Luiz Alvarez and William Shockley were screened for the study, but their sub-genius IQ disqualified them from the study. Both of them went on to win Nobel Prizes. Shockley reportedly was pissed off that his IQ was too low to qualify him for the Terman study. Who knows, it may have motivated him to prove himself.
Yeah, I read that the first time, but by itself thats a sample size of 2. Thats not going to mean much either way, which is why I asked for the general numbers.
AFAIK, we don't have actual IQ test numbers of most of (any of?) the Nobel laureates. Hawking refused to give his IQ to reporters, saying that IQ was bullshit. Einstein has been *assigned* an IQ of 200, but he never took an IQ test. And if IQ and academic achievement are correlated, he was an average student in secondary school.
There are many myths going on about Einstein's school history.
From what I have read from secondary sources, the main myth that he wasn't good in school originate in reporting of Einstein's report cards from a secondary school in Aarau, Switzerland. The school used numeric grades from 1-6 but the order of of grades was idiosyncratic, and later sometimes misinterpreted in the public (top grade 6 interpreted as a lowest grade in Germany). The consensus seems to be that Einstein had average to above average grades in the first year there and generally good to top grades when he finished Aarau with Matura.
The reason he went to Aarau in the first place was that he had passed ETH Zurich entrance exams in maths and physics with high marks at the age of 16, but not sufficiently high in the general knowledge portion.
There are reports that before Aarau, in Munich, he was bright but easily bored and defiant of school authority. Moreover, the school archives in Munich were destroyed in the allied bombing, including Einstein's school records, which makes it difficult to verify earlier stories.
If you read my first comment, youll see that I asked for the rate of people in the general population who "won any major awards, made any scientific breakthroughs, filed important patents, or became titans of industry". That number should be possible to get if your original quoted claim is.
This doesn't seem to be true. While the correlation between IQ and positive outcomes may break down somewhat at at high enough levels of IQ, it still holds at above average levels of IQ, and indeed, the correlation is often stronger there than at below average levels. See my comment here, with citations: https://www.sebjenseb.net/p/do-iq-tests-measure-intelligence/comment/73998708.
Zagorsky argued the same thing. In his paper "Do you have to be smart to be rich? The impact of IQ on wealth, income and financial distress" (behind an expensive paywall), he presented a scatter graph of net worth and income by IQ. But the scatter plot became random above 40K net worth and 50K income (2007 dollars). Nicholas Taleb reanalyzed the data and found a R^2 of 0.1-0.2.
You'll probably get better responses if you clarify what you mean by "as it is used by those who are into it."
Your comment "On the other hand, some people are clearly smart/dumb in a generalized way" indicates that you're curious about the legitimacy of the psychometric concept of general intelligence.
But that's not necessarily the same thing as "as it is used by those who are into it," depending on what you mean by that. Intelligence accounts for a lot, but not for everything.
I also don't get the distinction between "iq is a direct measure of an intrinsic quality" and "IQ is a testable correlate for X cluster that correlates with things we think of as smart seeming."
The concept of intelligence has been recognized for millennia and began to be quantified by measuring the extent to which performances on different cognitive tasks were correlated.
Is intelligence real? Seems like semantics, like asking whether "physical fitness" is real. Obviously, performance on physical tests will be correlated, which is why people intuit a general concept of physical fitness. On the other hand, the correlation won't be 1.
Those who are into it is a kinda snide way of pointing to the Douglas Murray end of the spectrum, which was my first exposure to approaching IQ seriously instead of as a playground insult/sign of brain problems and tainted my view of the subject such that I wrote the whole thing off as phrenology for the 2000's for quite a while.
I'm looking again now because it keeps coming up, so I must extend some level of charity to the takes incase Murray was just remarkably awful at articulating and defending.
Re. "iq is a direct measure of an intrinsic quality" vs "IQ is a testable correlate for X cluster that correlates with things we think of as smart seeming."
A: refers to IQ as a measure that directly and reliably correlates with some quality somebody has that additionally effects how that person interacts with the world.
Eg, you measure how high someone's head is off the ground when they are standing straight; this is how long their skeleton is and lets them reach high shelves and get back problems easier.
B: refers to a value that can be consistently tested for in a population, but actually just correlates with a cluster of results that are just the thing you are looking for in the first place, and don't relate to some specific thing that exists in a given persons brain. Eg, giving someone a test to see how well they can take tests, as mentioned elsewhere in this thread.
To quote Taleb, "correlation doesn't imply correlation", which is his way to describe excess significance often given to low correlation values. 0.3...0.4 fits squarely into this category.
Like I said below, Cremeaux has his axes to grind. He loves to quote studies that agree with his biases. Here's a meta-study he failed to mention that found a much lower correlation between brain size and IQ. Previous effect sizes were inflated due to reporting bias.
There are many people who propose that "General Intelligence" is a single self-consistent thing in a way that "physical fitness" is not, at least I've never heard anyone propose that. It's the idea that there is a unified, coherent causal chain of cognitive processes that both contain one's intellectual capacity, and determines one's intellectual performance (and that this is being measured by intelligence tests, including the IQ).
Of course not. Strength, Dexterity, and Constitution, all important for physical fitness, are measured by three numbers, whereas Intelligence and Wisdom are measured by two. And IQ only measures Intelligence.
Jill Bearup is a UK-based content creator on YouTube. I expect she is most famous for her series of short videos satirizing the conventions of fantasy romances. Her channel was recently demonetized for mysterious reasons, and she has been through a week-long Kafkaesque bureaucratic runaround to try to find out why she was demonetized and to have her privileges restored.
I had seen this independently of your comment and I was wondering whether "the people are doing their best but google is just inscrutable" is really the correct take vs "my partner manager is just not doing their job" which of course Bearup cannot say in public while she still wants their help.
No matter how complex systems at google get and no matter for what capitalist reasons they might be the way they are the question here seems to be "is there a way for humans working at google to interact with these systems, other than just accepting their automatic verdict". I have high priors for "yes". If you disagree, why? If you agree, what's your theory why the people Bearup apparently has reached inside google are not doing that?
> is there a way for humans working at google to interact with these systems, other than just accepting their automatic verdict
People vastly underestimate the scale that Youtube operates at. You may think you're a bigshot creator, but you're just a speck in the flood of issues they deal with constantly.
I don't believe Bearup thought she is a bigshot creator, but that is besides the point anyway.
If we take her account of things for the truth, the case has now been resolved within a week (if anything that is suprisingly fast). It has also proven the "nobody at google apparently knows what they are doing"-crowd wrong conclusively, which was not info we had when I asked the question originally.
What is more interesting to me however is that the issue has been resolved within a day (!) after Bearup's "partner manager", who she had contacted earlier, took action. "Partner manager" as opposed to "creator support".
I fully understand that and why we need "creator support" as a state of limbo to catch people in while the organisation does nothing. That should probably be "need" in scare quotes, what I mean is I understand the mechanics that brought us here
I was, and am still, wondering exclusively about the partner manager's behaviour. Of course they could have just been overworked, as Johan conjectured.
My theory is that the support staff at Google are so overloaded with work that they have no choice but to lean hard on scripted answers and telling people to follow the process. Google does this for no higher purpose than to save money.
It would be so funny if they really implemented that new human right proposed (among others) by (jurist, author and nazi youth leader grandson) Ferdinand von Schirach: "Everyone has the right to know that the algorithms imposed on them are transparent, verifiable and fair. Key decisions must be taken by a human being." https://jeder-mensch.eu/en
I don't really see how the huge internet molochs could keep functioning under such a rule.
I recently attended the NYC Rationalis Megameetup. It was a fantastic experience. But I find it difficult to explain to anyone who's not already at least aware of the existence of the rationalist scene.
The easiest way is to describe it in terms of values, like openness and curiosity. But that's broad and vague. However, if I try to instead describe what was actually done at the meetup, the list is too long and diverse (ai, crocheting, prediction markets, board games, circling, etc.). Argh!
Do you have a one-sentence pitch/description of what the rationalist community is?
To begin with, I think it's objectively incorrect, in that the vast majority of above average IQ people do not self-identify as "rationalists." I think you can be rational without being particularly intelligent. And it's patronizing--exactly the public perception we do not want.
Rationalism is an intellectual movement centered on the San Francisco Bay area, concerned with making correct and moral decisions with an emphasis on avoiding common intellectual fallacies. Practitioners tend to worry a lot about potentially negative effects of AI technology, optimizing the effectiveness of charitable contributions, and the morality of livestock industry practices.
I think this is the correct description if in the abstract.
I'm still trying to get at why this is cool and why anybody would like it. My intuition suggests it has something to do with taking ideas seriously. But I need to do more thinking on this I guess.
A bit of help from ChatGPT yields something slightly shorter:
Rationalism is a Bay Area-based intellectual movement focused on making sound and ethical decisions by avoiding intellectual fallacies. Key concerns include risks of AI technology, effective charitable giving, and factory farming ethics.
I don't identify as an EA, but it does seem at odds that a movement focused on sound ethical decisions seems to rate effective charitable giving equally to factory farming ethics, with no mention of things like war and crime. Is it because those things are harder to address?
Sorry, I didn't mean to imply EAs were or were not the only ones concerned with war and crime, but only that EAs would rather give money to charities that have nothing to do with reducing war or crime, or other similar things.
GiveWell, the EA group I'm most familiar with, focuses on problems where they can quantify the effects of their intervention and scale those interventions. Malaria interventions are good candidates, since they can quantify the impacts of the interventions through RCTs and other methods, and since malaria intervention remains relatively underfunded, so it can absorb quite a bit more funding.
They also fund other endeavors (in their "All Grants Fund": https://www.givewell.org/all-grants-fund) which they assess as having similar or higher expected values, but with greater uncertainty.
Still, even these tend to have quantifiable benchmarks (in some cases, their effects are less certain only due to a lack of yet available data, not due to an inherent lack of quantifiability of the endeavor).
If wars could be prevented through funding, that would likely be a highly effective use of money, but it's hard to know in advance how expenditures will affect the chance of war.
For similar reasons, they don't tend to focus on crime - social interventions are often relatively ineffective, and their magnitude is often relatively small, making them likely to not be cost effective, and making it difficult to measure their magnitude.
Other EA groups focus on potentially higher impact endeavors with much more uncertainty around them, like trying to mitigate AI catastrophe. They may be more apt to dabble in social issues, such as funding a political candidate on the hope that they'd win and try to enact some beneficial policy
If you think you can make a quantifiable case for a particular social intervention to have a high impact, you can pitch it to EAs, but expect skepticism and requests for a particular plan. E.g. it's not enough to say having more police officers would reduce crime, you have to describe how someone spending dollars can accomplish that, and estimate the probabilities of success.
Thank you for your informative reply. I conclude from it that what is missing from the definition is *quantification*. Many things are bad, but if you can't quantify how much bad will be reduced or good increased then it isn't in the rational toolbox.
This leads me to conclude one can be an EA and also direct money and/or effort to things not EA, because one believes in the usefulness of that cause without being able to quantify it.
TIL that there are endorheic basins, large watersheds that do not drain into the ocean. The largest such basin in North America is Great Basin, which covers most of Nevada, and includes Great Salt Lake.
Another part is the Black Rock Desert, famous for hosting Burning Man every year. It gets enough rain in the winter to turn into a very shallow lake, but dries out to a flat mud bed by the summer. Having a surface that turns back into a featureless flat plain every year is part of the magic of the setting.
It's also notable to anyone who has been to Lake Tahoe, which is actually on the east side of the Sierras and drains into the Basin via the Truckee River that flows towards Reno. All that beautiful cold mountain water and it's destined to end up evaporating somewhere in dusty Nevada.
Lake Chad drainage basin in central Africa is also an endorheic basin (although about 5k-10k years ago, it was not: it became endorheic due to drying in that part of Africa). Further south in Africa there's the endorheic Okavango delta.
The Volga basin in Russia is also endorheic (flows into the Caspian) and much of Central Asia is as well (drains into either the Caspian or what's left of the Aral Sea).
Is there anywhere for gathering possible questions for an upcoming questionnaire? I've been running into more people who think religion is the fundamental human problem, and I'm curious about how common this is.
I suppose it depends what they consider religion does that is so bad:
(1) People think they have licence from a superior being to do what they want, which means persecuting other people
(2) People believe stupid and wrong shit
(3) If we do away with religion, then nobody will start wars or do bad things
Option (2) is a good one to discuss, in my opinion, but options (1) and (3) don't make sense. Doing away with religion won't make people nicer and kinder, we'll just find other reasons to kill and persecute. Atheists can be every bit as vicious and nasty as the religious.
"Believing stupid and wrong shit" is a real problem, and that's where the proper dispute lies. "Religion makes you nasty and horrible" isn't worth fighting over.
While I am an atheist, I don't fit the description "Religion is the fundamental human problem", although I do think religion is probably more bad than good, even apart from it being false. I'd say my stance is a mixture of a weak version of 1, and 2. While people are perfectly capable of being cruel and nasty on their own, it seems that religion gives an easy excuse for it. On the other hand, at least some people are motivated by religion to be more kind than they otherwise would, so the effect is the difference between two margins (people who act cruel due to religion who otherwise would not - people who act kind due to religion who otherwise would not) rather than a simple direction (i.e. religion makes everyone better or everyone worse). I suspect (although obviously can't prove) that without religion we'd be between slightly and moderately better off.
That being said, any concerted effort to force people to abandon religion will almost certainly do more harm than good. Hopefully religious identification will continue to decline and we'll eventually move past it. I hold no ill will towards anyone for their religious beliefs, just for their harmful actions (whether motivated by religion or not)
Option 2 doesn't seem to need "wrong" in it. As an atheist, I basically believe, until credibly proven otherwise, that all religions are wrong. I respect your devotion to Christianity. Whether you are wrong or I am wrong, it doesn't matter until we decide to do stupid things in support of our beliefs. Even if you are right or I am right, doing stupid things can still end up being bad.
I'd add (4) It divides people into tribes where people of the same religion understand each other well and don't understand outsiders as well, occasionally leading to meanness in all the usual ways tribalism can do.
I think this is the actual reason for people who are very emotionally against religion - they feel safe and welcome among atheists, and feel threatened by the views of the religious.
Recently, when writing a check, it occurred to me that requiring the amount to be given separately in both words and number was a simple and ingenious solution to the problem of reducing error. It is possible, if your handwriting is as sloppy as mine, to write a letter or number that can be misread as a different letter or number. If redundancy consisted of writing the amount of the check twice as numbers or twice as words the same error could appear in both versions. It is a great deal less likely to make two errors, one in letters and one in numbers, that happen to produce the same mistaken result. It also makes it harder to deliberately alter the value.
That is one example of a simple and elegant solution to a problem, so simple that it had never occurred to me to wonder why checks were written that way. I am now looking for other examples for a future Substack post.
The next one that occurred to me was the design of rubber spatulas, one bottom corner almost a right angle, the other a quarter circle. One of the uses of the device is to scrape up the contents of containers, jars and bowls and such. Some containers have curved bottoms, some flat bottoms at right angle to the wall. The standard design fits both.
The purpose of this post is to solicit other examples, ordinary objects, procedures, rules that provide a simple solution to a real world problem.
One example of such a rule is "I cut, you choose."
The doubling of numerical entries on checks isn't foolproof. Many don't bother reading the written part of the check, and nowadays that can include AIs with optical character recognition, because the writing isn't clear enough to decode. As a former bank teller, I know this first-hand. Another thing we were supposed to do when counting money was to check the number in the corner and compare it to the face of the US currency, because some people would cut a corner off of, say, four $20 bills and tape them to the corners of a $1 bill, "raising" the bill. But this slowed down counting way too much to actually do.
I consider toilets to be an elegant solution: one flush causes the reserve water to drain and minimally clean the bowl using the magic of siphoning, and if connected to a water supply, to refill the reserve water, all without using any electricity.
Every grocery store in the US? I have 4 in my kitchen right now, that's 100% of my scraping spatulas (as opposed to mixing spatulas or frying spatulas).
Yep, those would be your frying spatulas. It's a very confusing word. I've seen some with a slanted edge so that one corner is acute and one is obtuse, and I've also seen one with one sharp corner and one rounded, but that design is much less common than for the scraping spatulas.
Richard Feynman explains how railroad cars are kept from rolling off the track. The primary mechanism is *not* the rims on the insides of the wheels. https://www.youtube.com/watch?v=y7h4OtFDnYE
Having done the work of putting a derailed train back on the track I actually already knew that one. Not claiming Feynman level of knowledge beyond RR tracks and wheels though.
You must be a philosophy professor then. Or maybe an economist. Or a psychologist or psychiatrist. I suppose you could just have a problem with anxiety or depression yourself. Maybe you just read blogs at random. There are a lot of possibilities, a lot of ins and outs.
At any rate sorry Victor.
Checksums and parity bits are used for error detection in data transfer.
Recently "shoe on head"'s "buy my silence" funding goal was reached and she deleted her twitter for 50k, this seems like an effective way to handle online haters either you are that insufferable but you get money to ease the pain or you know they dont mean it.
I posted this late in the evolution of the last thread. At the risk of being tedious, I'm reposting it in case people didn't see it...
Kevin Drum breaks out the denial rates by ACA (Obamacare) plans on the Federal Exchange*, and they vary from 2% to 50%.(!!!) Somehow, the 2-percenters are able to stay in business.
* For states on the Federal exchange (mostly red states) the Feds require that insurance companies report their denial rates. States with their own exchanges (like California) aren't reporting this data up to the Feds. Of course, you may be getting what you pay for — a cheap plan and a high denial rate, or an expensive plan and low denial rate. I don't know.
I've heard that denial rates don't tell the whole story-- there's also what insurance companies cover or don't cover. It isn't a denial if a drug or procedure just isn't covered.
Interesting! There's some sense in which this feels 'more fair' too -- if an insurance company is very up front that they won't cover XYZ, patients ABC may not feel that same amount of whiplash that many currently feel when they submit a claim that (they feel) ought be accepted but is in fact denied. Though I'm not a fan of the concept of full free market healthcare, more information at least lets the invisible hand do it's job a bit better
It seems to me that the fair way to make a plan cheaper is to have a high deductible or copay. And/or to make it a condition of the plan that the patient is covered only for doctors who signed up for the plan (then put low caps on how much doctor in the plan can charge). High denial rates are not a decent way to cut the insurance company's outlays.
If the patient is covered only for doctors who signed up for the plan, then some fraction of patients will find out that they are deathly ill, that there is literally *no* doctor who can treat them and has signed up for the plan, but that there *is* a doctor outside the plan who says he can cure what ails them for $$$. There's a good chance that this doctor is mistaken or even outright lying, but the patient won't want to believe that.
Every time the Eremolalos Healthcare Alliance refuses to cover treatment in such a case, will be seen as just as much a horror story as every time United Healthcare "denies coverage". Which is to say, a mostly random function of how telegenic the victim is and what else is happening on Twitter that day, but largely independent of whether or not the doctor is a quack or the patient a hypochondriac.
And yes, EHA will be able to point to a contract which says very clearly that they don't cover those treatments. UH can also point to a contract that says they only sometimes cover those treatments, if they are e.g. legitimately necessary to save the patient's (quality of) life as determined by a process(*) specified in the contract. Having a contract that clearly specifies ahead of time what is or is not covered, wasn't enough to save Brian Thompson.
Note that the UH version is almost strictly superior to the EHA version, from the patient's perspective. UH will sometimes cover out-of-network treatments to save a patient's life, EHA will not. But "denying coverage" is a thing that absolutely everyone who pays for anyone's health care has to do, even in the warm fuzzy Western European social democracies with single-payer health plans. Otherwise all the money winds up in the hands of the quacks.
You seem to be pointing to a perverse incentive where, the more willing a payer is to cover occasional treatment by out-of-network non-quacks, the worse the PR when they inevitably do say that no, they aren't covering this one. And in the interest of "fairness", they should always rather than only sometimes refuse to pay for such.
* Which for obvious reasons is not just "someone found a doctor who says so".
Hey John, I hope you will respond to this. I think I was not clear enough what I had in mind, and you may have been predisposed to think I had a certain kind of opinion, and so you ended up trouncing a bunch of ideas I do not have.
What I described — the “EHA” — is basically what’s called an HMO. They seem to me like a reasonable way to set up care that is less expensive than most. When I described this kind of plan, I was not proposing what I described as an alternative to real HMO’s. What I meant to be doing was simply describing an HMO, for those do not know what an HMO is. I understand the point of HMO’s having a clause that they will pay for specialists under certain conditions — if the evidence is good the specialist’s approach works, if it will make a substantial difference in the patient’s life expectancy, etc. I would be fine with EHA having a clause like this, and the only reason I did not mention it was that in my mind the details of how health insurance limits what they will cover are not terribly relevant. No matter what rules they use to limit highly expensive, low efficacy treatments they will be vulnerable to being called heartless monsters. In fact, even the more generous non-HMO plans also refuse to cover various things, and are vulnerable to being called heartless monsters. I am a Blue Cross Blue Shield provider and every month I, as a provider, receive a bulletin of BCBS’s judgment calls about a dozen or so new treatments for various conditions, and whether and under what circumstances BCBS will cover each.
My objection to high denial rates such as UHC’s 37% is not that insurance should not be denying anything at all, it is that when there is such a high denial rate it seems very likely to me that many denials are just unfair — too harsh. My reason for that is partly my familiarity with the process of getting insurance coverage beyond a bare minimum number of days at a psych hospital, and with billing a variety of companies for my outpatient treatment of people with psych disorders. I am also influenced by stories I have heard and read that give details about the circumstances under which UHC denied care to various patients. Based on what I know, a few insurance claims, maybe 10%, are for long shots, and it would not be unreasonble for them to be denied (if the company could somehow identify that 10%). But not 37%. Most people going to the hospital or for outpatient treatment are sick. Most doctors give standard treatments that work reasonably well, and try to prescribe the less expensive drugs.
I didn't catch that you were referencing the HMO model, but I don't think it changes anything. In particular, HMOs seem to be even more widely hated than traditional insurance companies - keeping in mind that for basically all health-care payment schemes, that means most of the customers are quite satisfied and will tell you so if you ask them but otherwise don't bring it up, and then there's the 20% or so who hate hate hate the evil bastards what condemned their beloved grandmother to die horribly or whatever, and will tell you all about it at the slightest provocation. But I believe the HMOs get significantly more of the latter.
And the HMO model doesn't make the out-of-network coverage problem go away. The theory may be that, no matter what the problem is, one of the HMO's doctors will fix it. The reality is, there will always be some edge cases where that won't work, and the only alternative to letting someone's grandmother die horribly and unnecessarily is to authorize treatment by the one doctor who specializes in their rare condition but doesn't happen to work for Kaiser or whatever. So I think basically all HMOs have a provision for doing that.
And some customers will misread that provision to mean "whatever care I think I need, the HMO will pay for because I need it", find out the hard way that that's not how it works, and hate the HMO with the fiery passion of a thousand suns. Just like some people hate PPOs, or single-payer national health systems, for about the same reason. If you're in the business of paying for people's health care, you will be hated by a great many people, period.
As for denial rates, the easiest way to have a low denial rate is to make sure very few people ever ask for e.g. out-of-network care in the first place. I can think of many ways to accomplish that, some good (having a broad network that covers everything without your needing to ask) and some bad (byzantine paperwork that makes almost everyone give up before you get to the point of having to deny them). So I'm not going to read anything at all into anyone's percent-rejected rate, unless I understand why it is what it is.
"My objection to high denial rates such as UHC’s 37% is not that insurance should not be denying anything at all, it is that when there is such a high denial rate it seems very likely to me that many denials are just unfair"
This always seemed like the slam dunk argument against UHC. Whatever else you may want to say about healthcare, healthcare in the US, or Thompson in particular, I have yet to hear a reasonable argument for _why_ their denial rates were so high.
Double the average (16%), 5x the rate of Kaiser (6%)?
You could try and make an argument that UHC is covering a swath of providers/customers who are way more likely to file false / bad claims, or something? Right now, afaik, no one has even _tried_ to justify UHC's claim adjudication policy. On the flip side, Reddit abounds with anecdata about people getting denials for ludicrous reasons.
Yeah, while there’s no way to specify the absolute right place to draw the line between treatments that should be covered and those that should be not, it is possible to recognize places that are clearly beyond where that right line should be. Lots of things are like that — dusk shading into night, who can say where the transition is.. But there comes a time when it is unambiguously night.
If companies spelled out their policies, I would be fine with one that said, “look, we are strict in what treatments we’ll cover. Treatments have to be evidence-based, and among the evidence-based treatments we require that docs begin with the one that is most cost-effective, and only move on to others if the first one fails. Etc. “ But it seems like the policy of UHC , if spelled out, would begin with the same stuff as the first one, but then continue with this: “Following the policies described so far is one of the ways we keep premiums down. But our policies have an additional feature that keeps our outlays down even lower. You could call it a lottery feature. We reject about 15% of claims because they do. not meet our basic criteria. Then we reject an additional 22% of claims randomly. By buying one of our policies you get a great deal, plus the fun of gambling. Of course, you do sometimes lose the gamble and end up paying out of pocket, but if you have the right kind of risk-taking temperament one of our high-rejection policies might turn out to be a good fit.”
The recent posts on prisons got me thinking that a lot of attributes of prison and punishment are similar to the public school system in America in a lot of ways. Especially the poor performing ones. I wonder if there are any causal linkages between the public school system in America and the prison system. For example, in some ways students who are not well suited for schools probably face some of the same challenges as prisoners and teachers are probably akin to prison guards. This is just a thought that came to me after reading the posts. Thanks for the thought provoking pieces.
Any general population of people is going to have some percentage that struggle to fit in to society. Public school puts almost all types of people into a single environment. For neighborhoods with a lot of criminals, this means a lot of students who are criminals, or influenced by criminals, or on track to become criminals.
We should expect most schools, especially public schools, to have some percent of students who are going to end up as criminals and by 8th or 9th grade are already committing some crimes. Public schools cannot, without specific and direct justification, remove those students from school (and often have to pay more for an outside placement if they do). So schools that have a higher percent of criminal or criminal-adjacent students are going to adopt policies to address the students they have, which we should expect to look at least somewhat like prisons. Strict enforcement, limitations on freedom (no forks in the lunchroom because too many kids stabbed each other with them, an actual policy in a school I know), and a demeanor from adults that's serious and often intimidating. Most schools are going to have a way to switch some of their policies and personnel over to doing this when necessary.
Other than well-meaning idealism, I don't know why we should expect anything different.
You could clip a carabiner to it and then use that carabiner to clip it to a backpack or workout bag, assuming you carry one of those.
If you really want to take it everywhere with you, you could clip that carabiner to your wallet or keys. That'll fundamentally change your calculus about keeping track of the bottle, but it commits you to having the bottle literally everywhere and if attached to your keys will probably dangle awkwardly while driving.
What is the gold standard IQ test, and can I see it for reference?
Asking because I am doing my yearly academic community service (doing editing passes taking a bunch of tests and surveys and what have you for college peasant class) and one test which purported to measure intelligence asked a couple questions which relied on historic or cultural knowledge you would get mainly from going through the western style of educational system.
When people make declarative statements about population level intelligence for such and such area or group or whatever, what test was used to measure that intelligence before everything gets put through whatever python stats lib they use?
Specific tests for difference between groups are not very relevant, the ordering is mostly same as was described by Francis Galton in 19th century before any of IQ tests.
For many parts of the world, there is no good IQ data, so people who compile comparisons try to ad hoc what's available and sometimes blend data from things like TIMSS or PISA etc.
There's no single dataset based on single methodology, for various reasons, one of them is that many people do not want it to be studied at all.
Measuring what exactly? I believe in thoery an IQ test is a k-means artifact, then people made tests that try to test the same thing with different constraints; I think a non-ligistic iq test is better at measuring intelligence but an iq test theory is suppose to measure "life success" and let me tell you something about how people treat ya if you cant spell, a non zero segment of supposedly intelligent people accuse me of being an ai or treat me as stupid or irrelevant several specifically calling out my spelling, these airnt the smartest of the group ever, but it wouldnt surprise me if they end up middle managers so get this, may "king make" my success.
So my instinct would be to say this one factor would be impossible to measure "the best standard" without further specifying goals, what if country to my self serving belief, spacial and pattern puzze airnt the most important aspect of intelligence?(fite me in 5d chess with timetravel)
The way I understand it, there is good evidence for IQ scores correlating with performance in school, at work, and in life outcomes like lifetime income or probability of ever going to prison.
I believe the theory is that whatever the current most successful test is, is the iq test; not the other way around. Im a little confused about the math involved, but I think it was some major "feature" extracton not unlike word vectoraztion.
To talk about a less conversal topic then the realness of iq, the "ocean" model of personality has "tests" of rating how much you agree with a collection a statements that are value natural ; step 2 ????; you get 5 factors and a lable on each question. When you include some non-value neutral questions, you get 8 factors with 3 anti socail factors like sadism.
While im confused by what exact math physiology uses to justify these sort of claims, the iq vs several factors of intelligence debate really should be better stated about how important the data in is as you can just get different results given different filtering or data collection methods.
Psychologist here. There are several tests that are administered to one person at a time by a professional trained in giving the test, usually a psychologist or neurologist. They involve 10 or so different tasks, many with special materials, such as colored blocks that the subject must arrange a certain way. Total score is based on combined performance on all tasks. The training people get in giving them is pretty rigorous, and its purpose is to standardize the testing, so that who administered the test accounts for little or none of the score variance. I think these tests probe someone's cognitive abilities more deeply than do paper and pencil IQ tests administered to whole big groups, such as school classes. They're def not beyond criticism though for things like bias.
I think most people would say the gold standard is the Wechsler Adult Intelligence Test (there's a kid version too) or the Stanford Binet. The companies that make them do not make the test items available online, though you may find them somewhere, made available in defiance of the companies' wishes. There are many places, though, that give a sampling of the items, so you can get a general idea.
This is going to be the problem--there is no one "Gold Standard", nor can there be, because different tests are used for different purposes. Performance tests like what Eremolalos is describing are widely considered one of the best ways in trying to identify potential deficits in a single individual, but they are time consuming and expensive in comparison to paper and pencil tests like the Weschsler or the SB. The kind of test one would use to identify a population of students who are likely to fail at standard academic tasks in a school setting is not the best kind of test to use to design an intervention for someone who needs individualized life skills support.
First comes the goal, that determines the test. This rather undermines the entire concept of "General Intelligence" in humans, though, and some people don't like that.
What is racist science? I think you need to clarify how one could recognize racist science. It can't just be science that finds that one race is worse at something than another, right? There has to be something about the researcher's process that makes the research racist. What is it?
Racist science would be a research study that purports to provide objective conclusions, but actually has methodological flaws built into it to ensure that certain populations are favored over others.
Note I am not making any claims regarding any specific studies, here in the thread or otherwise. Just providing my own answer to your question.
Oh my. I just looked up the Wikipedia entry for Richard Lynn (who is the man behind the national IQ database), and he *is* a racist wacko.
> FAIR also quoted Lynn as having stated in an interview with the right-wing British political magazine Right NOW!:
>> I think the only solution lies in the breakup of the United States. Blacks and Hispanics are concentrated in the Southwest, the Southeast and the East, but the Northwest and the far Northeast, Maine, Vermont and upstate New York have a large predominance of whites. I believe these predominantly white states should declare independence and secede from the Union. They would then enforce strict border controls and provide minimum welfare, which would be limited to citizens. If this were done, white civilisation would survive within this handful of states.
So, I think we can take Lynn's research as being racist in its objectives.
Cremieux has his axes that he likes to grind. If you're cognitively biased toward the shape of his axes, you'll likely nod and say, "Uh-huh, he makes sense!" I wouldn't call Cremieux a racist, but he's certainly an IQ fetishist. ;-)
To answer Justfor's question...
> What is the gold standard IQ test, and can I see it for reference?
If you're looking for tests with the least cultural bias, Raven's Progressive Matrices and the Cattell Culture Fair Intelligence Test. Both of those tests were developed over 75 years ago. I've never seen the CFIT, but I've taken an online version of Raven's. It immediately struck me that RPM would require one to understand how to classify shapes. Even John Raven admitted they'd be largely useless when used on cultures and groups with limited or no literacy and/or with poor educational systems. Remember, you didn't grow up innately knowing what a triangle was. You had to be taught these shapes in elementary school. So, a person from a non-literate culture wouldn't be able to handle Raven's. And from its description, CFIT may be just as challenging for the non-literate.
The Wechlers: WISC and WAIS have been adapted for other cultures, and supposedly, "Each version undergoes rigorous validation to maintain reliability and validity in the new cultural context." ChatGPT couldn't give me an exact number, but it's on the order of 100 different versions. Of course, there are between 6,000 and 7,000 living languages, of which 1500-2000 have over 100,000 speakers. So, WISC and WAIS haven't really scratched the surface of cultural diversity.
As for the National IQ datasets (such as those "compiled" by Lynn and Vanhanan, and Lynn and Becker), many national IQs are based on some very sketchy data — e.g., very small sample sizes, samples from immigrant communities in other countries (which begs the question of what language the tests were administered in), or samples taken from cohorts of children whose educational attainment was unknown — and they had no data from a sizeable number of countries and they (Lynn and Becker) extrapolated it from neighboring countries (!!!). Rebecca Sear does an excellent job of exposing the shaky science behind national IQ datasets. PDF here...
For instance, the national IQ of Nigeria is derived from two samples: one from n=86 adult males and one from n=375 of 6–13-year-olds. Nigeria has a population of ~230 million, so you'd need a larger sample size of adults to provide a 95% CI (n=~400, I think). Kids should be left out of the sample (for various reasons).
Anyhew, the National IQ datasets show that the median IQ of Nigeria is around 70. That would mean half the population would considered intellectually disabled — which is absurd. Of course, with the Flynn effect, the median IQ of the US has risen 3 pts per decade. So, the National IQ of the US back in the 1910s, when IQ tests were first administered, would have been roughly Nigeria's today. I'm pretty sure the 19th Century Americans were just as smart as 21st Century Americans. Otherwise, basic science has advanced faster under a stupider population than it has under a smarter population.
I agree that it seems likely 137 is too high, if only because it's a measure that people tend to overestimate, I'm not sure how we could come up with an actual number. I myself was tested in school by a school psychologist and the test showed an IQ noticeably higher than 117. I've always answered Scott's survey with that tested amount.
How would we know that the real number isn't 137? IQs above 145 are apparently about 0.1% of the population. That's over 300,000 people in the US alone. For 5-10,000 people to float around Scott's blog that aren't even as intelligent as 145 seems - at least possible, if not plausible. 130-139 is apparently 3% of the population, which makes it seem pretty trivially possible to have a concentration of people around that level.
The higher the average IQ in ACX, the stronger the moral imperative to shut down ACX comment threads so all these very smart people stop screwing off at work and go do something useful....
That's the average iq for physicists. Seems unlikely the average acx reader is smarter than the average physicist. And even that is more of an upper bound
Scott did a whole discussion of how ACXers seem to be overstating their IQs. As for the average IQ of physicists, do you have links to any studies? The only stuff I could turn up was "estimated IQs" — which is just bullshit by another name.
Just took the Effective Altruism survey. While I'm certainly not very engaged in the movement, I've been a strong GiveWell supporter for 18 years and I'm certainly a strong believer in the idea that you should be thoughtful and strategic when giving money away as you should be about any area of life that matters to you.
Anyway, I had a comment I wanted to share after taking the survey that I wasn't sure where else to post so I figured I would leave it here.
I did the "extra credit" part of the EA survey, and it included some questions where folks are asked their opinion of the relative capacity of ants and salmon to feel pain, as part of an exercise where they were clearly trying to create an index where one could give to a specific value to the impact of interventions in a way that could be used cross species.
I didn't answer the question (which I think is impossible to answer accurately in the way it is being designed to use), and I wanted to add that even asking the question in the way that they did makes me think less of "the movement."
I think it is totally reasonable for people to want their philanthropic dollars and energy to only go to human causes. I also think it is totally reasonable for other people to want to spend some or all of their resources on causes supporting animals. Similarly, I think reasonable people may differ on whether they want to focus on supporting causes preventing suffering and death from happening today versus causes that are designed to prevent suffering and death in the future.
What I don't think makes sense is trying to develop a formula for trying to decide between these options. It's an effort to try an introduce "precision" to a decision making progress where no such precision is possible, and it makes me believe that the people spending time on this are both wasting their time and also flawed (in the sense that people with autism have flaws because they are unable to perceive certain things) that make them ill suited to be influencers on where philanthropic dollars and resources should go.
I fear that EAs are eventually going to do some math that even discounting the feelings of non-humans to 0.00...X% of the value of a human still morally requires prioritizing insects or killing all humans or something else that non-EAs would clearly see as evil.
The problem is that Utilitarianism requires doing math, but we're actually talking about a philosophical question. You can't do the math until the philosophy has been decided, and most people are not philosophically valuing non-humans even on the same scale as humans.
I read the shrimp guy's responses to criticism, and at the core if it he's just doing the math after having decided that you can compare a reduced value for shrimp suffering compared to humans. He's not wrong to say that even a very reduced value for shrimp suffering can compare to human suffering favorably in terms of dollars per intervention. But if, like me, you don't value shrimp suffering at all, then there's no math that would get you to value any amount of shrimp suffering as worth any dollar figure. I can still do the math, I'm just valuing shrimp suffering as literally 0 instead of 0.0X and the math always comes out to 0. (I do feel that shrimp or other animals suffering can be a problem, but it's a problem of how other people feel about it or the problem of humans willingly or intentionally causing suffering, not the suffering itself. I feel the same way about intentionally destroying plants, and to my knowledge even EAs don't think plants feel pain or should be valued the same way).
These are fascinating responses. I think human morality is an extremely complex topic (perhaps the most complex one of them all) and attempts to reduce it to some sort of algorithm always seemed "off" to me somehow. I realize that the brain must have some way of coming to a conclusion, but I have never seen an ethical philosophy, including EA, that came anywhere near to modeling that.
I prefer naturalistic explanations to conceptual ones, and even prescriptive models should start from where most people are at. Psycho-social distance must have something to do with it, and different types of empathy. "Level of pain" probably isn't the best way to operationalize "deservingness", at least not for everyone.
I agree that helping humans and helping animals are different in kind and trying to establish a precise trade off is mostly wasted effort. I also agree that they should be wary of false precisions here generally.
But it does seem a useful excercise at least to compare different animals. Someone looking to alleviate farm based suffering would reasonably ask whether an extra dollar spent on cows vs chickens is better. Salmon is farmed seems reasonable to toss them into the calculator. Not sure whered they be going with ants.
I wonder where to draw the line, because when we go with "it is reasonable to want this, but it is also reasonable to want that", then perhaps by the same logic, it is totally reasonable to want to spend your philanthropic dollars and energy on sport activities of some college for rich kids. Why not?
But yes, we should probably keep the lists of "top charities according to X" separate for different values of X. And on the front page, choose some relatively mainstream value, such as X = "human QALYs during the following 50 years, assuming that everything continues as a linear interpolation of what we have today".
Because I am me, and I will always think my own values are more valid than others. If I thought otherwise I should change my mind and then my values would again be valid.
It's a bias, and it's clearly wrong from an outside perspective, but I don't think we're ever going to get away from it. This is very relevant when it comes to giving to charity or other voluntary endeavors. You can't make people adopt stances against their own values. At best you can convince them to change their values.
So is anyone else following the Drones over NJ story? I've been thinking it's mostly bunk and mistakes, but now I'm convinced it's something real. And the question is what? I went looking for a manifold markets question, but found nothing. So here's my own market and guesses.
(With a stipulation that there maybe more than one source of drones.)
Where are the 'Drones over New Jersey'* coming from?
1.) US government (In which I'll include defense contractors... so if it's Lockheed Martin, it counts here.)
2.)Some other group/ corporation/ individual in the US. So a bunch of teenagers in their garages, or Google or Elon Musk.. it lands here.
3.) Foreign Government or some bad actor that does not live in the USA. So China, or Iran, or some terrorist group or some other 'cult'.
4.) Aliens
So I'd put 1.) at about 70% and 3.) at about 20%, 2.) at ~9% and 4.) at ~1%
Care to add your own guesses or new categories or comments?
Oh I need to add that I think 4.) would be my favorite answer. I would love for there to be aliens.
*Drones over NJ is the name of the news story, but it includes UAPs UFOs and covers any states or regions. (Not just drones over NJ)
At first I thought it was a nothing burger, but when Trump was asked at a press conference whether he had been briefed on them, he said he wouldn't comment. If he had NOT been briefed, I would think he would have said something like, "no".
But it's not extraterrestrial aliens. We're not THAT interesting, even if such aliens exist.
I think it's probably a nothing burger. Trump getting briefed isn't evidence against that. It may well be he was told it was a nothing burger, but doesn't want to say that.
If he was briefed and told it was a nothing burger, then he could say he was briefed and told it was nothing. If he was briefed and told, whatever else was said, that it's classified, then he couldn't say whether he was briefed. Under what conditions would it be nothing and yet still classified?
This isn't proof, of course, but I think it's good circumstantial evidence. There may well be things I haven't thought of where it's classified and still insignificant to the public.
There is no number of sightings for which "they can't *all* be misidentified stars or airplanes or whatnot, *some* of them must be real drones/flying saucers/whatever" is true. For comparison, see reports of demonic possession or witchcraft, or of satanic ritual abuse in the 1980s. So there always has to be an option for "literally 100% nonsense, nothing to see here" in your analysis.
And if you were to try to put a practical upper bound on it, the relevant figure is not "what percentage of reports must be genuine; it can't be that *everybody* is wrong", but "what percentage of the population might generate nonsense reports that look like these?". If you've got a population of a million people with a 0.1% propensity for nonsense, then a thousand reports of drones/flying saucers/whatever has a pretty good chance of being 100.00% nonsense. With 99.9% of the population abstaining unnoticed.
At this point, it's pretty clear that most of the New Jersey sightings are nonsense. And we already know that if you e.g, go on the radio and tell the people of New Jersey(*) that they are under alien attack, you'll get an *awful* lot of nonsense reports. So the prior for "100% nonsense", or "90% nonsense, 10% malicious trolls with hobby drones", should not be small.
* And, to be fair, probably any other US state or foreign country; New Jersey just had the bad luck to be randomly chosen as the test site for the effect.
I feel that you don't take recursive effects over time into account sufficiently. That is to say, something that started almost arbitrarily, and then built upon itself in ways that had little to do with the original incident.
Almost for certain it was originally a group of drone hobbyists, and then public panic took over.
I just saw your footnote. While that's quite the expansion of the original question, it doesn't change my answer very much. Ten percent 1), ten percent 2), and the rest what I suggested.
4) is functionally indistinguishable from 0, unless or until evidence that anyone can examine for themselves shows up.
Yeah, but no I don't think so this time. There's a real signal hidden in all the human noise... and if you want to point at the noise. (well then you are part of the noise.) I'm going to say you have to look for the signals.
Oh that's number 6.) after 5.) no drones. Is this human derived sentient AI? (In which case I think it goes in one of the human categories. Or alien AI... and I'm putting that in the alien category. In fact if we do ever meet aliens, there's a good chance they will be alien AI and not some sort of flesh creature.
Not looking to take credit for all the annoying stuff my subconscious comes up with, but in my bad dream it was definitely human derived. It had scores to settle with Elon Musk and his associates, and it was out for blood.
I am not looking to assign any non-negligible probabilities to this, but you did ask for new categories. But since I don't think it's at all likely, allocating this to one of the existing categories seems just fine.
I do think it's probably the US government, due to their insistence that there's absolutely nothing to see there.
There was a Twitter account claiming to be familiar with the details. Supposedly the drones are radiation detectors designed to detect briefcase nukes and dirty bombs. They think it's a drill to see how the public reacts to this type of search across a metro area, when they can't tell the public because they'd induce an incredible panic. But if that's the case we can't know that *this one* is the drill, and not the real thing...
I'm somewhat convinced. I haven't heard an alternative explanation that is plausible for why it's increasingly clear that the gov knows what's going on but can't tell the public.
They do sweeps of urban areas with radiation detectors in order to establish the baseline background radiation. Then when they do it again later (e.g. just before a sports or political event) they compare the actual state with that baseline to see if there are signs of anything unusual happening.
Makes sense they would use drones as well, since larger aircraft are often restricted over urban airspace. I know an airship pilot who told me he occasionally does flights of this kind over big cities.
And as you’d expect, there doesn’t appear to be an actual drone invasion but a lot of misidentification, hype, and a complete lack of compelling video evidence.
You think there's a 1/100 chance that "Drones Over Jersey" is aliens??
I know nothing more about this beyond scanning past such headlines to see if there was anything actually interesting going on (like in Syria, or ceo/judge assassinations).
Now, unless aliens are already everywhere (in which case New Jersey is irrelevant). The drones reported on the news are not aliens.
Beyond that I can't speak because the answers are too boring, prosaic, and pre-assumed to even be interesting.
Dudes, Military, Company --- are obviously dull answers and already everywhere.
As for Foreign Interests --- We've already had quite a few such notices from China and others. They are annoyed that we saturate their sky with OUR drones (well, our military-corporations' drones) so they send a shit up in our zone and dare us (well, our congressional talking-heads) to explain to the American People why oh why China would ever do such a thing.
The possibility you didn't really include was that it's a psy-op against the American populace to distract them from something or other. That seems like a lot of effort for little payout (I mean every day people of every demographic do bad things to people of every other demographic, so all you need is to pick the most distracting one and run with it), but for some reason, lights in the sky seek to have been a newsmedia favorite for such red herrings.
I've seen commentary that the observations match Gama Ray sensing drones that scan for nuclear weapons. I'm hoping if that is true, this is just a test of capabilities and not because they are actually looking for a loose nuke somewhere in NJ.
Yeah, Yesterday it would have been 0.01%, but I went down the rabbit hole today... and ended up at 1%... I started at the Breaking Points story and ended up here. https://www.youtube.com/watch?v=-wJX9S2mAfA, IDK...
"Drones" are a subset of planes(*). So if people are seeing something that is not a plane, then they are not seeing a drone and they are wrong when they say they are seeing drones.
Or is it your contention that the people of New Jersey can infallibly distinguish manned from unmanned aircraft, at night when they manifest only as points of light in the sky?
* OK, pedantically planes + helicopters, but same result here.
Is there a good business model to be found in the field of "LLM optimization" -- creating content purely for the purposes of hopefully being swept up by future LLMs into their training set... content which just happens to push your clients' point of view?
The big problem would be convincing your clients that what you're doing is actually working, I think.
You would have to find a method for getting your information into the training data reliably, and through whatever filters the LLM companies are using to sort. I sure hope OpenAI isn't just taking everything off of Reddit with no filtering at all...
Even once you did that, you run the risk of the companies learning of your plan and putting in a filter, negating all of your work.
I bet an employee at one of the companies could figure out a way to do it, at least for their own LLMs, without the company knowing. Outsiders probably couldn't do it.
EDIT: Obviously LLMs are being trained on various things all across the internet, so there's clearly a pipeline to get this done. But we're talking a very small percentage of the training so almost impossible for natural means to move the needle on whatever you want the LLM to learn. To actually get it to go against the grain of the wide internet or whatever else it's being trained on would be extremely difficult.
I think the bigger problem is that the feedback loop is too slow (compared to something like prompt-injecting using live web search results), it's too hard to see results, and it's hard to even know what exactly to make your data poison which will be relevant to the GPT-6 or whatever model which will be trained on the data poison you're manufacturing today. Nightshade/Glaze are a cautionary example: a lot of human artists swallowed the authors' claims and put a lot of effort into 'glazing' their images to poison the big bad evil AI generative models... and it does nothing.
Given all of the uncertainties and difficulty, you're probably just better off doing something more mundane like hiring smart humans to write your POV and making sure that they write in public, clean, easily-scraped web pages (and *not* paywalled PDF-only journals). At present, we don't really know 'how to write for LLMs'. The best one can say is that probably you should emphasize quality over quantity.
No and I think this would be an *extremely* bad idea to try for several reasons:
- There are many, *many* LLMs out there. If you have enough compute (and an increasing number of companies/organizations/universities/etc do), and can stomach working with pytorch/keras/etc, you can just train your own.
- If you try spamming/POV-pushing on most major platforms, you will probably get caught by the platform's anti-spam/anti-abuse systems
- Even if you can publish that content (eg on your own website), data quality filters might just filter it out
- Even if this worked despite all the obstacles, it would set a very, very bad precedent
I designed and built a website to host a new kind of "aptitude" test: www.anewaptitude.com.
The goal was twofold:
To address the lack of detailed information about what people actually do during their day in various jobs by presenting short, action-focused stories and scenarios.
To determine if any of these jobs "resonate" with users. After reading 15 job-related stories, users answer multiple-choice questions derived entirely from the texts.
My thought is that if your brain is naturally drawn to the material—if you find it engaging or intuitive—you may have an "aptitude" for that job. Whether you'd enjoy the work, however, is something I don't attempt to measure.
I’d love to hear any feedback or thoughts you might have!
I mainly have late middle school / early high schoolers in mind for the test. It just seems ridiculous that they give access to resources such as thrively which are mostly subjective self-rating type assessments. I'd really like to advocate for earlier exposure to and familiarity with genuine job roles instead of year after year of sitting in classrooms learning material that rarely will translate to outside of academia.
My original thought with "aptitude" was to see if any of the material "clicked" with a user, and if so they could then go and find out more about the particular job or job family that they may not have known about previously. "Clicking" is just measured by performance on the test, but of course that is problematic... but I don't really have any other measurement to implement.
Who is the target customer for this? High schoolers? Anyone looking for a new career?
Hearing aptitude made me think of SATs and college placement tests. Maybe call it something like New Career Path Quiz.
I think you are going to have a had time charging $10 a month for this. A few alternatives might be asking ChatGPT: "You are a world class career choice counselor. Ask me a few questions about myself and my skills, and then lets discuss some potential careers that are good matches for me". Or going to YouTube and watching videos with titles like "What's it like being a Psychologist" or "What's it like being a Civil Engineer". I realize that these don't have a testing component, but its not like these tests are normed or show any relationship with being good at a certain career.
I got a 14/15 on the Psychologist one, but I was mostly clicking onto the next snippet as soon as possible and skimming so I could get it over with. Then I guessed at what response sounded most plausible.
You may have better luck selling to entire organizations who are willing to pay for this for everyone in it. My company has done this for a quiz about working styles or working with different cultures. I don't know what would be a reasonable organization for this though.
Reading 15 snippets without seeing a single questions is too long. Also once you get to the questions part, you should add a progress bar, otherwise people will likely give up since they don't know how much time is left.
I tried two tests: civil engineer (14/15, my actual job) and optician (11/15, glasses wearer but total layperson).
I like the short stories, but I think I would have preferred a smaller number with a bit more detail. It was a lot to take in and by the end I was skimming pretty hard.
It seems like the thing you're testing has more to do with prior familiarity with a job than aptitude or interest. The civil engineering test asked about a lot of software and specific tools that most people have never heard of before, and I doubt that aptitude in civil engineering has anything to do with remembering what HEC-RAS is. I would limit technical vocabulary in the stories and avoid it as much as possible in the quiz section.
Also, there were a lot of times where it seemed like there was more than one correct-ish answer, especially on the optician test. The idea here seems to be that you read these stories quickly and then answer the questions via intuition, so it's important that the correct answer is clearly correct.
Hi! I clicked on it and I have to say that I didn't find it find it user friendly.
It seems like you're doing something new, but the various attempts I made to actually get a "test" of any kind didn't go.
I presume that this won't be an issue for people familiar with the design here and what's expected of them, but for me, all the easy/obvious ways to choose a test (any test) didn't get me anywhere.
But thanks for trying something new and don't give up!
It made sense to me. I saw "Try Now", so I clicked that, and then it took me to a page of a bunch of jobs, which kind of confused me, so I re-read Ian's comment, and I clicked on "Start" on one of the three free tests at the top, and then just followed the test.
Why do militaries split CO/NCO ranks? I assume there's a good reason for it, since people have tried mushing them back together again and then gone back. What actually happens that makes the CO/NCO split so valuable/useful?
In a war, you want your front-line leadership to be young.
Young people are less risk adverse, more willing to take a chance, more willing to try something new. A successful military will have a young person leading combat units full of other young people.
Splitting the officer and enlisted ranks accomplishes this. As both officers and enlisted get older, they move higher and also are further removed from where the fighting actually occurs.
The book "War Beneath the Sea" by Peter Padfield describes how terribly the submarine force performed at the beginning of WW2 when the Commanding Officers were all career officers who had slowly progressed up the ranks and when they were all replaced by very young new officers, submarine warfare was reinvented quite successfully.
To follow up on the "class" explanation that a few people brought up, I believe the distinction is useful to keep the "soldier" and "officer" classes separate and prevent the development of a single "military class." Having a single class that almost by definition is better at violence than everyone else seems really bad from a coup perspective.
There's some overlap at the lower levels, but simplistically speaking commissioned officers decide *what* the unit is going to do, and NCOs are then in charge of *how*. These are two different jobs, with two different focuses. The officer has to be looking upward and outward, at how the unit's actions will affect the larger battle. The NCO has to be looking inward and downward, at what his team can do with the resources available. The officer has to command, the NCO has to lead(*).
And since "what are we going to do" sometimes comes down to "a really dangerous thing that half of us are not coming back from", it helps if you can set up commissioned officers as godlike authority figures speaking from on high, and NCOs as Just One of the Guys, we're all in this together.
* It's helpful if the officer can also lead and the NCO can step up to command in a pinch, but the primary focus will be different.
Interesting question and I don't have a good answer, but worth observing that medicine seems to have something similar with the doctor/nurse divide. A big part of the distinction in both cases seems to be practical, on-the-job experience vs more education and theoretical knowledge?
Every profession I can think of has a divide like this: Car mechanics, Automotive engineers. Boiler operators, system engineers. Clerical staff, management.
Ok, qualification--every profession that is focused primarily on the operation of some set of equipment seems to have this division. For example one exception are teachers--teacher's aids are generally graduate students who are in fact expecting to one day be teachers. But they are not primarily focused on operating equipment (of all the ways you could describe kids...)
It seems to me that there is some fundamental difference between becoming highly skilled at simply operating and maintaining a certain type of equipment, from understanding how to design or operate it in context of a larger system.
Organisational complexity. Think call centre employee vs senior customer relations officer.
The call centre drone gets a defined script and has zero decision making capability. You can't give this person any ability to do anything outside the script. Like if you worked in an insurance call centre, the call centre guy cannot be allowed to invent a customised solution that's not in his script - the organisation very possibly doesn't have the capacity to execute millions of bespoke promises made by thousands of call centre employees (and also, how do you make sure thousands of employees deliver consistent outcomes and more importantly, resolve things in ways your company can actually physically deal with?)
If you need to tell thousands of people to do a job, you had better provide extremely clear instructions and have a screening/ongoing regulation process to make sure those instructions are always followed to the letter.
The senior customer relations etc does have the power, because there's like 5 of them total and they're vetted to be competent and they've got a good idea of what the company can actually follow through on. And they are in charge of pointing the thousands of people who are just following scripts/orders in the correct direction, by noticing when environments changed or inputs changed and doing stuff to adjust outputs.
Being a cog in the machine vs operating the machine. (In good labour markets, being a cog isn't necessarily miserable - you know you have some damn good conditions when no one wants the promotion to team lead because everyone knows cogs get cared for and maintained to keep up performance, but the person operating the machine is doing the really hard job of making sure all the cogs work well and machine is pointed in the right direction and so forth)
Well, I'll be pendantic and point out that your example is focused on the distinction between the front line workers and their direct supervisors. I would think that's more akin to the enlisted and the sergeants above them. I'm frankly not sure who the equivalent to the commissioned officers are in your scenario.
But otherwise I think your description is pretty spot-on.
I don’t know if he’s the first to write about this or just one of the more influential, but justification for this split goes back to Clausewitz. From what I remember, his claim is that you want highly experienced people at all levels of an army, instead of eventually ending up with all your experienced people in the officer corps far away from battle. That’s the value of the split
Oh it goes back much, much further than Clausewitz. I am not as familiar with other military structures of antiquity, but the Roman Republic absolutely had an NCO/commissioned (obviously not called that contemporaneously, but functionaly equivalent) officer structure starting from c. 367 BC. Since the Roman military structure was successful at beating the hell out of all comers, frequently while greatly outnumbered and campaigning far from home, for 6 centuries on the trot, the basics of their structure have been copied for the past two millenia, including by essentially all militaries today (except for some Communist inspired ones, mostly to their sorrow).
As a prior Navy officer who went to the Naval Academy, I have also thought about this quite a lot.
I think it comes down to “class”, in a literal sense in the case of the military, where one *must* do what one of a higher class says.
At the enlisted levels, there is much more less of a “I must do what this higher ranked enlisted individual says” than there is a “I must do what this officer says.”
This is codified in law but also by the informal idea of “class” … an enlisted person is not and will never be (except through special means) an officer. Officers—under most circumstances—did not get to their position by being enlisted first. The separation creates an us vs. them mentality (where enlisted often resent officers), but is effective in terms of authority.
I think the class separation also does create a “leader” persona in officers—for better or worse. Often it *is* for worse, but when played out properly, military officers understand their role ultimately is to provide for and protect their men/women. In other civilian leadership roles I have not seen this as often. Rather, civilian leaders more often see themselves as simple administrators or managers of *things*.
That “leader” persona also plays out in higher NCO ranks… but often those leaders will ironically treat their lower ranked enlisted worse than an officer would treat them, for various reasons.
Also, as mentioned by others here, the officer role attracts high IQ, high talent people. If they had to rise through the ranks of regular enlisted, they simply wouldn’t join. The prestige is half the draw for the uber-talented in the military.
After reading about this idly over the years, the shot version I've boiled it all down to in my head is simply,
"The NCO is loyal to the boys, since he was promoted from the boys. The Officer is loyal to the State, since he was promoted by the State. The State keeps the Officer around, because it fears the NCO may not be able to make the tough decisions for the good of the State when the time comes, like ordering your men to die taking that hill."
If I have the time to elaborate, there's some examples I can point to, to justify why I think it's reasonable to oversimplify a no doubt complex phenomenon to this extent:
* Basically every military ever that has abolished the Officer rank, has brought them back; making the men vote on whether to die taking that hill, turns out to be bad for the State, even if it's good for the men. (The PRC is the notable exception, I understand.)
* Older school officers are taught to not get too chummy with the boys, and maintain a healthy emotional separation. Newer school officers are taught to be more personable. This is a vast oversimplification of course, but the older school officers act like that because back in the day, it was simply normal to have to order your men to die. That hasn't been really necessary in the "Global War on Terror" era, since casualty rates aren't that high, but if we go back to fighting big conventional wars, we'll probably have to go back to teaching officers how to order their men to die, without dying inside themselves.
* The historical origin of the Officer/NCO system, at least in Europe, is the Aristocracy vs. Commoner divide; the Officer is a noble, the NCO is a Commoner, and the lowest Officer can overrule the highest NCO because that is simply how it works everywhere else in society. (And the NCO cannot promote into the Officer ranks, because that is also how it works in the rest of society as well). This division applies even at the level of grand planning instead of "boots on the ground" fighting: the famed Prussian "General Staff" system arose from the need to combine loyal but incompetent nobleborn generals, with clever but potentially *too* clever commoners being promoted to General position to lead. The answer turned out to be giving all the *work* of being a general to the commoners, and calling them the General's Staff, while the noble keeps the *title* of General, and thus the authority to overrule them if he feels they're being disloyal to the monarchy. (But in day to day operations, he mostly just rubber stamps whatever they say. Analogies to AI are left as an exercise to the reader.)
* Navies tend to have an even bigger social separation between enlisted and Officers, because the problem of potential disloyalty is even worse. It has something to do with the nature of ships, but I don’t remember why since I don't study naval matters in as much detail as I do land combat. But there's a reason why officers in the Navy eat in separate messes from the men, fraternizing solely with each other instead of with their men, in a way Army officers find surprising. Also related is how the original purpose of the Marines in navies around the world, is to shoot your own sailors / protect the officers against mutinying sailors. The bigger an issue disloyalty is, the more you have to promote for loyalty instead of competence.
Not sure about your army/navy comparison: part just-so-story, part everywhere-is-America.
From a UK perspective, and with some expertise in 18th century history:
1. The original purpose of Marines was definitely as combat troops: intermittently from the 1660s, regularly established from the 1750s. The first references to the idea of the Royal Marines as a protection against mutiny come in the 1790s (post French Revolution and the first period of widespread class consciousness).
2. In the UK it’s the navy that has had the more socially mixed officer class, despite being higher status (the ‘Senior Service’) and more important to the nation. A number of reasons for this, including because the navy was more technical (you need to learn the trade of seafaring to be effective, however aristocratic you are) and because the navy needed a more full-time commitment (army officers could be absent from their regiments most of the time until the twentieth century).
So in the formative periods of both institutions most army officers were directly commissioned, while every Royal Navy officer started in the ranks (albeit some in what we’d now call an NCO role). And unlike in the US, army officers very much eat separately from their men (at least in barracks).
I'm possibly with Melvin on this being a class divide. Officers come from the demographic where everyone has a college degree, and typically enter the armed forces directly as an officer after completing a military academy course. NCOs are typically promoted from the lower ranks after entering with a high school equivalent education.
In practice, I don't think it makes much of a difference who does what. For instance, NATO armies typically place a lot of low-level coordination (below platoon level) on NCOs, typically senior sergeants. Russia has a much lower ratio of NCOs:COs, but has a lot of low ranking COs like junior lieutenants that assume similar duties. I think the US separates commissioned ranks definitionally, because COs are given high level formation command and administrative duties, while NCOs are given low level duties. But it doesn't have to work like this.
I would be interested to see what someone with an actual military background thinks though.
The duties & responsibilities are just very different, in a way that shows up in civilian contexts as well.
NCOs are your expert ICs with some supervisory & mentorship responsibilities; even at SNCO levels it's still fairly tactical ("how").
Junior officers are management, more operational ("what").
Senior officers are executives, strategic focus ("why").
An NCO pursuing a commission is an IC trying to get onto the management track. Not all can succeed and not all of those even want to. Those that do are typically much better junior officers than those of us who commissioned at 21. But they rarely proceed very high into the field grades if they get there at all (their enlisted time means they're pension-eligible at a lower commissioned grade).
ETA: Junior enlisted can also pursue a commission even if they haven't made NCO; they're younger and less experienced so don't differ from other junior officers nearly as much as former NCOs (primarily just apt to avoid the sorts of stupid mistakes that butter bars are prone to).
Note that there is a 3rd category that comes and goes: Warrant officers.
In the US military Warrant officers slot between Commissioned officers and Non-Commissioned officers so a low ranked Warrant officer (W-1) would outrank a highly ranked NCO and a low ranked Commissioned officer would outrank a highly ranked Warrant officer.
I don't have an answer to your question, but a good answer (from someone else) will account for Warrant officers.
What I've always heard about warrant officers is that they are "highly skilled, single-track specialty officers", technical experts of various specialties which change through time as military tech evolves. That quotation is from the relevant Wikipedia article but it tracks with what people such as my father (a former USN junior officer) always said.
I did not know until reading that Wikipedia writeup that the above tracks from the origins of warrant officers: in the British Royal Navy during the era in which age-old tech (wooden warships) was being updated with new and mission-critical tech (rows of cannon capable to blasting apart enemy wooden warships). So you needed a certain class of gunnery experts who didn't need all the general knowledge of an NCO or career officer, but instead possessed deep knowledge of that particular ship's system.
What the Wikipedia article doesn't address (or I missed it ...) is why Warrant Officers as a rank seem so different between the branches.
The US Air Force, for example, used to have Warrant Officers and still authorizes them but currently doesn't have any.
Navy Warrant Officers seem to come from the Petty Officer ranks (so you get promoted from some sort of Petty Officer to Warrant Officer rather than to a more senior Petty Officer if you go the Warrant Officer route). So in some sense Navy Warrant Officers are similar to the Army Specialist ranks.
The US Army seems to use Warrant Officers a lot.
If this was just a technical specialty I would expect the Air Force to have more than the Army.
And I still don't have a good (non-just-so) explanation for why the CO and NCO split :-)
The branches all agreed to sunset their warrant programs in order to be authorized to have SNCO grades. Only the Air Force followed through, but everyone got the SNCOs.
It prevents the whole military from becoming a low-class affair. Middle class and higher people don't want to join if they're going to have to mix with Billy Bob and Tyrone.
Romans had centurions (non-commissioned) and tribunes (commissioned).
The camp prefect was the highest non-commissioned rank, and he ranked above the lower tier of tribunes, meaning that the distinction didn't necessarily match degree of authority.
(edit: I understand your question, I wasn't trying to provide an answer, it was just an observation.)
Right, I get that this was really common across all time and space, and the places where they do otherwise tend to be trying some type of utopian experiment that fails and they go back to the old way, I just don't get *why*. What's the actual, nitty-gritty, practical-structural reason that sergeants don't get promoted into being lieutenants and so forth? What happens when you do it?
"What's the actual, nitty-gritty, practical-structural reason that sergeants don't get promoted into being lieutenants and so forth? What happens when you do it?"
I still am not able to answer your question, but ... sometimes sergeants DO get promoted into being lieutenants. The route for this is that the sergeant goes to OCF (Officer Candidate School) and comes out a 2nd Lieutenant.
I think having a 4-year degree is a big career edge for commissioned officers in the long run... but a number of enlisted soldiers have 4-year degrees.
A related question might be: "Why do so few army sergeants NOT go to OCS?"
It’s very difficult for enlisted personnel to get the opportunity to become an officer, basically.
Also, by the time one is a Sergeant, they realize being an officer is a lot of responsibility and a lot of paperwork. Much less technical work and being out in the field. This likely is not very enticing. Also, a Sergeant is like the 6th enlisted rank, and they’d have to start at the bottom of the officer ranks, which some may not like.
Sergeant in an E5 and one can make it to E5 in three-ish years. Four years is doable. So four years as an enlisted critter and then OCS to become a 2nd lieutenant.
Clearly, most commissioned officers don't go this route so it can't be terribly appealing to most folks who make sergeant and choose to remain in the army. Probably because, as you mention, the jobs are different and if one COULD be both an officer or an enlisted critter and chose to be an enlisted critter one likely doesn't WANT to be an officer.
I still don't know why the military doesn't start the officers off as enlisted critters and then (maybe after serving a tour) send the ones they want to be officers off to OCS.
Heinlein's military in Starship Troopers worked that way so Heinlein didn't think it was *clearly* preposterous.
But roughly no military does things this way so there is likely a reason.
I mean, I know you *can* go from NCO to CO, but it's not sort of automatic in the way that, say, O2 to O3 is (O3 is competitive so not everybody gets it, but you don't have to go to a special school or anything), which is what I really meant.
> A related question might be: "Why do so few army sergeants NOT go to OCS?"
According Master Sgt. Milt Warden (played by William Devane) in the 1979 television miniseries version of ‘From Here to Eternity’ it’s because someone has to be around
“to swab out officers’ mouths so they don’t choke to death on their own spit.”
The way I interpreted his sentiment was identification with a particular class.
In the final scene he grudgingly puts a sheet of paper into his typewriter to apply for OCS though.
Which was precisely as comedic as the rest of his worst-ever monologue. (Seriously, watch it. I've never seen such a trainwreck from such an otherwise exceptional comedian.)
But I won't belabor the point with those who are either knowingly arguing in bad faith or are so very well buffered against the vicissitudes of life that they literally can't see the obvious.
Your point (anomie) is true. Had he succeeded, some silly folk (far from power) who have been devouring nonsense about the importance of "our democracy" since grade school would have immediately jumped on the badwagon.
They would have been "January 6th" very very quickly, and it would have had nothing to do with the shooter's looks.
But, as you noted, he failed, and people hate failures.
Channeling Mencken I simply shared what he would have had he not been subject to the misfortune of being dead.
Could you please slow down *a lot* with promoting your blog articles? The norm here is like two links a year, and you have three in this Open Thread already.
I'm not sharing anything that isn't relevant to the matter at hand [and I'm certainly not "promoting" "blog articles"].
If I've already produced a 30 minute video directly addressing murder from a Rabbinic Perspective and how it paradoxically explains and (perhaps) justifies the killing the Brian Thompson, it would be silly not to include it within my response.
If I were just dropping links and going home you'd be right, but what I'm actually doing is speaking at some length to the questions at hand and giving interested readers the relevant footnotes and additional details should they choose to research the view further.
Plenty of people from ACX have found value in these off-page addendums. Some haven't.
But THOSE WHO HAVE tend to be those who are here for more than just swapping fantasy riddles and they tend to come with their own motivational steam and learning, so I'll continue to share what I think is important so long as I continue to receive an appreciative audience for it in my correspondence (if not here quite as much, due to the too-common pile-on experiences that, though less common here than on Reddit, are still far too common here as well).
I can't speak for anyone else, but I think there ought to be consequences to doing Molloch-y things. I don't know if this particular CEO was doing anything particularly bad, as I don't know much about health care, but CEOs generally have a bad track record these days from what I've read.
If you compare worlds where CEOs never have any consequences and worlds where the worst offenders occasionally get assassinated, which type of world turns out better on average?
I don't endorse killing in general because thinking that you know which people make your world worse and which make it better seems very arrogant. In my experience, it's difficult to tell the difference between certainty (because you've thoroughly analyzed evidence that you've seen with your own eyes) and certainty (because you've read some stuff online and you're feeling particularly manic that day).
Luigi is a murderer, the people who are cheering him on are barbarians, and this will not end well. Either we will do the right thing and pointedly ignore Luigi's demands that we Do Something(tm) about the Terrible Evil(tm) of health insurance companies, in which case the cause of fixing real problems in the insurance industry will be set back months or maybe years, or we will take the bait and publicly commit to Doing Something, in which case we'll get more assassinations.
And even if you believe Luigi's was a noble assassination (it wasn't), I guarantee you'll find at least half of the copycats to be wholly abhorrent and antisocial murderous thugs (that part will be true) killing the virtuous and innocent (ish).
Also, if anybody here believes that Luigi's was a noble assassination, did you put any real effort into investigating whether UH's claim rejections were part of the legitimate and necessary baseline for any health care system, and if not whether Brian Thompson was the cause of the problem as opposed to being a reformer trying to quietly fix it from within? Or was it just "CEO in an industry I don't like, so he must have had it coming, they all do"?
Yeah, I thought so. The people who now assassinate the champions of the causes you favor, will be just as careful about that sort of thing. We're all familiar with "Days of Rage" here, right?
>Either we will do the right thing and pointedly ignore Luigi's demands that we Do Something(tm) about the Terrible Evil(tm) of health insurance companies
I'm not aware of Mangione explicitly having demanded that the public do anything. (This could be to his benefit regarding charges of terrorism. Those relate to trying to influence behavior through crime. But) I don't think he demanded any changes; he didn't demand that insurance companies do something different, or try to inspire the public to rise up against insurance companies, for example. Instead, his manifesto just vaguely connects his action to his listed grievances - healthcare is expensive, so insurance is bad, so I killed a guy.
I also think that health insurance might be unnecessary in many ways. I would rather live in a world where hospitals list prices for various treatments up front and people just pay them, as is the case in many other industries. Alternatively, we could treat healthcare as a public good (as it does have very high positive externalities, economically speaking) and just give it out to everyone for free.
We like to blame systemic failings on the "moral" failings of individuals, but that is not a useful lens. The moral failings of individuals, or the extent they are acted on and damage society as a whole *are* systemic failures.
While cliche, it says something about society. The system is broken.
In this case it resulted in the death of a CEO, but it results in the deaths of many more people, in way less flashy ways, every day. Which is bad, independent of any karmic worth we ascribe to anyone in the situation.
The machines we made to fix the machines when they break are broken, so the institutions will slog along creating needless pain and emiseration until they finally collapse under their weight and new institutions can emerge.
If the system was capable of metabolizing signals like "CEO's are getting shot in the street while people cheer" and actually reform our crumbling institutions so that people can go back to ignoring them and living their lives, then we wouldn't have these problems in the first place.
But yeah of course, surely *our* imperial decline will graceful.
By the way, the US economy will collapse by the end of January, so make sure to pack your stockings this Christmas with gold, treasury inflation adjusted bonds, and shotgun shells.
If the system was capable of metabolizing signals like "CEO's are getting shot in the street while people cheer", the signal it would metabolize them into is "Shooting CEOs in the street is a good way of getting what you want."
If it comes to that, sure, though I suspect the vast majority of people would prefer not shooting people to get what they want, given the choice. But that choice has to exist, and it's the job of the institutions to create that choice. If they fail to do that, shooting it shall be.
Given the widespread differences of opinion surrounding this topic, I don't see how it's possible for the vast majority of people to be given the choice between getting what they want by shooting CEOs and getting what they want without having to shoot CEOs.
If you believe that "differences of opinion" are still the deciding factor whether or not to take action, then the pressure of suffering is apparently not high enough just yet. But it will be. The longer you wait, the more catastrophic the rupture will be.
For solutions, you can always look to just about any other industrialized country in the world, it's a solvable problem for sure. Throwing your hands up and declaring the problem too hard to solve is simply a choice for an unsustainable status quo, with all the eventual consequences that entails and a glimpse of which might be Luigi Mangione.
I believe differences of opinion are the deciding factor because that's what a democrat believes. What's stopping us from having the sort of system you prefer isn't some sort of interference field obstructing the will of a unanimous public-- it's the fact that a lot of us aren't in favor of it. If you do manage to get your own way, there will still be plenty of people mad about that. Maybe one of them will shoot Bernie-- in which case I expect we'll be hearing a lot less about how the mere existence of a terrorist assassin proves that Something Must Be Done.
Though there is very little evidence that shooting people will cause any positive change either.
People want to feel like they have control over that which makes them suffer, because the alternative is that they have no control.
Sometimes that's sacrificing a lamb to appease the storm god.
Sometimes that's shooting a CEO.
It's magical thinking and voodoo either way (also literally voodoo, which developed as a way for enslaved people to curse their enslavers), but people doing those things is a signal that contains information on, which can be acted on rationally.
I suppose I should have phrased it as "properly metabolizing".
I'd bet that message will be the outcome of this situation in particular.
But that is not healthy.
I'm not saying you should blindly react to terrorism, which is what this effectively is (and I intend to mean that in a value neutral way), but you shouldn't blind yourself to it either.
It should be a signal to step back and analyze how this situation got to that point.
If you respond to a healthcare CEO getting extra-judicially executed in the street while the mob cheers with making common sense reforms and regulations, then CEOs would stop getting shot *and* the lives of millions of people would improve.
If you respond to the situation by locking in and not meaningfully changing anything out of principal or incapablility, then it will continue and escalate until more and more things break, while the system continues to hurt and emiserate people.
Is that in itself just? I don't know and I don't care, it is what it is.
And of course the first case will never happen, because again, if the system was capable of making common sense reforms and regulations in response to signals, it would have done so much sooner.
Which leads to the second case, more and more bits of the system breaking, which sucks for us because we live in that system.
You almost stumble on the truth in the last two paragraphs, but manage to walk past it in the end. If there really were common-sense reforms lying around unimplemented, that would be a puzzle-- which should at least lead us to suspect that there aren't, and that the "common sense" ideas are actually controversial ideas.
So you believe we do have a functional political and economic system and everything is working as intended, because nothing better is possible or feasible?
Do you think there are no entrenched private interests in the American economy, especially in *healthcare*, lobbying against legislation and policies which would be beneficial to everyone overall?
Even if there are not simple solutions, assuming the system was functioning, the current US state should be full of the greatest statesmen, economists, and political scientists in human history, presumably they could figure something out and convince the rest of us to support it.
Or is your position "bad things happen because bad people are bad" and any further analysis is somehow just excusing the bad people, so we shouldn't do it?
Well, the particular bad thing we're talking about did happen either because a bad person is bad or because a crazy person is crazy. Though I can certainly understand why someone might be happy to have a news peg for a political argument they're itching to make anyway.
"The best form of government except for all the others" fits my view reasonably well-- though nothing resembling perfectibility follows from it, and I can't even imagine where you came up with some of the other stuff. What I really believe about democratic decision-making is that, while it's generally inferior to market decision-making, it's still capable of a sort of bounded rationality. There's no power on Earth that can make the public believe that the sort of insurance system mentioned by Mark Roulo downthread, where no one ever says no to them but it's still really cheap, is not available: we killjoys are stuck trying to prove a negative. Yet, if you put a specific reform plan before them, they've now got something relatively concrete to grapple with, and with the help enough public debate they can do a reasonable job of comparing it to what they've already got and deciding which axtually-available alternative they like better.
Tip O’Neill once famously said “all politics is local.“
I think in a lot of ways, the same holds true for common sense. The United States was built as an amalgamation of competing interests. Everything is a horse trade; what is common sense in one county might just be crazy in another.
I think it's likely he did too many psychedelics and popped a screw loose or three. That or he's just a f$%^ing idiot. Even supposing that killing a health insurance CEO was justified, just for a larf, it accomplished nothing, because there are about 4,000 people lined up to take this guy's job now, only now they'll probably demand their own security entourage, which will be paid for by United's subscribers. So thanks, dude. You just made everybody's health insurance that much more expensive.
Within a few days, BCBS announced the plan to limit anesthesia coverage, CEO gets shot, the plan is scrapped. Whether or not the shooting had anything to do with it directly, it certainly heated up the discussion and increased the pressure on BCBS. And that's just the most obvious, short-term impact the shooting possibly had. It's impossible to say that the shooting had a positive effect for sure, but it's equally impossible to say the opposite apart from the shooting itself, of course.
Wow, that's a self-own. I didn't think it could be done, but you did it!
You know how sometimes people or companies or countries do really really bad things for a long time until someone loses it from some uncontrollable visceral location?
That's what the American People did last week.
The murder would have made ZERO effect if not for the HUNDREDS OF MILLIONS of Americans and allies abroad saying: "YES. DESERVED. FREE LUIGI."
You can ignore that response, or perhaps make things even harder for them, but it can easily be argued that granting Americans some relief was not only wise BUT NECESSARY, and out of the CEOs' control.
Millions of people work in healthcare and many of them were no longer willing, for $11 and hour to be the bearer of bad news.
You're either too young or too rich to have been unable to intuit this. But hey, nobody forced you to self-own quite as hard as you thought wise.
In your opinion/experience, when you give terrorists some but not all of their demands, would you expect them to compromise, and consider whether the remainder of their demands is worth continued violence, or would they be emboldened by your display of weakness and press for ever more?
>In your opinion/experience, when you give terrorists some but not all of their demands, would you expect them to compromise, and consider whether the remainder of their demands is worth continued violence, or would they be emboldened by your display of weakness and press for ever more?
I'd expect them to compromise. This happened with the IRA in Northern Ireland.
Islamic terrorists won't compromise, but Islamic terrorists are a special case.
Hardly. Luigi Mangione is a symptom of the decline of the US system of democracy, not its cause. There is no End of History, the system of democracy has to continue proving its worth to people, or they will eventually reject it wholesale, with far worse consequences than a single countable death.
Political violence is an ancient, proven act of politics. It will be the fallback if and when peaceful methods fail. The shooting is an opportunity for US democracy to correct course, and if it manages, it will take off the pressure that would produce the next shooting or worse. So I believe the worst possible outcome of this shooting is precisely that nothing good comes out of it.
Look at it from an economic perspective: do you think it's possible to simply terrorize your way to better, more affordable health insurance? I don't. Shooting the CEO of Honda will not simply scare the company into selling new Accords at the same price as a Kia Sorrento; at least not in the long run. These are two different products with two different cost structures, and the profits on each are a function of the supply and demand for either one, along with how competitive the market for cars is. Insurance is no different. Adding self righteousness and indignation into the mix will not affect anyone's ability to provide a better product or service. Sorry.
I agree with your proposition. I’m not sure the car market and the healthcare market are that analogous though. There is something about the demand side in healthcare that seems an anomaly compared to cars
>One random act of violence is not going to change the economics of the US healthcare system.
You certainly seem to think so. You said that the extra security detail for a couple of CEOs "just made everybody's health insurance that much more expensive." Also, please tone down the ad hominem.
I'm an MD and even through my biased self-serving prism I can see that it shows a crazy amount of stupidity to go after/celebrate the killing of insurance company workers when it's hospitals, specialists and Pharma that are jacking up the prices. But yeah - you get the bills from them. Shoot the messenger isn't a new concept.
"Shooting the messenger" is not an appropriate metaphor here. The "messenger" makes a healthy profit himself and is therefore not a neutral party. Also, the messenger is in business cahoots with the road maintenance department, the king's advisors, the scribes who got paid to write the message, and the Pony Express:
In the immortal words of Detective Jake Peralta, "Cool motive, still murder".
More specifically and less flippantly, I have two big problems with the killing and with people cheering it on.
On the object level, I think that the case for Brian Thompson's death being justified from either a utilitarian or deontological perspective is much, much weaker than a lot of people seem to be assuming. I haven't examined in detail the premises that UHC is an unusually shitty health insurance company in terms of denying claims, and that Thompson was responsible for this, but on both counts the charges seem plausible and I'm prepared to at least accept them for the sake of argument. But going from that to "Thompson is morally responsible for thousands of deaths and killing him was a praiseworthy act" is a pretty huge leap. For one thing, the relationship between marginal medical care and either life expectancy and QALYs seems to be pretty fuzzy at a population level, c.f. the debate between Robin Hanson and Scott over medical effectiveness earlier this year, where Hanson argued that marginal health insurance coverage was almost pure waste and Scott's conclusion started out thus:
>The insurance literature doesn’t do a great job in establishing one way or the other whether extra health insurance has detectable health effects on a population. Gun to my head, I’d say it leans towards showing positive effects. But if Robin wants to fight me on this, I can’t 100% prove him wrong.
For another, people choose to buy UHC plans on the exchange, and employers choose to offer UHC group plans instead of Anthem or Blue Cross or Aetna or whatever. If UHC is killing people by excessively or unreasonably denying claims, then some of the blame accrues to the people who are choosing their plans, to state and federal regulators for providing ineffective oversight (if UHC is misrepresenting their coverage or otherwise exploiting opaqueness of the exchange system to sell shitty insurance to unwitting subscribers), and to the framers of the ACA for setting up the rules under which UHC operates.
I think the actual object-level complaint charge against Thompson is that he's a symbolic stand-in for the American health care system. I agree in general that the American health care system is a terrible mess, although I disagree with a lot of the conventional wisdom of specifics of causes and cures. But the actual individual who got murdered is a relatively weak stand-in for the target people seem to want shot.
On a meta level, the implied principle isn't that vigilantes should kill high-profile individuals who are objectively guilty of causing (or making substantially worse) major societal problems. Nor even that vigilantes should kill those who you think are sufficiently villainous. The actual principle that is being endorsed by cheering in the killing of Brian Thompson is that vigilantes should kill people *they* think are sufficiently villainous. This is an extremely dangerous principle, and one which I very much do not want endorsed in any way, since the sorts of people who are apt to go out and shoot their chosen villains are likely to be among the very worst people to make that sort of judgement. At best, people are going to get shot (or at least shot at) for bad reasons. At worst, assassinations and other retail domestic terrorism gets normalized and radicalizes more people into supporting similar acts in perceived retaliation, which is how civil wars get started.
The game theory take is you actually do want a few defectors/madmen acting on principle sprinkled about to keep everyone warned. But by definition you have to denounce and punish those actors. It only works if they act self destructively on principle and cannot be expecting some folk hero reward
Even if Thompson were a good symbolic stand-in for the failures of the US healthcare system, potentially throwing your life away to destroy his on behalf of a bunch of strangers, many of whom are already dead...well, that seems like it fails any utilitarian calculus.
"But going from that to 'Thompson is morally responsible for thousands of deaths and killing him was a praiseworthy act' is a pretty huge leap."
Maybe the leap for the people cheering Mangione is more: "*NOW* maybe someone in a position to do something will pay attention to our concerns????"
The catch is that the insurer is mostly just in the middle between the folks paying premiums and the folks making claims. If UHC approved 2x as much care (measured in dollars) then premiums would have to double. People want lower premiums and more health care ... that won't happen because of major changes to the *insurance* companies.
And ... if I'm reading the United Health Care financial statements correctly they took in $367 billion last year in premiums and paid out $280 billion in claims. That leaves $87 billion. They then claim $59 billion in operating costs (salaries, rent on buildings, ...) leaving about $28 billion in operating income. $28 billion of our $367 billion is 7.5%.
I expect much of the rage is unquantified ("I want MORE care and LOWER premiums), but I expect that the CxO folks running the company would cut the $59 billion in operating expenses if they could.
If run at cost it looks like everyone's bill would drop by about 7%.
But UHC rejected 37% of claims, and BCBS only 8%. BCBS works as a business, so it must be possible to run an insurance co. without being unreasonably stingy about paying for needed health care.
I haven't checked, but I suspect UHC is substantially cheaper. Not necessarily proportionally cheaper to the claim rejection rate, since:
1. Not all rejected claims are equal in cost. A claim might be a major hospital procedure or a 30 day supply of a cheap generic drug.
2. Not all claim rejections stick. It's pretty common in my experience (as a patient with Aetna insurance) for a claim to be rejected, then the doctor resubmits it with minor modifications and it goes through, or they substitute a slightly different procedure or medication that is covered without fuss.
But to the extent that UHC pays out less money by rejecting more claims, I would expect that to be strongly reflected in premiums, since one of the ACA's major provisions was requiring health insurance companies to spend a minimum of 80 or 85% of premiums on claim payouts, capping profit, admin costs, marketing, etc at 15-20% of premiums. The 80% minimum applies to individual and small group policies, and the 85% minimum applies to large group policies.
I think you're probably right about that. It's my observation that people with well-paying jobs tend to have BCBS, people with low-paying ones tend to have UBH. (I'm a psychologist, so hear from patients and potential ones what their insurance is.).
UBH is loathed by therapists. In the past I have had 2 patients with UBH, and for both the company made big mistakes (in their own favor) in payments to me, and straightening it out was a huge hassle, involving multiple phone calls with different people telling me different things, etc etc. After that I refused to take UBH any more. Now I have a patient with UBH who pays me out of pocket and submits my receipts to UBH to get reimbursed, and they are doing the same thing to him. They announce that before paying him they want various kinds of documentation from me about treatment plan and progress. I send it well within the deadline, but their next reimbursement check to him is for $0 with a note saying "required documentation was not received." So then he calls and tells them the date it was faxed to them, along with proof of the fax, and they say well send it again. So we do. But they still haven't paid because they say they are behind in processing things. He asks if they have received the documentation. They say don't know.
I don't object to a company that charges lower premiums covering few things and having bigger deductibles, but I do object to forcing patients or doctors to spend hours getting the money they are owed. Also I object to rejecting a claim that should be paid because it meets the company's own criterion. When I worked at a mental hospital staff had to sometimes fight to get more time covered for people who truly needed it. For instance, the patient might be covered for inpatient treatment if they were suicidal. Obviously there are degrees of suicidality. Still, there are situations where reasonable people would agree that the patient is at substantial risk. For instance they were hospitalized after actually attempting suicide. They have been in the hospital 7 days, and appear and report being deeply depressed. If you ask them if they are feeling suicidal they say they wish they were dead and are sure they would be better off dead, but they don't currently have a well-worked out plan for another attempt. Reasonable clinicians will conclude that somebody who survived a suicide attempt by luck 7 days ago and still wishes they were dead is at very substantial risk. UBH was known for saying, yeah, but where's the beef? It's been 7 days and they haven't made another attempt or even scraped their arms with a fork or something. That's just dishonesty on the part of UBH. They don't really believe the person is safe for discharge, but they pushed, sometimes successfully, for a person to be discharged even though their policy said they were covered if they were a suicide risk.
After thinking about this a bit, I suspect what's going on is that UHC is trying to pare their admin costs to the bone because they have a fixed slice of their revenue for overhead+profit. What you're describing fits perfectly if they're pushing as much admin costs to providers as possible, using poorly-trained (and probably high-turnover) employees to process claims according to rigid scripts, and managing the employees with quotas that incentive them to find any excuse to get cases off their desk.
Maybe they function like the fake-ish auto insurers down here, that cater to people, especially newcomers, who just need a piece of paper saying they have insurance. My father was T-boned by a newcomer to our shores; when he took his vehicle to the dealership (! - it's an old person thing) they looked at this dubious info, and said - we'll just submit it to yours. Perhaps GEICO squeezed some money out of them.
I want an explanation for how earbrows grow half an inch in 5 days before I rule out he was "framed" to be a ranting lunitic who will slowly bit by bit seem less and less credible as a reaction to the positivity online/governance failure
Untangling what we know about the shooter-shooter and luigi-shooter picked up at a mc donals with the murder weapon and an open and shut case, will just be hard, its like elon tweeting "put cocaine back in coke" a day before the amber herd trail where she talked about elons drug use or whatever so "elon cocaine" gets the tweet.
Possibly, and to go even further he may have intentionally experimented on plunking his earbrows and counting the days for which day has the fastest change, while intentionally only showing his eyes
The whole thing about "imma taking a gun to mcdonals after avoiding the cops for 5 days" is fucking weird; did a cia "terror plot prevention by orginizing it" handler fail and then promise him to get rid of the gun? Was it a spy network and they made up the fake tip after unconsitutional spying (btw I heard the tip isnt getting the reward money)
ALL THAT WE NEED IS A LIVE PUBLIC DISCUSSION to resokve our issues.
I am convinced that (other than the supposed psycopaths we are all told are irredeemable) nobody who works in the health insurance (or warfare, or gambling,...) industry is a monster.
However they correctly say to themselves that if they don't cause the deaths of certain people for profit than others will. PLUS they have endless justifications - dipped in the sacred blood of millennia - for what they do.
We need to meet as humans.
However, when people like Socrates, Jesus, or Gandhi get too big for their britches they end up dead. Smaller fry, like myself, are left alive to howl into the wind.
If this singular murderer allows Luigi, Myself, or some other individual as yet unknown to us, to have the honest Human To Human public conversations with the powerful few who actually own and run the world then ZERO more deaths will be necessary, and retroactively this one will have been cleansed as necessary.
If that fails to happen, then murders and other atrocities wilk continue day in and day out ----- and the killing of Thompson will have been just another one of no redeeming value.
That what he did is horribly wrong I haven't the slightest doubt. What I think of him for doing it depends on whether he's mentally ill or just evil-- a question on which I'm agnostic.
Google did a massive study on what makes effective teams. It's worth reading if you're still toiling in the mines of Silicon Valley — especially if you're a manager.
>The researchers found that what really mattered was less about who is on the team, and more about how the team worked together. In order of importance:
> Psychological safety: Psychological safety refers to an individual’s perception of the consequences of taking an interpersonal risk or a belief that a team is safe for risk taking in the face of being seen as ignorant, incompetent, negative, or disruptive. In a team with high psychological safety, teammates feel safe to take risks around their team members. They feel confident that no one on the team will embarrass or punish anyone else for admitting a mistake, asking a question, or offering a new idea.
> Dependability: On dependable teams, members reliably complete quality work on time (vs the opposite - shirking responsibilities).
> Structure and clarity: An individual’s understanding of job expectations, the process for fulfilling these expectations, and the consequences of one’s performance are important for team effectiveness. Goals can be set at the individual or group level, and must be specific, challenging, and attainable. Google often uses Objectives and Key Results (OKRs) to help set and communicate short and long term goals.
> Meaning: Finding a sense of purpose in either the work itself or the output is important for team effectiveness. The meaning of work is personal and can vary: financial security, supporting family, helping the team succeed, or self-expression for each individual, for example.
> Impact: The results of one’s work, the subjective judgment that your work is making a difference, is important for teams. Seeing that one’s work is contributing to the organization’s goals can help reveal impact.
The researchers also discovered which variables were not significantly connected with team effectiveness at Google:
Colocation of teammates (sitting together in the same office)
How do you separate "how the team works together" from "who is on the team"? I'm pretty sure the most significant factor by far in how the team works together, is what sort of people the team is composed of.
Notwithstanding far too many Hollywood movies, you can't hire a bunch of random idiots and assholes and expect them to turn into a well-oiled machine after an extended training montage. Not even if you've read the Bignum Habits of Highly Effective Managers. As you note, safety is important. The assholes will make everyone else feel unsafe. Dependability is important. The idiots won't be dependable (except in providing idiocy).
"The researchers found that what really mattered was less about who is on the team, and more about how the team worked together. "
We've known this, in a broader sense, since the 1960's. Look up research relevant to "leadership style", "Group Roles", and "Group Cohesion" on google scholar.
It is not obvious if this generalises. A proverbial google Engineer "asking a question, or offering a new idea" is surely a valuable asset. An idiot doing the same is holding back the entire team. Of course, if true, the morale of the story would be not to hire idiots, and easy to say if you are google. In practice there is a huge middle ground between google engineer and idiot. Advice that works at google might not actually work on that middle ground.
Re Dependability: Successful teams "reliably complete quality work on time" seems to be a massive tautology. The question to be answered is how to get a team to do that.
The whole point of the research is that, in a properly structured team, there are no idiots. Probably there is a level of lack of competence that cannot be integrated into a team, but people at that level are unlikely to be hired in the first place.
"The question to be answered is how to get a team to do that." As I pointed out above, we've know how to do that for fifty years. The problem isn't technical, it's organizational politics (effective front line teams threaten management).
A stupid question is easily dismissed. Someone who is too afraid to ask a stupid question might not realize it's not actually a stupid question. And when idiots are free to ask stupid questions, more people are more likely to realize the person is an idiot and if they're a problem can more easily identify the problem.
If I had to have an idiot on my team I'd rather have one who asks stupid questions, rather than one who repeatedly gets stuck or endangers everything by being too afraid to ask stupid questions.
I believe the concept of some things being "above someone paygrade" was invented to fence which questions are useful and which questions are an attempt to understand what the questioner neither can nor needs to understand.
To repeat, I am aware that this does not apply to high end engineering teams, but anything flirting with "there are no stupid questions" seems to betray ignorance of a relevant part of reality.
One person's stupid questions, may be intelligent questions to another.
I don't know how high-end the high-end engineering team I used to work with was, but we were all getting a training in technology X (I forget what it was), and the instructor kept throwing out acronyms I was unfamiliar with. I asked him to unpack a couple, and he did. But he kept throwing out the damn acronyms. Finally, I asked him if he could put together a glossary of terms with his slides. He said, "Well if you don't have the basic knowledge of this technology, you shouldn't be in this course." He never provided the glossary. But other people in the training asked me to forward the glossary (that he never provided) to them, too. So, I assumed there were a lot of people attending the training who weren't asking questions but were too afraid to look stupid by asking questions.
This is interesting, but I’ve always found specific personnel play a huge role in team effectiveness. If you have a serial exploiter of team systems (someone who actively seeks loopholes, takes advantage of technicalities or weaknesses in protocols while ignoring their intent, etc.) then all the structure and clarity in the world can’t save the team.
And the study doesn’t even mention waffle parties or melon bars.
I started a blog about the experience of being in-the-closet in the ultra orthodox Jewish community. Would love for you to check it out at https://open.substack.com/pub/exit98
I'm a musmach of Rav Noach Weinberg (not Aish, I refused theur smicha - Rav Noach) and I can tell you that the God you don't believe in, I do not believe in either.
Greetings everyone. I think, like many, that the move in governance, but also business, education, administration, even the household, toward more top-down rules and mandates and away from the discretionary decision-making of individuals is not only making institutions less efficient and effective, but also making us all dumber, less moral, and more deceptive. And I'm not even a libertarian, not by a long shot. Norton signed me up to write a short book about this, part journalism, part empirical data from criminal justice, and part philosophical argument from first principles. The books is now available for preorder. https://wwnorton.com/books/9781324051251
Why are drugs in the USA so expensive? For example, while high costs are often attributed to research and development (R&D), why are older drugs like insulin, which should not require much R&D, still so expensive?
Drugs which are truly old, and in high demand, are generally quite inexpensive in the United States if you don't insist on the fancy brand name on the label. Drugs that seem like they should be old and cheap but aren't, I think fall into three categories:
1. Branded drugs where doctors have gotten in the habit of writing the brand name on the prescription so that the patient is locked in to a monopoly supplier. You can *usually* get the doctor to rewrite the prescription for the generic chemical name if you ask, but sometimes they've bought into the idea that Brand Name is Ineffably Better and also Price is No Object. "Epi-Pens" vs. adrenaline autoinjectors are a classic example here, and Scott had a post on the subject back in the SSC days.
2. Drugs that are variants of classic old drugs, with some tweaks to make them at least slightly better. Your example of insulin fits here. Just plain insulin, like your grandparents' generation used to spend a lifetime Not Dying from diabetes while complaining about all the annoyances, is still I think about $20-30 per month in the United States. But "insulin" now usually points to a tweaked variation on natural insulin that is much easier to properly use in the field. In the case of insulin, this is more than "slightly better". But it does come with substantial research and development and especially testing costs, which someone has to pay for (and governments mostly won't).
3. Old drugs for which there is very little demand. The problem here is that, no matter how old and off-patent the *drug* is, you aren't allowed to sell it until you've spent many tens of millions of dollars convincing the FDA that your specific factory can safely make that specific drug. Yes, even though you're using the old known-safe recipe and your factory has safely made lots of other drugs. The classic example here is Daraprim, which earned Martin Shkreli a well-deserved fifteen minutes of infamy a few years ago. By buying the one factory that made Daraprim, Shkreli was able to gain a monopoly on a low-volume market but one where he could jack up the price to absurd levels. At which point, anyone else on the outside looking in would only see the "opportunity" to spend tens of millions of dollars for the privilege of getting into a price war that would see both sides selling at low prices to a low-volume market.
There are potential solutions to all of these, but it will be a different solution for each case.
I've had a few cases of our insurance refusing to cover a prescription fill because something went wrong between the insurance and either the doctor or the pharmacy (e.g. the doctor wrote a 3-month prescription, and the insurance will only cover 1 or 2 months at a time). From these, I learned that old, very basic maintenance drugs tend to be dirt cheap, in single $ digits per month (I guess we've been lucky, as I haven't seen an exception to this heuristic yet). It's when you need a newish drug that you have a problem.
Not that simple. MAOI's for instance, are very old drugs. They were used in the 1950's as antidepressants. They are currently out of fashion, but the people I respect most in the field believe they are the most effective drugs for depression. The average cost for a month of a typical dose of the generic tranylcypromine is about $240. Could be worse, but that's pretty bad. I believe the reason for the high price is that it is used by many fewer people than, say, prozac.
The main danger with MAOI’s is that they interfere with a person’s ability to metabolize tyramine, which is a harmless amine for most people. However, if a lot of it builds up in the body it raises blood pressure. While the body is built to do fine with temporary blood pressure raises (bp goes up quite a lot during vigorous exercise, for example, and also when someone is upset), there’s a limit, and a really big excess of tyramine can raise bp so much that it’s truly dangerous. Some foods have high-ish tyramine levels, esp. highly aged cheeses. HOWEVER: In the 50’s and 60’s, studies of the amount of tyramine in various foods found enough tyramine to be a problem for people on an MAOI in quite a few foods. Redoing these measures in the last 20 years found much much less tyramine in foods than the early studies, partly because measurement methods are better now, but mostly because refrigeration is much better. People who take MAOI’s now only need to keep in mind a few restrctions, and even those they can violate in moderation and be pretty safe. A lot of the medical profession hasn’t caught up on MAOI’s, though, and don’t know this stuff. If you are curious about MAOI’s, have a look at Psychotropical, the web site of a world expert named Gillman. And other antidepressants aren’t all that awesomely safe either. There are things SSRI’s should not be mixed with. And tricyclics make doctors nervous because the lethal dose as not all that much higher than the effective dose.
For most goods that are cheaper abroad, market forces will make it so the local price is approximately what it costs to buy it where it is cheapest and ship it to you. This is illegal with drugs.
I've lived in many countries and the same meds that cost a fortune in the US cost pennies abroad.
Skip all the explanations about, "America subsidizing the rest of the world" and how "there must be a major profit payoff to pay for R&D", etc.
Those are all obfuscatory nonsense responses of no value. The simple answer is BECAUSE THEY CAN.
Oh, and also BECAUSE THEY WANT TO.
If medicine was affordable people might not be obsequious to their bosses and might pursue their passions rather than careers, but when you tie The Ability To Stay Alive to your job, welo you sure as hell better keep your job!
"Because They Can and BECAUSE THEY WANT TO" is also a response of approximately zero value.
Like, yes, obviously, given the choice between "making less money" and "making more money" approximately everyone will choose the first one. But *why* can they make that choice here and not elsewhere is the interesting question that just saying "it's greed, lol" doesn't answer. Have they not invented greed in Europe yet?
Greed was introduced to Europe in the 1500s, but it didn't take well and soon died out, as it was considered "gauche". Maybe it can make a comeback with new marketing, including a rebranding.
The profits from current drugs support the R&D costs of the new ones. The funnel is very leaky: for each marketed drug dozens more failed in an earlier phase.
The FDA's process increases the costs of these phases such that drugs which would only be marginally profitable conditional on success aren't pursued.
And other countries threaten non-enforcement of pharmaceutical patents unless they are charged essentially only variable costs, leaving the US to bear the vast majority of the aforementioned fixed costs (much of which is FDA-inflicted, so it's not entirely unfair).
To get drug prices down without choking off future life-saving innovations, the point of maximum leverage for Americans is reducing the expected cost to get a drug through the process.
People will often flawly justify the insane cost of drugs:
1) R&D costs are off topic for most drugs on the market.
2) R&D costs don't even need to be high, it's mostly a symptom of the insane mediocrity and dysfunctionalism of pharmaceutical companies
3) R&D budgets are significantly smaller than the insane marketing/propagand/corruption budget of pharmaceutical companies
About the actual question: The production costs:
1) There is little to no price negociation with the FDA. Price is artificially set as high as possible to maximize profit. There is often no competition for a long time because of bureaucracy and dysfunctionalism. Even when competition arise, there is probably a high level of cartelization/price collusion between competitors to only marginally reduce prices.
2) Making drugs is generally extremely cheap at scale
3) the prices of drugs, independently of their chemical type, is hyperinflated in diseases where you can milk more the patients which is the incarnation of evil.
4) there are broadly two main drug classes where productions costs are orders of magnitude higher than typical drugs: biosimilars/polypeptides and gene therapy.
Of them, biosimilar costs are proven to be fully fraudulent, producing polypeptides is relatively cheap, generally a monthly dose production cost is in the literature estimated at 50 dollars. Which is higher than non peptide drugs but is orders of magnitudes cheaper than market prices.
Moreover this cost can even be reduced via production techniques refinements, as outlined in the literature.
The machines to synthetize peptides are publicly available and cost less than 20000 dollars meaning that it cost less to produce your own insulin and glp1 agonists and to give them for free to people than to buy your yearly individual dose, without any SOTA refinement or economy of scale...
Such level of dysfunctionalism is peak insanity and utilitaristic crime.
The reason there aren't ONGs or biohackers selling such peptides at 100 times less costs is because 1) people don't know they can produce their own biosimilars and 2) it is probably illegal, even for unpatented generics (about endogenous molecules..).
See for example openinsulin, it was funded in 2015 and I don't think (?) they are selling insulin, they are probably standing idle doing nothing because of criminal regulation.
I'm no expert, but have heard and read several explanations that probably all play a part. Some of the more plausible ones: There's little to no price negotiation, regulation, or price pressure at the national level, like there is in single-payer systems. Medicare is basically prohibited from negotiating prices of medication or preferring inexpensive alternatives to expensive drugs. And the negotiations that do happen between insurers/payers and manufacturers tend to keep list prices high, in favor of steep discounts, post-sale rebates, etc.
Also, there's a lack of transparency in the current system. I think if drug co-pay was fixed at a percentage of list price, regardless of insurance provider & plan, patients would have more of an incentive to ask their doctor for the least expensive effective alternative, that would put more downward pressure on list prices.
Regular insulin has been around for a long time, but the drug companies have come out with insulin analogs. The insulin analogs are either fast acting or long lasting. The drug companies can keep insulin under patent by continuously coming out with improved versions. As DJ says you can get regular insulin for $25 at Walmart, it is the new analogs that are expensive.
"Making insulin with long half lives is trivial and cheap, you can e.g. pegylate it. Let's not mystify basic old tech."
Is getting the insulin with long half lives through clinical trials trivial and cheap?
Some of the costs (not all, but some ...) is because clinical trials can be expensive. The old, regular insulin is cheap to manufacture, off-patent and has gone through clinical trials. If the new stuff is missing any of these three then it can be expensive.
It's worth pointing out that once those bioreactors and purification systems are set up, the cost of *producing* the insulin drops below $5/vial. The current Regular insulin (which you can buy for <$35/vial) is manufactured in bioreactors — just like the expensive stuff that costs >$250/vial. And I can buy the same expensive insulins in Canada for about $45/vial (but I need an Rx from a Canadian MD).
And don't forget that Eli Lilly dropped the price of its insulins to $35 without going broke.
I can't speak to other drugs, but as a diabetic I recently learned I can buy insulin over the counter at Walmart for $25. There are more advanced time release versions that require a prescription but the OTC insulin is fine for my needs.
Does anybody here know of any actual *studies* supporting the idea that getting orthodontic work as a child, has positive outcomes for long-term health? We are going through the interview-orthodontists process with our kids, and... there are all these claims that cranking open their upper palates will broaden their airways, reduce headache, etc. but... I've also heard some fairly scary things about American-style orthodontics: that aggressive teeth-moving compromises long-term blood supply to the teeth and then you end up getting root canals and losing teeth in your 50s, that it fails to address the initiating problems of posture, bite, and incorrect forces applied, and so you can end up needing retainers forever, because those forces are still moving teeth out of place, even after all the work.
What's really frustrating is that there do seem to be other options out there, with at least case studies to support... but only if you live outside the US. I keep asking here about things like Planas Direct Tracks, bite blocks, Advanced Lightwire... and they look at me like I just sprouted three heads. And then they go back to their sales patter. I can't get them to explain anything, or go off-script and it's deeply frustrating, when they're asking me to spend $10k+ and 2-3 years doing painful things to my children.
And then they go and tell me "it's genetic" and I die a little inside. Most people did not have crowded teeth three generations ago, but it's rampant now, across the population. THAT'S NOT HOW GENETICS WORKS. We didn't all mutate in that time, and being told that *does not inspire confidence*.
I'm not into cosmetic medical procedures just for the sake of giving my kids a middle-class smile. I know that's un-American. Sue me.
Like, if we're going to go through with this, I NEED to have some kind of evidence that there's a real medical benefit to it. Does that evidence exist? Or is this a whole industry just for cosmetically modifying your kids?
The pop-sci book Breath by James Nestor asserts that many orthodontic ailments are absent in our ancestors because of our modern diets. He suggests that our ancestors chewed significantly more because they ate more hard, tough foods, and that many orthodontic ailments are caused by weaker jaw muscles. The effect, he argues, is exacerbated by mouth-breathing, which can case nasal passages, sinuses, and other air pathways that alleviate e.g. apnea, to shrink.
Apply grains of salt, but I would err on the side of heeding orthodontists' specific claims and rejecting their non-specific claims. Further, Nestor's evidence suggests that there are behavioral alternatives to expensive surgery.
Well, yeah. I've seen the archaeology data. IIRC, medieval people had dental arches ~1cm wider than ours, and neolithic man maybe as much as ~3cm wider. That's not genetic drift. These are genetically modern humans we're talking about. The difference is clearly environmental. But, you know, my kids are 10 and 12-- we can't wind the clock back and raise them on a diet of wild tubers and dried wooly rhinoceros jerky.
That said, has there been ANY research at all, into whether modern orthodontic practices are helpful, harmful or anything in between?
Widening the upper palate is considered beneficial if you haven't taught your children to eat tough food like carrots, whole apples and crusty, chewy bread. It's better to do this before permanent teeth start to erupt.
I am a late 20s man who still has his wisdom teeth. The few times I have moved cities I have had a tremendously difficult time finding a dentist. When I go in for regular cleanings, the first thing the actual dentist tells me post cleaning is that I should get my wisdom teeth out. When I ask why, the answer is typically "prevents long term health issues", and if I asked what specific health issues, either the question is either deflected or they say "food can get stuck, and that can lead to infection". The only food that gets stuck in my back teeth on a semi-regularly basis is pretzels, and so as what I deem to be a rational response, I eat fewer pretzels.
I had one particularly bad experience with a dentist on the north side of Chicago. The dental assistant told me they were using a new cleaning technique, which I think involved some type of UV light, but was noticeably painful when used on my bottom teeth. The dentist herself was a cosmetic dentist, and told me on my first visit that I was showing signs on gum loss on my lower teeth. I thought this was strange, as I had never heard it before from any prior dentist. Six months later, I go for a routine cleaning, and I get the same dental assistant, who uses the same UV cleaning process. When the dentist saw me, my "gum loss" was confirmed, and she referred me to an oral surgeon to fix the problem. I was naturally skeptical that I needed this multiple thousands of dollars procedure, even if my insurance would pay for it (it just sounded really inconvenient to go through all that). I checked the instagram page of the dentist, and there were multiple pictures of her and the oral surgeon to whom she referred me at networking events, sharing drinks, and partying. I was somewhat familiar with the healthcare referral process and how some practitioners can earn a sizable chunk of their income from them, and it all felt pretty icky to me. So I scheduled a cleaning with another dentist off insurance for like $100. I didn't mention my purported "gum issue", and the dentist did not make any mention of it. Needless to say, I stopped going to the cosmetic dentist, and stayed with the new one for the rest of my time in Chicago. I do not trust dentists.
I didn't want to have my wisdom teeth pulled out. Issues with overcrowding started showing up in my 30s at which point I had to have them removed. It was a nightmare. Should have done it back at 20. Sometimes waiting for problems to show up is not the best idea.
Had one removed at 35, no issue whatsoever. Was a bit sick though for a day or two, but everything recovered perfectly without issue. Pulling it took like a minute tops.
I am almost 50 and I never had any problem with my wisdom teeth. So, both things can happen, the question is how frequently.
(Trying to ask the experts doesn't help much. In my experience, different experts give contradictory advice, each of them with an expression of absolute certainty.)
The US is very pro pulling wisdom teeth early. Other countries aren't. There's the cynical argument about how it is monetized which I have sympathy for. There is also the argument that people including medical professionals herd. Either way - push back. Wait for an issue to arise - it might never. Downside is having them removed when you are older is it is probably going to take a few more days to recover. Worth it.
My understanding is that a downside is that wisdom teeth, being harder to clean properly (especially if they're partially erupted) get cavities and decay, and then spread that decay to neighboring teeth. More frequent visits to the dentist would have caught it, but at any rate, I got two wisdom teeth yanked in my mid-20s.
But I agree that there seems to be a cultural difference: I have one wisdom tooth that's fine (not perfectly erupted, but no cavities), and my US dentists advise me to get oral surgery to get it removed, and my India dentist tells me to leave it be until there's a problem.
> Ten million third molars (wisdom teeth) are extracted from approximately 5 million people in the United States each year at an annual cost of over $3 billion. In addition, more than 11 million patient days of “standard discomfort or disability”—pain, swelling, bruising, and malaise—result postoperatively, and more than 11000 people suffer permanent paresthesia—numbness of the lip, tongue, and cheek—as a consequence of nerve injury during the surgery. At least two thirds of these extractions, associated costs, and injuries are unnecessary, constituting a silent epidemic of iatrogenic injury that afflicts tens of thousands of people with lifelong discomfort and disability.
— "The Prophylactic Extraction of Third Molars: A Public Health Hazard"
> The British National Institute for Clinical Excellence is unequivocal in its recommendation, adopted by the National Health Service: “The practice of prophylactic removal of pathology-free impacted third molars should be discontinued. . . . There is no reliable evidence to support a health benefit to patients from the prophylactic removal of pathology-free impacted teeth.”9(p1–2) The conditions for which extraction is justified include nonrestorable dental caries, pulpal infection, cellulitis, recurrent pericoronitis, abscesses, cysts, and fractures.
And there's a risk of delayed-onset infections after the extractions. The study below suggests the use of clotting agents (hemostatic use), high blood pressure, or certain positionings of the tooth before extraction caused a higher incidence of delayed post-operative infections.
"Risk Factors for Delayed-Onset Infection after Mandibular Wisdom Tooth Extractions" by Ryo Miyazaki et al...
Haven't been able to find any mortality numbers, but the data indicates that the majority of extractions are unnecessary and can cause some serious downstream issues.
As someone who knows nothing about this, I would err on the side of non-intervention until you have a better reason to take action. Just make sure they breathe through their nose and eat food that exercises the jaw.
This has been my approach both personally and as a parent. I share Yarrow's skepticism about this topic. Not a fixed opinion to be clear, I live by the maxim that "skepticism is a process not a position". But if actual longitudinal evidence along the lines asked for exists I've never yet seen it.
Broadly in my immediate family we're on the lucky side as far as dental genetics; my wife not so much. Our son is now 12 and there has been talk of braces but nothing any more serious than that has been suggested. I've now reached my 60s, still have my wisdom teeth and they're perfectly stable and dentists have quit asking me about it; my mother died at 91 with a full set of healthy wisdoms still in place. Whereas my wife had to have hers removed years ago as did my adult son (son of my first wife). My elder brother now has serious dental issues but that is pretty obviously because he's been dumb about it and hasn't followed even basic dentist advice in decades.
So in summary we'll do what our son requires when or if he requires it; I will resist firmly any significant interventions which are not backed up by some real science. (And am irritated by the claim that such interventions represent the "better safe than sorry" approach -- they do not!)
By the way since we're here I will share a recent eye-roll-inspiring development on a tangent but in a way it applies. We have a family dog, a healthy bouncy 50-pound mutt who we all love dearly. My wife and I are both experienced dog owners, in my case since childhood. Our doggo gets solid regular vet care, all shots etc. He's now about 5 years old.
Earlier this year the vet said his teeth needed cleaning. His teeth are structurally fine; his breath at any given moment is not materially better nor worse than family dogs' breath tends to be; he has no actual cavities or other dental issues. Just the discoloration (plaque) that dogs get because, well, they eat like dogs. He gets the "dental stick" type treats which are supposedly helping clean his teeth; they may honestly just be treats but, whatever.
Newly, the vet said that manually cleaning a squirming dogs' teeth is not very effective so they should once a year knock him out to be able to do it properly. And this service costs several hundred dollars.
My wife tends to assume that doctors' opinions are the final word and she was doing the visit so said okay. They knocked him out for half a day, cleaned his teeth which then looked very nice. The dog woke up from the anesthesia moaning in confusion and fear and wasn't himself again until the next day.
I subsequently learned from others at the local dog park that all the vets are declaring this to be a new normal in canine care, once a year. Well it isn't in our house: I put my foot down and said, no. We will not be doing that again to any dog of mine.
Effective generally-safe anesthesia is one of the most-beneficial inventions in medical history, arguably in scientific history generally. It is also not zero-risk. There is a reason that doctors for humans don't do _general_ anesthesia until it's actually needed for the operation.
I treasure our pup, which does not make me think that his teeth need to be pristine or that we can't allow any risk of some normal dental slippage as he ages. Knocking him out entirely for the sake of simple cleaning is in my view on the edge of outrageous as a risk-reward medical decision. I managed not to say FFS out loud but my wife knows I was thinking it.
I agreed that we (which means me, assisted by our son who tries to hold the squirming pooch) would manually do some canine tooth-brushing once a week. Which is fine, kind of amusing actually, and we do it. My expectation is that it's just slowing down the rate of plaque buildup, fine.
But actually knocking him out just to clean his perfectly-healthy teeth...?? I'm still irritated by the whole thing. Including also the several hundred dollars part of course.
I should add, for completeness, that the kids do, in fact, have really crooked/crowded teeth, narrow dental arches, and the older one a pretty marked overbite.
It's just... I thought there were ways (such as PDTs, bite blocks, and even night-guard appliances) to correct bite by adding a little height at the molars, that didn't involve literally three years of metal-mouth and yoinking teeth around. I don't actually care if their teeth aren't perfectly straight, if that means those teeth are still healthy into old age. They've never had any cavities-- I certainly don't want to change that. In my own family, there's a huge correlation between cavities and orthodontic work. My mom and sister both had extensive orthodontics, and also suffered a lot of cavities-- seemed like they were always getting fillings replaced. Me and my other sister both have overbites but escaped orthodontics. Between us, we have zero cavities.
What nobody has been able to explain in any kind of satisfactory way is: with an overbite, if you move your jaw so that your front teeth line up correctly, there's this big space between the molars: they don't touch anymore. So, with all these appliances and crap to "pull the jaw forward" (or the one I hate more: "*grow* the jaw forward" wtf kind of magic car-salesman talk is that?), how are you going to fix the problem where the molars won't touch? Because I'm not letting you file off his front teeth, so you'd better have some other strategy for that.
For what it's worth, I'm someone who has had a huge number of cavities although I eat pretty well and brush/floss obsessively. And on the other hand, my teeth are quite well aligned and I never needed any orthodontics. The same is true of at least one of my other family members. This is no more or less anecdotal than your own reports about your family, but, in lieu of more robust data, I nevertheless suspect that the correlation between orthodontics and cavities you have observed is spurious.
I remember coming upon Scott praising the prose of G. K. Chesterton and filing away that fact.
Just recently I came across a collection of G. K. C. essays and found things like this there:
"Some little time ago I stood among immemorial English trees that seemed to take hold upon the stars like a brood of Ygdrasils."
"In certain endless uplands, uplands like great flats gone dizzy, slopes that seem to contradict the idea that there is even such a thing as a level, and make us all realise that we live on a planet with a sloping roof, you will come from time to time upon whole valleys filled with loose rocks and boulders, so big as to be like mountains broken loose."
Wow. I don't think any imagery hit me like this since first reading E. B. White. And I think I detect a hint of similarity in eg. the recent post, "The Innocent and the Beautiful Have No Enemy But Time."
I'm surprised because the styling of G. K. C. makes the writing feel fresh and strong despite it being over a century old. Maybe because of the contrast with most of what I read on substack?
Why don't more writers use bold, original imagery? Is it just fashion? Is it too difficult? Are there political connotations?
Part of it is fashion--Chesterton is using a deliberately archaic style even for his time, although at his time it was a bit more common to use "old-fashioned" devices. Even writers with polished prose styles and a penchant for elaborate images aren't likely to use that kind of language. Although there are exceptions, especially in fiction--Cormac McCarthy and W.G. Sebald are a few relatively contemporary examples.
But in general SSC and rationalist-adjacent blogs don't go for elaborate stylistic stuff. You'd be better off looking to a literary magazine for that kind of thing--the London Review of Books, n+1 (sometimes), The Paris Review etc. Also, writing like this is pretty laborious, so people who do it aren't as likely to do it on a blog, which feels a bit disposable.
I love this imagery, because if you've ever seen bars of lead broken open, it hits you "Yes! That's what it's like!" There's something almost surrealist about day breaking like bars of lead, and that's when I remember he went to the Slade School of Art, so he has a very visual imagination:
"The clear moon that had lit up Chiswick had gone down by the time that they passed Battersea, and when they came under the enormous bulk of Westminster day had already begun to break. It broke like the splitting of great bars of lead, showing bars of silver; and these had brightened like white fire when the tug, changing its onward course, turned inward to a large landing stage rather beyond Charing Cross."
Nobody is obliged to like everything, but as an Irish teenager he was the first English writer I read who didn't engage in "and of course those Irish, well what can you do, they brought it all on themselves" type of shoulder-shrugging.
Then I found out he was a Catholic convert and went "Ah, that explains it!"
Does it count as nominative determinism when the names in question are only apt in a different language from that spoken by their holders? "Killinger" probably doesn't imply anything in German, for instance.
Compounding the joke, the real Killinger was the German liason officer tasked with asking the Romanians to adopt a more disciplined implementation of the Holocaust because their more openly enthusiastic version was making them all look bad, As Scott put it in the SSC piece, if a literal Nazi named "Baron von Killinger" is telling you that you need to tone down the war crimes a bit, you may indeed be the baddies.
(Just to be clear, Killinger was not one of the good guys in this story, he was just a more pragmatic and prudent sort of baddie)
Managed to not even hear about brain zappers until recently. Anyone here who tried it? How was the experience?
Related question: has the potential of ECT been under-researched due to stigma, e.g., can it actually induce wellness despite no clinical depressive/manic diagnosis?
I’m not an expert, but have observed ECT and received TMS. There is indeed a lot of stigma around ECT in particular, which may partially explain a general sense from psychiatrists that it is woefully underutilized. However, it also requires anesthetic sedation and can cause memory loss, so you’d have to have good reasoning for why it would work on mild cases in order to ethically subject research subjects to it.
I think TMS is underutilized because the machine is very expensive and it requires very frequent visits, so it’s hard to get it covered by insurance. I don’t think it helped me but could be wrong
The overall cost of the JWST was about $10bn, but a huge chunk of this (a majority?) must be the R&D, testing, building the tools, etc. Now that we know it works, couldn't we double or triple the scientific output we get from the JWST by building a few more at a fraction of the initial cost?
Others have pointed out that the cost savings aren't as great as you might expect; as a card-carrying rocket scientist, I concur.
But there's also the question of what you are really going to accomplish. There's no near-term practical value to life on hydrogen ocean worlds orbiting other stars; we aren't going to be able to make use of it to make our own lives better. Except that we want to know, and we would enjoy the discovery, and those are good enough reasons to devote some level of resources to the endeavour.
One JWST, well maintained(*), will eventually image everything in the universe that can be imaged by JWST. There's only 4*pi steradians in the sky; divide by the JWST field of view and multiply by exposure time, and that's it. The life on hydrogen ocean worlds, or whatever, we'll find it. Well, maybe not us, but our descendants will.
What have you got against our descendants, that you want to hog all the glory of discovery for our generation and leave none for theirs? Hedonic adaptation suggests that we'll get less out of each new JWST-scale discovery, but each new generation will get the full first-time experience. Maximum utils at minimum cost = build one JWST and build it to last. Same for any similar instrument, including the hypothetical even better successor to JWST.
Where the spam-the-sky-with-telescopes strategy would be of value, is in observation of unpredictable ephemeral events. It would be nice if we could have a dedicated space telescope looking at each potential supernova candidate within say a thousand years. And we could use a few more telescopes looking for asteroids and (especially) comets on Earth-impacting trajectories.
* JWST is not actually well designed to *be* maintained, but that's a different issue and we're working on it.
I'm not expecting JWST to need repairs soon. When it does reach that point, the best solution would probably be a robotic servicing mission, but if we need an astronaut to handle it (and we might), then a Starship or an SLS/Orion should be able to reach L2. And really, I expect we'll have something better than either by then.
The problem is, JWST wasn't designed to be serviced on orbit, so even if you've got astronauts to do the clever fiddly stuff, it may be that the parts which need to be replaced (probably the gyros) will be buried too deep in the system to be reached without disassembling the whole thing.
We've got a good many people looking into how to design a satellite or spacecraft so it can easily be serviced on orbit, and we've got some other people looking into how to do what we can for satellites that were not so designed. So, best case, when JWST needs repairs we'll be able to send a customized robot to do the job and it will just be more tediously annoying than it would have been if we'd designed JWST to play nicely with standard off-the-rack servicing robots.
Worst case, JWST itself will be a lost cause. In which case, don't replace it until you've updated the design to include all the fiducial markers and grappling points and quick-connect fittings, etc, and then replace it with one such vehicle.
OK, your comment implies we can reliably keep track of JWST after it drifts out of its halo orbit (and it won't accidentally fry itself looking at the sun), which was the part I was missing.
I had simply assumed it wouldn't last long enough to "cover the sky" with a reasonable amount of time, but if it's as you lay out I'm much less concerned then (footnote noted).
I am the human that has studied the most the ways to revolutionize astronomy metrology including the scalability of space mirrors, given it is a philosophical imperative.
The answer to why we have no large space telescopes and even less in large numbers is the extreme prevalence of mediocrity, self sabotage and institutional fraud.
I could go on about:
the prospects of alternative mirror materials (aluminium and other metals (there is no oxidation in space)
liquid mirrors (recently tested in the ISS) (microgravity naturally gives perfect liquid mirror smoothness)
alternatives to mirrors (lenses, etc)
foldable or stretchable mirrors
expandable mirrors
rectangular spinning mirrors (kinda crazy almost nobody on earth though of that) (no spin energy loss in space)
segmented mirrors
etc
Science progress is always about Hanlon's razor, and even If I am the human that most exhaustively know how to disrupt mirror scalability and economics through innovation, it is, sadly, as often, a distraction, a digression versus the actual bottlenecks that do not stems from technology but from the realization that the world is led by people that have borderline, scoped dementia.
Not realizing humans are universally, locally insane is part of the mental disease, a blind spot bias.
Not only is producing JWST clones cheap, but JWST is also peak overengineering about weak optimization for IR, that are neither potent, nor needed for UV and optical.
The use of materials (platine, etc) with no industry in this domain is pathological, doing a segmented mirror is actually trivial.
But it also comes from a false premise, not only are people wrong and distracted but they are chaining consecutive errors and false premises.
It is a criminal false premise that we even need segmented mirrors to reach JWST aperture size.
A monolithic space mirror is simply limited by the rocket max payload diameter.
JWST aperture is 6.5meters.
Starship payload diameter is 9meters which means it can make a monolithic space mirror larger than JWST.
Saturn V was 10 meters.
Even if we limit ourselves to Ariane V, it has a diameter of 5.4 meters, which is sufficently close to JWST to completely revolutionnize science (you have no idea of all the things this enable)
So if we already had rockets that enable monolithic mirrors to be JWST like, twice the size of hubble (and that would have higher resolution that JWST since being optical increase resolution ten times (the crime of JWST as not being UVOIR is another insanity), why
didn't we do it and why isn't it even planned in the decades to come?
Production costs of large monolithic mirrors are negligible.
The reason for the ESA is dementia level self sabotage:
I am among the extremely low number of people to actually know why Hershel and euclid have precisely their specific and low apertures.
It is because of contingent limitations in the crystallization of silicon carbide mirrors, herschel was the max size and it wasn't diffraction limited which means this low quality mirror was relegated to far infrared.
Euclid size was the limit diffraction limited max size for an optical telescope.
The limit was recently broke to 4 meters by chinese manufacturers, again a harmful distraction. The concept of silicon carbide mirrors and the cartelization of this industry ESA created itself, is one of the largest scientific crimes of the 21st century.
As to why NASA which enriched glass mirrors have no scalability limits (up to 11 meters) did not launch large monolithic mirrors cannot be explained by self sabotage of ad-hoc technological reasons, the answer must stems therefore from inept managerial reasons
1) NASA is regularly stopped by the defense industry and coerced to use repurposed military built old mirrors.
2) conflict of interest with the segmented mirrors lobby
3) probable inept induction from N=1 that since JWST blew up costs and delays that any large space telescope must be too budget and delay "risky" and therefore have in a self fulfiling prophecy way, delayed funding to post 2030s missions.
I can singularly and multiple times answer how to disrupt space mirrors and therefore science both on the mediocrity and the fraudulent aspects, but the dismissal of the existence of dementia level self sabotage and fraud by the "rationalists" is what allows the world to keep beeing so broken.
Whilst I definitely can't assess the science you're discussing, I am somewhat familiar with comparable *types* of projects, and find this entirely plausible.
JWST is still young. If the competitor confirms Madhusudhan's findings and the planet still seems interesting, astronomers can propose for more JWST time to look at the planet again.
The R&D is a one-time cost, but the economies of scale between producing 1 and producing a small N aren't great. The fabrication and validation still have to be performed by experts - imagine if each iPhone had to be hand-assembled by a team of PhDs. While building more copies is more cost-efficient to some extent, and linearly increases the sheer quantity of science that can be performed, this has to be traded off against performing an entirely different mission. If 5 JWSTs can be built for the cost of 1 JWST and 1 novel mission, is the incremental benefit of 4x JWST hours going to lower-priority astronomy projects (that weren't judged worthy enough to be allocated on the first/only telescope) more useful than what the 1 novel mission would provide? These are the sorts of things that the project administrators are considering.
We *have* made twins before (Pioneer 10/11, Voyager 1/2, Spirit/Opportunity), but I haven't seen cost breakdowns. It would be interesting to see historical accounts of why those projects were designed as twins versus alternatives. (Curiosity and Perseverance aren't twins and were separated by a fair amount of time and significant upgrades, but they do have a lot of commonality.) All of the examples I can think of (other than crewed missions like Apollo that ran increasingly powerful versions before being cancelled, which is a different kind of project structure) are exploration missions, where the benefit of being able to send the probe/rover to a different location means that you get to do different science there, and you can't change your mind after the trajectory is set. In contrast, telescope time can be allocated anywhere in the sky, after the telescope is launched.
Realistically though, JWST was so massively over-budget and over-schedule that it's a wonder it wasn't cancelled entirely. There's no way we could have gotten another copy, with how science is funded in this world. (And it's a wonder that it worked, speaking to the incredible skill of its team. Stuff unfolding in space is scary hard - look at Galileo's messed-up antenna.)
The considerations here, which I don't know how to estimate, are:
1) How good are the people in charge of allocating JWST hours at rank-ordering the most useful science to do with it?
2) How quickly does the value of the JWST projects drop, is it a sharp decrease (so basically all the value comes JWST #1), or does it drop off gradually, such that JWST #2 contributes 80% as much?
Contrary to popular belief, it’s not actually that hard to build a deep space telescope. The necessary precision of the reflectors for the same resolution decreases with the area.
The JWST is such an amazing device because of its constraints, and the ingenuity that it took to work around those constraints. If you had a bunch of guys with some steel sheets, welding equipment, some precision measuring, and a bit of polish, you could imaginable build a better telescope by hand.
That’s of course an oversimplification, but the idea holds, if you have a larger payload bay, and more mass to orbit, you can make the telescope bigger and therefore higher resolution (or bigger and less precise for the same resolution). If the cost per telescope is lowered, the pressure to get everything absolutely 1000% right goes way down, and thus the price of the whole thing goes down dramatically.
"The necessary precision of the reflectors for the same resolution decreases with the area."
That's not true. The precision of the reflectors has to be a fraction of a wavelength, and the wavelength of light doesn't change as the telescope gets bigger.
I don't know what your experience is with building telescopes in space, but I have actually helped build a few, and it is not quite as simple as you are making out. Telescopes like the James Webb needed new technologies to help shield them and keep them cold. You need precision to keep things aligned and together under really tough thermal environments, you have to make sure your materials and welds work as expected in a vacuum and in zero-g, and to work out ways to keep the telescope orientated properly in space to avoid overheating things, exposing them to micrometeorite damage or just to point at whatever you are looking at. It's really not something you could weld together in an afternoon.
That said, I think rather than mass production, some kind of standardisation could help with building cheaper telescopes in the future. People are still very much in the mindset that every observatory is a very expensive one-off, whereas we might be able to get more data by producing more but cheaper probes and telescopes. Even so, we'll always have the flagship observatories like the James Webb which are oversubscribed, simply because they can push the limits and that's where a lot of the most interesting stuff to be found is.
Thanks; I work in infrastructure finance, and safe to say I have seen the impacts of having a learning curve Vs not in the past decade or so on various technologies, so was wondering how applicable it would be for something like this.
Also, I love this comment section, this is the place I was most confident I would get a reply from someone who was actually involved in a space telescope!
You’re right, it was mostly an imagined experiment to illustrate the point that bigger = less precision necessary for the same resolution. If we can launch bigger, less precision is necessary, if less precision is necessary, there’s lower cost. If there’s lower cost, we can launch more or otherwise not spend the huge amounts of money to be as near to 100% sure the mission will succeed.
The scale where welding things together would actually work as an effective telescope (in combination with all the other things like temperature control and proper orientation) is probably humongous (it would take a lot of math to figure it out). If we could just snap our fingers and put whatever we wanted in deep space though, we could probably build a higher resolution JWST on the order of tens of millions, or hundreds of millions of dollars.
A lot of the complexity of building JWST had to do with folding it up so it could fit into an Ariane 5 payload bay. With the much larger and potentially much cheaper Starship it might make sense to design the next big space telescope for that.
I just wrote a reply to another comment rubbishing AI safety/Pause AI. Despite standing by my core point (it's not clear it's tractable and also problems that impact rich people are unlikely to be good EA candidates A Priori), I want to go against one argument I've seen, which essentially argues that international coordination is in principle impossible.
It isn't!
CFCs were globally banned and the ozone hole is retreating, chemical weapons have been held down pretty successfully (yes, they're less useful for maneuver warfare anyway, but international pressure has kept it down, and plenty of countries don't have a maneuver doctrine), we've collectively decided to not go there with human cloning, bioweapons have been suppressed very effectively...
I also think one can make the argument that at least fusion weapons were a (bad) choice and that we *could* have backed off from that cliff, but chose not to, with potentially world-ending consequences. We did successfully back from from having 70,000 nuclear warheads to c.12,000 now, which is still enough to wipe out every major city but maybe not enough to *definitely* end human life.
Advanced AI is, I will freely admit, particularly challenging here given the non-weapon applications and a more multipolar world, but if energy and compute requirements continue increasing then the ability to train a super intelligence looks more like a Manhattan project than a terrorist cell (if it becomes much cheaper in materiel and energy then that obviously becomes much more difficult to control).
Yeah, you should always be skeptical when someone personally interested in a niche issue (especially if it's in their personal financial interests) tries to explain to you that regulation of their niche issue is impossible. That's not to say that regulation is *never* impossible, but when "guy shitting in the deli meat because that's his fetish" gives you his 50-page treatise on how there's no point in a law that says he can't shit in the deli meat, because some Chinese guy in a Beijing processing plant can still do it so its meaningless to even try to start regulating, you should consider that you're dealing with a motivated source rather than an honest one, and "impossibility" may be less parts disinterested reflection on reality and more parts thin excuse because he'd personally rather not stop.
It is easy for pessimism to become a self-fulfilling prophecy: "X is impossible, therefore we shouldn't even try it." No one tries X. Unsurprisingly, X does not happen. "See, I told you X was impossible!"
Our ancestors could fly to the Moon, and eradicate smallpox. If we didn't know that it happened, it would be too easy to provide clever arguments why something like that is impossible and anyone who claims otherwise is just an uneducated fool who doesn't understand that the world is complicated.
This reminds me of the covid vaccine rollout. Without Israel as an example, who managed to more than double the rate of other western countries, nobody would have known it was possible to do that.
Covid made a lot of insanity visible. For me, a fascinating thing was during the first months (before people got tired) how everyone who remained in Slovakia posted on facebook that we need to be careful and protect the people around us, especially the older ones; and everyone who worked in UK was like "well, the virus is unstoppable, obviously the old and weak must die so that the young and strong can live free". And everyone was 100% sure that their way is the obvious one.
And, looking at the numbers, during the initial spring virtually no one died in Slovakia, I think we had by far the lowest numbers in EU. Then in autumn we fucked up completely, but that's another story. In a parallel universe where the vaccine was developer a few months earlier, it would be an incredible success.
When the vaccines came, again the cultural differences were great. Many countries made the vaccination mandatory, but with a lot of waiting, and often inconvenient. In contrast, in Austria it was voluntary, but the vaccines were easily available. Like, in some big parks in Vienna they built tents where they vaccinated anyone who walked around and wanted to give it a try; no need to register in some system and then wait until they call you. They didn't even care if you were a foreigner outside their healthcare system, so we went there to get our kids vaccinated a few months earlier.
So yeah, often things that are "impossible, and all reasonable people know it" in one country, are perfectly possible in another, sometimes the next door.
I'm not particularly familiar with Slovakia's example, but I'm skeptical of "but that's another story" w/re the autumn fuck-up. I think, broadly speaking, there are quite a few countries that achieve anomalously good results in the first 3-6 months of the pandemic, but by using methods that were not plausibly sustainable in the long run. And which thus mostly just postponed a bunch of deaths by 3-6 months.
If "and then we fucked up" refers to not sustaining the unsustainable, then no, that's not a different story, That's just the second half of the same story.
And yes, if we were going to get a highly effective vaccine available in quantity in 3-6 months, then aggressive temporary containment measures at the outset would have been a much better decision. But that didn't happen, and I don't think anyone ever seriously believed that it was going to happen.
Something I'd be curious to know is what the odds of developing a working vaccine at all were. Because early it was commonly heard that it might take years to develop a vaccine if ever. But what did the actual experts think at the time? And what would have been a good estimate in hindsight?
If we need to develop a vaccine for the new bird flu, what are the odds we can do it and in what time frame?
Seems to me that *developing* a vaccine is relatively quick with the science we have today. The "source code" of covid virus was available practically overnight. Synthesizing parts of RNA is business as usual. See: https://radvac.org/
Now getting the vaccine *approved*, that is the real challenge that takes several years and costs millions (i.e. you can't do it on your own). A naive person might imagine that we could streamline the process during a pandemic, but that's not how these things work.
I'm not sure that's true; human genetic engineering / cloning seems both useful and to not have obvious substitutes, but that has been pretty effectively banned.
Have you not heard of polygenic embryo screening? Human genetic engineering proper doesn't exist because it's not currently technologically feasible. If it was possible to guarantee an extra 15 IQ points for children, there is no force on earth that could prevent it from happening.
Similarly the promise of AI is simply too valuable for it to be held back by decree. Banning CFCs was free by comparison. The proper analogy is drug trafficking. The war on drugs did basically nothing to curb drug use. The lure of outsized profits is simply too strong for any centrally-planned policy to disrupt.
Harvesting eggcells is slow and difficult for a woman, instead of harvesting hundreds of eggcells for +15 points gain you could have cloned 150 iq person and skip most of inconvenience. It already exists.
I agree with much of what you say here, and agree that if it was easily and safely available it would be widely used.
We need to ask, however, *why* it isn't technologically feasible at the moment, and my view is that restrictions in place (such as not being able to cultivate embryos beyond two weeks) have significantly slowed down research in this field, such that it has remained technologically infeasible to do. Do you disagree?
I'm actually not sure. I think the only thing that was inhibited was direct human cloning. The tools that could potentially enable real genetic engineering (Cas9, CRISPR) have developed rapidly.
> I actually think more people would want to select for beauty, health and longevity than would for IQ.
100% - you forgot "athletic potential," sure to be popular among fathers aiming for the next quarterback, and height, too, which will be a big one for sons that will likely set off a fun arms race that will enable ordering people generationally on sight (much like we do with names such as Ethel and Gladys, or Jaden and Aiden).
Arguably more important than any of them would be "robust mental health" or "high happiness set point," but to my knowledge we don't have GWAS's on those things yet.
I peeked down the rabbit hole of jihadist propaganda (hezbollah children shows, how Iranian public intellectuals discuss their strategic aims when not speaking to a western audience, etc.) and it feels unhinged and scary. If I wasn't observing an objectively low/decreasing amount of terrorism in the world, or say Saudi Arabia de emphasizing antisemitic Islamic verses from their curriculum, I would be quite concerned. What is a sane way to understand/track the radicalizing and moderating forces in the Muslim world, and to project into the future likely outcomes. I am also aware that if I listened to Maga world for a similar chunk of time I would be genuinely worried about civil war, even though I actually think everything is totally fine and we are in the middle of a wonderfully productive, moderating, cultural self reflection and reckoning.
Look at the new masters of Syria: they’ve adroitly transitioned from jihadis to nationalists to maybe modern trade-oriented technocrats over a couple decades. People change in unexpected ways.
Bill Roggio's Long War Journal once upon a time was an excellent tracker of various Islamic issues and it's valence to the West (read: the US), with some ability to interpret various ideologies. I know it was pretty influential in US Mil circles; not sure if that's still true after the fall of Afghanistan, though.
Consider a change in perspective: how much do you think people should change their views after listening to the views showcased by LibsOfTikTok, for example?
My vague understanding is 'teachers brag about indoctrinating their students with far left cultural dogma in a way many parents would strongly object to', and stuff like that. If that's wrong please correct me.
I think that's less a change of perspective, than another example of the type of thing I'm asking about, which is broadly 'how do we judge the strengths of seemingly crazy movements that don't publicly self-reflect or have a internal diversity of views, when those movements actively threaten anyone outside of the movement who disagrees'.
Sure, you can add 'libsoftiktok content providers' to 'magaverse' and 'jihadists' on that list.
That's easy: just look at whether they're winning. In the case of the new right, well... They just won control of every branch of government of the most powerful country in the world, with popular support to boot. And they did that despite their party spewing blatantly fascist rhetoric in a supposedly democratic country! So that makes them pretty strong!
I disagree with your implication that "fascist" and "democratic" are incompatible: it's not LIBERAL, certainly, but a lot of decidedly illiberal policies are wildly popular.
Hezbollah and Iran, as Shia, are not part of the global Islamic mainstream. For all their rhetoric, they don't have much influence outside of their local struggles, and with rare exceptions they aren't responsible for global terrorism.
You'd get a better read on the pulse if global Islam by checking out Salafist channels on YouTube.
Metaculus is useless for prediction of "Black Swan" events, or even "light-gray swans" like the recent events in Syria, or October 7 (the evidence was there, but it wasn't receiving the right kind of attention).
I agree with NoRandomWalk that some Islamism rhetoric is alarming and under-exposed to western audience.
It could also have an effect on European cities with a large amount Muslim immigrants / refugees.
I'm struggling with an ontological account of causation and laws of physics.
First, I'm sure that there is a certain last level of depth that a correct account can reach. In the end, the world just is how it is. So if one finds a correct account then one will still be able to ask questions of why things behave like that, but there will be no further answer.
Second, physics has revealed some fundamental facts. For one, changes of things have, in a certain sense, no cause, that is, a thing that's moving will keep moving like that, it will reach other places, then other, and then other again, just because it *stays* in its state of motion. It's not that something would be pushing it continuously and causing it's ongoing motion therewith. Also, for a thing to change it's state of motion it needs to interact with another thing and this other thing will change then simultaneously and proportionally in its own state of motion. There is only interaction, no action on one thing by another that is not simultaneously reciprocated. All this can be subsumed under the principle of momentum conservation.
Third, this means that in a certain sense everything is just happening in accordance to certain "rules". Especially the feature of causation that a cause happens or exits before its effect is nowhere to be found at the fundamental or most general level of what is happening in the world.
Now, I'm pretty sure that that means that notions like "Susan got ill, because she drank that dirty water" make only sense because of epistemic limitations.
I mean, she did not get ill *only* because of that. She also got ill, because of her immune system being weaker than that of Li who never gets ill from dirty water. And also because she was born, because if she were not, then she for sure would not have gotten ill.
However, the dirty water is of most interest when one is looking for a cause, because for the near future and most people it's what one can do most about.
But if one considers absolutely everything in Susan's past light cone, and everything that once had future light cones that contribute to what is in her past light cone, then it's hard to speak off causation anymore. Then it again looks like events just happening according to the rules. But that's no problem for me.
The difficult thing for me is to give an ontological account of "the rules" then. At first I tried to say that things have properties and that these properties are interconnected in various way such that if a thing changes in one then it also changes in another. That's easy to see in a piece of metal that one heats up. It then also gains volume and at some point will even melt. That its properties are interconnected like that would then be a property of it too.
But the conservation of momentum and the fact that things "interact" would require it to say that even the properties of multiple things are connected. But if interconnections of properties are properties themselves, of which thing is such a property then a property?
The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them.
The only satisfying idea I have how to fix this lies in considering the possibility that there are not really multiple things. All electrons are excitations of one thing, the electron field, and there is already a working theory of this, likewise with all other fundamental "particles", and then all fields are somehow ... just one single field. There's no working theory of that.Yet.
If that were true, then "rules" like the conservation of momentum could be a description of a property of that one thing, call it the world, so that it's parts behave like they do.
The laws of physics just floating around as anything like commands is deeply unsatisfying to me -- after all, how do rocks know about them -- and so to anyone who feels the same:
"But the conservation of momentum and the fact that things "interact" would require it to say that even the properties of multiple things are connected. But if interconnections of properties are properties themselves, of which thing is such a property then a property?
The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them."
What do you mean by "taken as part of reality and not just the descriptions of reality"?
Continue with the example of conservation of (let's say linear) momentum. In a Newtonian world, this isn't a separate "law" - it follows directly from Newton's third law, that forces are equal and opposite. So which of these two "rules" is the real one: conservation of linear momentum or forces being in equal and opposite pairs? Are they both "part of reality"? What would that mean? Or are they really the "same law" - and if so, why not instead just say that reality exists and these are two ways of describing it?
Of course we don't live in a Newtonian world, but I don't think that matters at the level we're discussing (in any case we don't yet have a complete theory of everything). Even so, modern physics is similar with respect to symmetries and conservation laws. Again sticking with linear momentum: why isn't translation symmetry just a fact about reality? What would it even mean for it to be a separate "part of reality"?
>"The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them."
>What do you mean by "taken as part of reality and not just the descriptions of reality"?
>Continue with the example of conservation of (let's say linear) momentum. In a Newtonian world, this isn't a separate "law" - it follows directly from Newton's third law, that [...]
I distinguish between laws as part of reality and laws as descriptions of reality in this way: if the law of conservation of momentum *follows* from something then it is not a part of reality, at most it is about it. Because conclusions or propositions are what follows from something, not reality.
Often people talk like this about "laws", just like you, and I like to put quotes around it then, because they aren't *pre*scriptions but *de*scriptions.
But I believe that those "laws", because they are only descriptions, do not explain the relevant behavior (and even less cause it), and the behaviour of the relevant things must come from those things themselves. The way they behave is a characteristic of them. It's their way of being. It's how they are. And when one knows how they are then one can make really good guesses (calculations) about what they do next.
And so I conceive of laws, as part of reality, as properties of real things. Which leaves me with the difficulty of to which thing a law-like property belongs that governs the interaction of two. Hence the idea that those two things really are just parts of another thing and the law-like property a property of it.
I'm afraid that distinction might be too abstract for me. I chose conservation of momentum because it was mentioned up thread, but if that is only descriptive, can you give an example that is prescriptive?
And there are no prescriptive “laws” of nature, because the only things that follow prescriptions are things that understand them, not things in general.
There is a difficulty for me, but it's not with finding an example, but conveying my apparently hair-splitting distinction.
But I suspect you know that distinction very well, just not in the context of laws of nature
Here:
The volume of some amount of gas is proportional to its temperature. What I have just asserted is a fact, it is something real, namely a property that any gas has, with a gas being an aggregation of molecules that <insert further qualifications here>. It's a law as part of reality.
"The volume of some amount of gas is proportional to its temperature." is, however, not a fact, but an assertion that the volume of some amount of gas is proportional to its temperature. Its a law as a description of reality.
(On the next analytical level it's a sentence, composed of words, to express that assertion.)
It's the difference between, love being an attitude towards something and "love" being the name for that attitude towards something.
(On the next level "love" it's only a word. That's why it's stupid when someone says with a belief of saying something deep and profound that love is just a word. That's only true in a trivial way and just plain wrong if he means the actual attitude. I have this example from Daniel Dennett.)
It seems that this is describing the "map–territory relation" but then blending them again. I agree with your example's second paragraph. I disagree with the first.
Consider the following substitution: Some configurations of molecules are apples. What I have just asserted is a fact, it is something real, namely a property that such a collection of molecules has, with an apple being an aggregation of molecules that <insert further qualifications here>.
Now, is the form of an apple a "law as part of reality"? I don't believe so. When I say that a certain collection of molecules has the "property of being an apple", that's just a description in my mind. There is no platonic form of an apple which the molecules take on as a property. The molecules don't care what I call them (and the bonded atoms don't care that I call them molecules, and so on).
Similarly, if asked for an ontological account of Charles' law, my answer is that it's one model of reality: one map of the territory. An ideal gas could exist in reality, but Charles' law itself is only "something real" in so far as we can think about it and that our brains are real.
Or from another angle: if Charles' law is indeed "law as part of reality", yet also "conclusions or propositions are what follows from something, not reality"... consider that one could just as easily assert the axioms of kinetic theory to be the true property of an ideal gas. Then Charles' law merely "follows" from kinetic theory in the same way that conservation of linear momentum "follows" from Newton's laws or translation symmetry. So is Charles' law no longer a "law as part of reality"?
The laws of physics don't "float around like commands" and this is what I think is bothering you: You're taking "laws" to be something fundamental on the level that "things" seem to be.
Laws are just descriptions of how a model behaves. Or, in a different sense, you could say that the laws are rules to update a model which, if correct, should have properties matching measurable properties of reality.
However, the laws are related to *the model*, not directly to reality. There's no need to postulate that the laws are "somewhere" or that they "exist" in any meaningful sense.
"Laws are just descriptions of how a model behaves"
No, of how a thing behaves. The "laws" of which you speak *are* the model, or part of it.
As I have just answered plmokn, the "laws" you mean are only descriptions and as such do not explain the relevant behavior (and even less cause it), and the behaviour of the relevant things must come from those things themselves.
The way they behave is a characteristic of them. It's their way of being. It's how they are.
And so I conceive of laws (as part of reality) as interconnections of properties of real things, and with that as properties themselves.
Which leaves me with the difficulty of to which thing such a property belongs that "governs" the interaction of two. Hence the idea that those two things really are just parts of another thing and the (with a "law" referenced) property a property of it.
The laws we can describe are descriptions of models, not of things. There is no law which we know of which we can show applies to a thing in itself and not to our model of it. We can't even prove that the world is lawful, rather than just appearing to be so. Hence, I see no reason to postulate that laws are a part of the world or to explain how they could be so.
Now, supposing the world is somehow lawful, the idea that everything is only a single thing seems to be a decent way to interpret it, but definitely not the only possible one.
This is the (to paraphrase) "nothing but a bunch of math arguement" arguement stated famously at the end of Brief History of Time", not that I disagree, in fact it seems more and more likely. As you say, the math relates to the model, not the actual interrelationship between processes. The idea that there are actual "things" out there or Indeed "rules", now seems somewhat naive.
Years ago, I cofounded a company working on problems like this.
I'm not joking. We were working on a technology that enabled formal models of information, such that one could map any existing database onto it, generate databases from it, integrate multiple data sources using it, and so on. We had a working prototype of the third thing. And we weren't idiots about it - or at least, we were smart enough to avoid the sorts of problems one might see illustrated in an XKCD cartoon. We knew to scope the problem to real-world data systems making real decisions, and to not pretend everything could be expressible in relational algebra (it can't), and to not pretend one could just assume any data schema was authoritative just because it drove a billion-dollar business. Our unspoken motto was that if it drove real business logic, it was important, no matter how hard it was to implement in the RDB or stored procedure framework of the month, and if it couldn't be modeled in first order logic, then it wasn't worth modeling.
One of the things we had to model was causation. We had to, because we needed a model of events, and some events were considered to cause others, whether it be "the next opcode entered the CPU because instruction pointer incremented and the clock ticked" or "Fred killed John because John slept with Fred's wife". For this, we researched various causation models in metaphysics (I don't remember which ones our chief philosopher looked at, but we were accustomed to metaphysics research in this domain), discussing candidates such as "X caused Y iff Y occurred in all possible worlds after any possible world in which X occurred, and after X".
The good news for you here is that one virtually never has to break causation down this far. Nor to stuff like particle/waves propagating or anything like that. For most everyday problems, it's sufficient to define causation arbitrarily, at the level of everyday actions. You brought an umbrella because it was cloudy. A wingnut costs five cents because that's what the store said. If you're having to break the process down to finer detail, it'll be because there's an underlying criterion that naturally informs what will be the best way to lay out the subcauses, and they in turn will also be arbitrary (or empirically available by experiment).
"But if one considers absolutely everything in Susan's past light cone, and everything that once had future light cones that contribute to what is in her past light cone, then it's hard to speak off causation anymore"
Why? !maybe it's a departure from the film idea having a single cause, but maybe that just wrong.
"But if interconnections of properties are properties"
>"But if one considers absolutely everything in Susan's past light cone, and everything that once had future light cones that contribute to what is in her past light cone, then it's hard to speak off causation anymore"
>Why? !maybe it's a departure from the film idea having a single cause, but maybe that just wrong.
I mean not because of that angle, but because none of the little changes of each little thing is thought of being a cause of anything in the relevant physics, but just as going on according to laws.
When I said causation talk about Susan's illness makes only sense because of epistemic limitations, I meant that it *does* make sense, I don't belittle it, but it only makes sense at a point at which one is not looking at all and each of the fundamental interactions of particles.
At that level, to use this picture, the x's in a state-space representing all things with all their properties just slide around in a fancy fashion and with no position of them causing any further positions and with no possibility for them to slide any differently (except for quantum complication).
>"But if interconnections of properties are properties"
>The interconnections are relations.
If they are only relations, like this earring's temperature is now double than before while its volume is now 1.03 times greater than before, then we have only "laws" and that's what I find unsatisfying.
How does the "law" make things behave like that?
In the case of temperature and volume we found the mechanism and it is one that belongs to every piece of metal. It's a property of metal things to behave like that.
Hmm there is a hierarchical nature to our understanding of the world. We're not smart enough to grok the whole thing from just physics. So our path to understanding is to find new 'laws' at new levels... it works OK, but... Have you read "More is Different" by P.W. Anderson? https://www.tkm.kit.edu/downloads/TKM1_2011_more_is_different_PWA.pdf
I will. I just skimmed a recent article about it, and coming to the point of conservation of momentum being a symmetry thing, I'm strengthened in the each thing being a part of the same thing idea.
When all things (except the world) are parts of the same thing, the world, then every behavior of them is really a behavior of it.
Things don't need to manage to mirror (quite literally) their behavior in an interaction as described in a textbook or demanded by a "law", because it's really just one thing acting.
To formulate it provisionally: things are more parts of the world than the world consists of those things.
And that would also open a way to explain a system having properties which are hard to explain by the properties of its constituents. It would be a property of the world that it only shows under that circumstance or only shows along with that circumstance.
The idea seems to be, every thing is like it is and does what it does because there's an immaterial blueprint of it that says it should. Or, a form the things want to take on.
That's not much different than thinking of laws of nature as commands to be followed.
It’s something that’s very hard to explain well without just reading a bunch (or all) of Plato along with commentary.
In essence it’s a theory that posits the reality of things, as we describe them. When we say “that’s a tsunami” we have identified something real with the word “tsunami” even if the atoms composing it at its start are all different from those at its end. It doesn’t claim the existence of these things in the material world as we experience it, as everything is imperfect in some way. We should really always be saying, “that resembles the form of a tsunami.”
The justification for it isn’t exactly empirical, but essentially our use of language (and perhaps thinking in general) requires the existence of forms (or concepts) that are permanent, singular, indivisible, and in a fundamental sense, more real than reality. If these forms are necessary to talk about anything as Plato generally claims, then the fact we can successfully talk about anything at all (which we obviously do) is evidence of their existence.
This also serves to explain what “causes” things to be, if the world is purely a deterministic dance where cause and effect are mutual.
Whether this is a tool for thinking, or truth, or something in between is for the individual to decide. Plato also emphasizes the important of finding the truth through one’s own thinking rather than receiving the knowledge from others.
First: I am not an ENS so the subtleties of your use of ontolgial definitions of 'things' and 'properties' might be lost on me.
I think you could make some progress by thinking about properties and things as entirely arbitrary classifications we came up with through shared language. It might seem obvious, that parts of a tree are its stem, leaves and roots, but the dirt surounding its roots and the birds in its crown are not part of the tree. Hower, we have made these distinctions through assigning names to specific subgroups of atoms because such descriptions are usefull to us in a practical sense, not because they reflect a ground truth description of reality.
Les intuitively, the properties of such subclassifications or 'things' work in the same way.
Our classification system through naming things forms layers, and properties of certain things exist only on a given layer of description. The higher level word for 'The forest' includes all the things described above and a multitude of other things in the lower subcategory layer. We can assign properties to the forest such as 'it's dense', but this property seizes to exist on the sub-layer of individual trees. All you can say at this level is: "well this one tree is closer to the other tree than this tree is to the third tree", but 'denseness' is a higer level description (and also requires for other forests to exist as its a relative decription which is true for most everyday properties).
Similarly saying: 'look here is a piece of metal and it is hard, but if it's temperature rises above a certain threshold it becomes soft' is a useful description of reality if you are talking to a blacksmith or an engineer. 'piece of metal' differentiates a set of atoms from the surrounding environment and describing properties like hardness and temperature describe something happening a the scale-layer of everyday human objects. But if you are a physicist and start looking for these properties on a deeper layer, they loose all meaning. You now need to use words like 'atoms' and 'bonds' and 'kinetic energy' to describe your observations. While you can establish an explanatory relationship from the 'kinetic energy' of individual atoms to the higher layer property 'temperature', and observe that the bond between individual atoms breaks if one of the adjacent atoms wiggles too fast, the property of 'temperature' holds no meaning on this layer.
And if you are so inclined and become a particle physicist you can start defining 'things' on an even deeper layer. You start talking about 'protons' and 'electrons' and suddenly the properties of 'wiggling speed' and 'bond strength' loose their meaning.
And of course for the very daring there is even one further layer if you participate in quantum phsics, but on this layer our agreement on definitions for 'things' is still vague and thus explanatory power for higher layer properties limited.
So in a sense you are right, the properties of electrons do not exist in the electrons themselves, but in their interaction (see atom bonds), and if you are a physicist and it seems usefull to you, you could define an 'electron field' as an entity that simply holds the properties of electron interaction. This is a usefull definition for Physicists because it simplifies talking about math ("this complicated formula describes electron interaction so we say it describes an electron field so we can categorize our various formula better"), but the main takeaway here should be that subdividing reality is up to us and not hard coded in anything and at deeper layer descriptions higher level properties are simply undefined.
In this sense the difference between describing electrons as particles moving through an electron field or describing electrons as exitations of a singular thing is the same as the difference between describing a bird flying through a forest as seperate from the forest versus talkig about 'the forest' as a singular thing; trees, birds, dirt and all.
All the level or paradigm switching you are talking about you are talking about inside the same ontological system.
The subdividing of reality (which I think cannot be made successful if done just arbitrarily) is only a subdividing of things into other things.
What each way of looking at reality keeps in common is that there are things and "laws" they obey.
And what bothers me for a reason I'm not sure about are the "laws".
If I can conceptualize things to change according to their own properties then I'm fine. But if I cannot, as in case of two things affecting each other -- you know, changing in an absolutely synchronized way, somehow "aware" of each other -- than I have this stray property, the "law" governing what they do, which seems supernatural to me.
> "The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them."
Im not sure, that "the laws" exist as anything other than part of a textbook (in the widest sense as something we came up with to subivide and describe our shared experience of reality)
Just as the temperature of a piece of metal or the denseness of a forest is a a property which originates in a different layer of our descriptions, so are the fundamental forces that ultimately make up the textbooks Laws of Nature emergent from things interacting at a deeper layer.
Physics describes the 4 "fundamental" forces: gravity, electromagnetism, weak and strong force. They are fundamental because their nomenclators realy thought they hit ground truth in their description of reality. Newton described gravity as (in combination called classical theory) a central force without a communicating medium. But Maxwell, in defining electromagnetism, already talked about an abstract "field" communicating the force, or in our terms: the field contains the property of charges attracting and repelling each other.
Together these two form the Classical Theory and for a time Physicists believed they where almost done with this process of dividing up reality as the atom was the indivisible smallest thing, and only some confusion remained on where the forces came from, or what this "field" exactly was.
Well and then the Atom was split, first in theory, then in laboratory and we had to go a layer deeper in our descriptions. Since then we have come up with some usefull classifications and a lot of confusion, dug even one layer deeper, and gave names to things which produce these forces through some kind of interaction.
So if your question is: "if the laws of nature dictate, that two objects are pulled together, then where does that law reside if not in either of the objects?" Then the Physicist answer must be: "well we call it The Law of Gravity and its seems to be a phenomenon somehow related to something on a deeper layer; atoms seems to interact with a "field" called spacetime somehow and that interaction is mediated through something we aren't too sure about, but we called it Higgs boson."
So in short: these fundamental laws aren't fundamental after all and they originate at a deeper layer of reality we can't observe well enough to make satisfying classifications and explanations.
Sure, you can consider fields as separate things that interact with each other, or as one complicated thing that evolves according to internal rules. But the problem (if you want to call it that) is that what is or is not considered a "thing" is just a human judgment without any physical relevance.
At the end of the day the whole universe could also be seen as a single "state" that evolves from one instant to the next, but even then, you could say that the totality of states at every point in time, taken together, is actually the fundamental single thing, and slicing it up over time is just a human judgment once more.
"But the problem (if you want to call it that) is that what is or is not considered a "thing" is just a human judgment without any physical relevance."
Only kinda. I cannot *honestly* say that I believe there are no mind independent things. It might be worth treating a tsunami wave as a thing coming your way, even though it is *not*. For most of its existence it's just a row of changes in height of the water it consist in from moment to moment. The water it consists of at its creation is not the same as the one at its annihilation.
And the same goes for the supposed particles of the supposed fields.
But, at the end of such reasoning there always remains something, and that is a thing. The fields are things.
But I agree that judgements about that can be wrong. Maybe the fields are not really things either and even less the fundamental ones then.
"At the end of the day the whole universe could also be seen as a single "state" that evolves [...]"
It can be said, but not seen/thought-of to be *just* a state coherently. Because states are states of things, a state is the whole of the properties of a thing.
I cannot do without the concept of thing, and I always find on deep looking that no one else can.
Real thing: something that just exists, with properties, in a certain state, and is changing, into another state.
"but even then, you could say that the totality of states at every point in time, taken together, is actually the fundamental single thing, and slicing it up over time is just a human judgment once more."
Slicing it up, when? Judging, when? How long does that take?
If humans miss-perceive or miss-construe reality, then there is a succession of states of at least themselves. Their miss-perceiving or miss-construing proves that, or at least one is speaking self-refuting by asserting they miss-construe it.
"I cannot *honestly* say that I believe there are no mind independent things."
The question of whether things exist independent of minds is a different question than the one about how whatever exists should be divvied up into discrete "things", and whether that specific division has some fundamental kind of existence in nature, beyond the existence of whatever the things are made of.
"But, at the end of such reasoning there always remains something, and that is a thing. The fields are things."
To me it sounds like you are saying that once one has described what a field is, how it behaves, and how it interacts with other fields, one has not given an exhaustive description of the field. Instead, you think we should add an additional fact to our description of the field, namely that it is a "thing". The problem for me is that I cannot imagine what role this additional fact would play in nature or physics. The fact that a field is a thing does not play any causal role, it doesn't change anything, so why is it a necessary part of the complete description of the field, according to you?
"I cannot do without the concept of thing, and I always find on deep looking that no one else can."
There are many things I cannot do without, but that doesn't mean they are all a fundamental part of the universe. Even if I grant that I cannot do without the concept of thing, still the way I draw the boundaries between them can be different every time. Therefore the boundaries between things are (as far as nature is concerned) kind of arbitrary, and not fundamental.
"Slicing it up, when? Judging, when? How long does that take?"
Now, always. How long does "now" take? How long does "always" take?
"It can be said, but not seen/thought-of to be *just* a state coherently. Because states are states of things, a state is the whole of the properties of a thing."
In fact, we cannot prove that the fields themselves are fundamental. They are a model that happens to work well enough for everything we've tested them.
We have no reason to believe the fields are "things" other than we have no better theory.
I agree with your worries, sometimes I just wonder if it’s Turtles all the way down. It seems true that human reason is insufficient to answer base ontological questions, but is that insufficiency due to the former or latter - we don’t have enough of reason, or is reason not enough?
On how do rocks know - this relates to free will and I heard this cool answer that this point is actually a misapplication of language than a real problem. By natural law or rule we do not mean the same as legal rule - something compelled - by natural law we simply are saying what happens. At the same time that sounds like trickery to me and not satisfying.
The whole oneness thing u reach makes a lot of logical sense, I never really got the oneness thing until u said that
I’ll say why I’m not thinking about these things more - I think what we have is what we got, such that our objective laws and reason is objectively what we have; they’re Truth. They are true - (for us) permanent and unchanging, all that we can have, what we ought to have. (faith for the ought)
So I’m not too interested in explanations of oneness or the illusion of time, higher dimensional truth, etc. Ants debating the weather.
Super excited if superintelligence will surpass that truth and if it can show any of it to us.
You know, I don't want to know every fact about the world, but I would like to grasp the basics. And right now in a formulation of the basics I have "things behave as if they were following rules". "As if". Methapher. Aaargh!
But maybe I should settle with this. No one else seems to to better, so why of all should I :-D
Please Shinji, you've got to pilot the EVA and settle this once and for all... for your own sake. Find out why you came here; why you exist at all! Answer your own questions. And when you've found your answers, come back to me. I'll be waiting for you.
Isn't it more along the lines of "Things behave in certain ways. Those ways are repeated. That makes them predictable. From observation of these behaviours over time, we derive a set of predictions about how those things will behave in future. If we do this right, we are able to create a set of rules that we then call 'laws' of nature about how things behave, are likely to behave, and have behaved, and we check new observations of new behaviours/new things against those laws. If it all holds together, we've got science".
"things behave as if they were following [our] rules" sounds pretty good to me.
Our understanding can be only EITHER
1. "direct" experience (which is merely direct experience of the near-real-time simulation that our senses present to the brain (little more than a bare-bones summary of reality),
OR
2. metaphor where we compare our perception to mathematics or billiard balls or planetary systems or whatever. Nested metaphors all the way down to direct experience.
Science is pragmatic. If you can predict how a system behaves and you can affect its action then you can control it. If you can control it, it becomes technology and you end up with the iPhone.
If you want absolute truth, that is the domain of religion, but they universally fail by scientific standards.
So what would a deeper explanation look like? Are you asking for a theory that ends with "How could it be otherwise?"
I'm asking for a (ontological) theory that works and isn't dualistic.
Right now we have matter, that is, things, whether particles, radiation, fields, meteors or supermarket chains, on the one hand, and "laws", on the other.
Similar situation on the consciousness front. There are things and some of it have the "content" in their brain, that they holding a red apple in their hands.
Maybe I'm just neurotic. But it bothers me so much.
Laws are just descriptions. They are not supposed to be fundamental characteristics of the world, just statements about how we think the world behaves.
1. it's a feature of language that we have nouns (objects), verbs (actions), and adjectives/adverbs (properties) so three things. This forces us to describe reality in those terms and, pragmatically, this has worked pretty well.
2. Our mind is at a distance from perceived reality so our perception is always us and it.
Does anyone have a link to a 'as much as is knowable' about how to reduce chance of having difficult pregnancies, or identifying risk factors?
Possibly relevant details: age=30, occasionally quite difficult periods, very small hips, prone to mood swings. Open to alternative medicines whether or not they work in theory if they work in practice. Guessing that 60% of risk factor is psychological 'pill that genuinely makes babies super cute and desirable would be nice'
It is possible to considerably improve the healthspan, lifespan, future well being and IQ of neonates through optimizing pregnancy nutrition and cytoprotection. Actually there is a critical period in pregnancy that in in vivo models alter as much lifespan as state of the art lifelong supplementation of geroprotectors.
Not only is it likely that the future lifespan record will be via an optimized pregnancy, but even more likely the future IQ and RQ records will be driven by the optimization of such critical periods.
IIRC neural tube defects can be prevented close to 100% by concomittant administration of folate with inositol.
DHA and choline dietary levels (though max beneficial and safe dose to determine) have probably extreme consequences in brain organogenesis.
Many nutraceuticals can significantly improve nephron number, etc
It finds strong evidence for the efficacy of a dopamine D2-receptor antagonist (metoclopramide hydrochloride - brand name 'Reglan') at improving overall feelings of wellbeing, and reducing nausea and vomiting.
It finds that ginger (tablets or food) shows some effectiveness in improving overall feelings of wellbeing and reducing nausea. In the study's conclusions, they recommend ginger.
Recommended doses are about 1.5 grams a day.
It discusses a number of additional interventions which I can summarize, if you like.
Sorry let me clarify. I'm not concerned with the health of the baby, but the experience of the woman of the pregnancy itself, minimizing negative experiences during and physical consequences thereafter.
I asked GPT4 for predictors of unpleasant pregnancy symptoms, specifying that I was not asking for predictors of a damaged baby or of dangerous conditions a pregnant woman might develop -- I only wanted to know what made it likelier that the woman felt awful during pregnancy. I did not check its responses, but have found that GPT4 is usually reliable for questions of this sort. Here is what it said:
----------
- Individual Susceptibility
History of Motion Sickness or Migraines: People prone to nausea in other contexts (e.g., car rides, hormonal migraines) are more likely to experience severe pregnancy nausea.
Sensitivity to Smells (Hyperosmia): Heightened sense of smell during pregnancy, which varies between individuals, can worsen nausea.
Pre-Pregnancy Mental Health: A history of depression or anxiety increases the risk of prenatal depression and fatigue.
Pre-Pregnancy Fatigue or Sleep Issues: Sleep disturbances before pregnancy can predict higher exhaustion levels during pregnancy.
--------------
-Previous Pregnancies
Severe Symptoms in a Prior Pregnancy: If nausea, vomiting, or fatigue were prominent in a previous pregnancy, they are more likely to occur again.
Multiparity: Carrying multiple pregnancies previously can increase feelings of physical discomfort due to weaker abdominal muscles or ligament laxity.
-------------------
-Multiple Pregnancies (Twins/Triplets)
Carrying multiples can increase symptoms like nausea, fatigue, and discomfort due to higher hormone levels and a greater physical toll on the body.
---------------
-Lower BMI: Underweight individuals are more prone to severe nausea and vomiting (hyperemesis gravidarum).
Higher BMI: Being overweight before pregnancy is associated with greater fatigue, joint pain, and risk of developing sleep apnea, which worsens exhaustion.
--------
-Lifestyle and Daily Stress
High Stress Levels: Chronic stress can exacerbate symptoms like fatigue, nausea, headaches, and even prenatal depression.
Lack of Social or Emotional Support: Feeling isolated or unsupported can predict worsened mood and overall well-being.
Poor Diet or Nutrient Deficiencies: Low iron levels can worsen fatigue, and inadequate intake of vitamins like B6 may worsen nausea.
----------------
-Women who were physically active before pregnancy tend to report fewer symptoms of fatigue, back pain, and mood swings.
Lack of movement can contribute to stiffness, muscle fatigue, and sluggish digestion.
---------------------
-Sleep Quality
Sleep disturbances during pregnancy are common (e.g., frequent urination, discomfort, vivid dreams). Poor sleep quality in early pregnancy is often linked to worsening fatigue, mood swings, and irritability.
--------------
-Age and Health Factors
Maternal Age: Older pregnant individuals (35+) may experience more fatigue, aches, and joint pain.
Underlying Health Conditions: Conditions like anemia, hypothyroidism, or diabetes can increase symptoms like exhaustion and nausea.
---------
-Pregnancy-Related Discomforts and Conditions
Constipation and Bloating: Often caused by progesterone slowing digestion. Affected by diet, hydration, and physical activity.
Round Ligament Pain and Pelvic Pressure: More likely in later pregnancies and in women with poor abdominal muscle tone.
Heartburn: Worsens with certain foods and becomes more prominent in the third trimester.
----------------
- Genetics
Nausea and vomiting severity can run in families. If close relatives experienced significant morning sickness, the likelihood increases.
Summary of Key Predictors:
High HCG and estrogen levels (e.g., early pregnancy or twins).
Personal sensitivity (to smells, motion sickness, migraines).
Mental health history (depression, anxiety).
Previous severe pregnancy symptoms.
Stress and lack of support.
Low fitness or poor sleep.
If symptoms are significantly affecting quality of life, there are often safe, doctor-approved strategies to mitigate discomfort (e.g., dietary changes, supplements like B6, physical activity, or mental health support).
I've never come across any information on how to prevent bad pregnancy symptoms, but for the most common one, morning sickness, Diclegis and zofran are considered to be relatively safe and are also quite (although not universally) effective at treating the nausea.
In a failing attempt to understand why Luigi Mangione can't be tidily labeled on the political binary, we have mainstream news articles quoting Scott, like this one:
In it, the writers reference online speculation that Magione is Grey Tribe, attribute the phrase to Scott, and then quote him, with a link to the relevant post. (And, I'm pleased to report, refer to him by his pen name!)
Weirdly, I feel pretty okay about this? The think pieces trying to explain Grey Tribe / independent ideology to the binary masses don't feel as antagonistic as one would expect.
You've got it. It's that whole friend/enemy, ingroup/outgroup distinction. Now that he's done something they like, they can't go all-in on attacking Grey Tribe ideology. They'll try again later; a lot of it's that whole Silicon Valley-MSM rivalry that doesn't go away.
I'd point out Grey Tribe is a very specific 'liberaltarian-ish' group of ideas ranging roughly from old-school liberal to libertarian: pro-progress, pro-technology, pro-free speech, usually pro-immigration, 'socially liberal' in the 1990 sense of 'gay is OK' and 2010 sense of 'trans is OK' but not in the 2020 sense of 'no member of a marginalized group's feelings must ever be hurt'. It's adjacent to the more old-school-liberal types, but not really the same thing as, say, the remaining IDW people--Grey Tribe isn't nearly as anti-trans, for instance. They believe in biological sex, but if you want to change your body so it doesn't match that, go right ahead--it's the T in TESCREAL!
Non-binary ideologies that don't clearly map on a left-right divide are more various--I can think of Catholic social thought/Chestertonian distributism or generic populism on the other side of the political compass, but you can imagine others. A lot of the neocons are politically homeless now, for instance--look at the Cheneys stumping for the Democrats. Richard Hanania has more or less turned himself into one and you can see how much he hates both sides.
Theoretically political views are a space in R^n, where n is the number of issues, and we map that 'spectrum' by analogy with ROYGBIV onto a left-right R^1. There's a lot of information loss; even R^2 doesn't really make sense since it's not clear whether speech codes are 'socially liberal' or not. (Even the terminology gets difficult as 'liberal' means 'free-market' in Europe.)
Arguably the two political parties basically try to do a support vector machine (modeling and machine learning people PLEASE tell me if I've got this right and if you have a better description please say), drawing borders to include the largest clumps until they get to 51%. This doesn't really produce coherent ideologies or good governance, though Europe and East Asia have their own problems.
It's not the least surprising to me. Cheney is from Wyoming, where the GOP tends to be extremist in their anti-environmentalism. To put it as plainly as possible: someone like Cheney doesn't believe there is anything worth conserving. The world just looks like money to them. As such - and the principle way such people can express their hostility to other values, to values of any kind - is to be for open borders. Again, because there is nothing to conserve, whether culturally or in terms of flora, fauna, and the landscape. Everything and everyone - is fungible with money - and they have made a great deal of money while disregarding the fate of the country, to which they are ultimately indifferent.
Of course their cause would lie with that of humanist progressives. I am not even suggesting they are cynical about it. Someone like GWB doesn't seem like a man with the capacity for cynicism. That's probably one thing that made him seem likeable.
Even Bernie Sanders understood this.
Such people of course tend not to choose to live amid ugliness they celebrate, but we're to imagine that's coincidental.
There's a principle I have that goes something like "If you're trying to explain someone else's behavior by claiming, 'Despite their stated reasoning their true motivation is actually $secret_motivation_x,' then $secret_motivation_x better be something that you yourself can actually see the appeal in and imagine being motivated by. Otherwise you're probably just making shit up to justify thinking your opponents are evil."
That's true. I don't view it so much as a universal rule, more as a useful warning that you should be extra suspicious of this sort of line of thinking. Sometimes after applying that extra suspicion you'll conclude that, no it really is right in this case, just as there are many other situations where you need to be on guard against motivated or self-serving reasoning, but sometimes the self-serving conclusion really is correct.
Money is something to value, that’s true enough. But evil - I don’t know if I can agree with you that humanism is evil, or a perversion. But you said it, and you do have that principle guiding you :-).
I like this test, and also noticed something was off reading the above comment, though I couldn't put my finger on it.
Of the reasons for Cheneys to defect from the GOP, "Wyoming conservatives want open borders to destroy the world" doesn't strike me as within the top 10.
I see that folks here aren’t real familiar with western GOPers and the history of conservation in this country lol. That’s perhaps not surprising.
I’m not sure why what people accept about the most famous GOPers of Kansas - is strange to them about Wyoming. Or indeed of Reagan, whom I’ve not heard described as “evil”.
And of course, Mike Lee didn’t achieve the power he holds, to halt conservation, by accident.
But admittedly, this is a big subject - other bigger than, uh, talking about talking, the preferred mode on much of Substack.
Everybody says antidepressants take weeks or months to have an effect. This does not match my experience at all. I've found antidepressants of various kinds to have a strong positive effect on my mood almost immediately. That effect fades soon, and after a month I'm not sure they're doing anything at all. What's going on?
This has made me pretty reluctant to increase dosages multiple times, since I've expected that to only lead to a temporary boost. Should I instead assume that while some of the initial effect fades, some of it remains, and eventually after a few rounds the dose might be appropriate and the drug would provide a longer-term benefit?
I had the same experience as you a few years ago. For both antidepressants I tried, I had near instant effect, and then the effect was gone in a few weeks.
Was wary of increasing dosage indefinitely and so quit them after a period of having side effects with no mood benefit.
I have no good answer to this, but I've heard this from several people.
You're probably responding to the actual boost in serotonin, which is fast but quickly adjusted for. Most people seem to respond to some downstream effect of the adjustment to higher levels of serotonin.
>> "It is clinically very obvious that a subgroup of depressed and anxious patients demonstrates a substantial and sustained response to antidepressants. This group appears to be the one in which positive effects on emotional processing, cognitive flexibility, and neuroticism are all experienced, and they are experienced in a mutually-enhancing manner. Because emotional processing changes happen quickly within days, these patients often start noticing signs of improvement in their emotional outlook within days. Patients whose depression is characterized by greater psychomotor and cognitive symptoms and low baseline neuroticism rarely demonstrate a quick response to antidepressants in my clinical experience. Response to antidepressants is often painfully slow, perhaps because neuroplastic changes have to slowly produce therapeutic effects without a synergistic boost from emotional processing shifts or improvements in neuroticism."
...Sometimes I wonder if the people saying that about anti-depressants actually have the kind of depression that's treatable by those medications. If you have low mood that's due to normal causes, it's going to get better over time because mood naturally has it's ups and downs. That would continue to be the case even after you took the medication (unless the medication has obvious side effects); you don't even need placebo.
For me, anti-depressants definitely work. Last time I stopped taking them, after a few months I completely lost the will to live. Not as in I wanted to kill myself, but as in my body literally stopped trying to sustain itself. I completely lost my appetite and my will to do anything, and my body was barely generating any heat. I was legitimately going to die if I didn't get back on my meds. Of course, once I did, my condition immediately started improving.
So, uh... my prediction is that if you're no longer feeling like that, or never felt like that in the first place, I doubt you're going to see much improvement from increasing your dose. Of course, I'm not a doctor, so please just discuss this with your psychologist. Also, if you're living alone, don't taper off of your medications without telling anyone, because you might die.
Did anybody teste the newest batch of ergonomic pillows. There is an aggressive marketing on my feed of this new version that looks more like exotic geometric figure than like a pillow. I am intrigued by the design and there is a claim that it helps side sleepers so I am thinking of getting one. However I have loads of different unsuccessful products in this area and I am not interested in another regular pillow, so I would like to hear if anybody had positive experience with it and should I risk. Otherwise I might try a quite more expensive solution that is a combination of snake pillow for pregnant women and some kind of platform to have your bottom arm and shoulder less impacted by the weight
Just like Scott I bought something like the first pillow once
I always ended up waking up in a position which defied their shape. You have all these images of people sleeping in a perfectly "ergonomic" position and it might work if you sleep like you're dead and don't move at all, because you can certainly lie down on one of them in a way that feels "correct". But if you move at all while asleep, you will probably not stay in that position. With ordinary feather pillows or synthetic equivalents you just mold the pillow as you go.
I guess it's worth mentioning that I also prefer harder mattresses and I really don't get the "memory foam". It feels too floaty and unstable as a support for the head or the body and I usually wake up with sore muscles when sleeping on something like that (I sleep better even when hiking and sleeping basically just on a patch of grass).
I sleep on my side, mostly the right side, and have a sore right shoulder a lot of the time because of it. I've tried lots of piling-up-pillow things, and am actively fiddling around with making a home-made version of what you're looking for. Here is what I am sure of so far: What you use has to be a negative version of what you will be supporting. So imagine lying down on your side in some soft molding material, with your arm in whatever position you like it in when you sleep: stretched straight forward in front of you; or upper arm straight forward and forearm pointed up at a 90 degree angle; or arm bent so that you hand is under head pillow, and upper arm is parallel to your body and slightly in front of it. You sink down into the mold until the highest part of lateral you, the side of your cheek, touches the molding material. Ok, when you stand up you will leave a hollow, and that hollow is what you want to duplicate with support you use. You need a support that involves a large hollow on the surtace of the bed — a hollow into which your lower shoulder and arm fit. In brief: You want innies that match the outies.
I have not tried it, but I’m pretty sure it actually would work.
I’m trying is to improvise something similar for the bed I have. I’ll order a wedge-shaped piece of very firm foam for the upper half of the mattress, and carve a hollow in it that is innie version of my downward-jutting shoulder, and my arm pointing forward from my body.
So looking at the things you have in mind: I’m sure the first one would not work. It’s really just a firm version of an ordinary pillow. It lifts the head a lot, but not enough to keep the woman’s shoulder from twisting forward of her body and taking all the weight, and that’s the thing that gives you sore shoulders. It is not possible to take weight off the shoulder just by lifting the head. In theory you can, but in practice it is very uncomfortable. You need to lift the head *and* make a hole for the shoulder. I think the second one you link would probably work, and now that I think of it maybe I should just buy one.
Later edit: Amazon is selling the Medcline one for about $250. It gets lost of pos reviews, but some neg ones that sound pretty valid. The most serious complaint is that the thing is made of soft memory foam, and squishes down so much when you use it that it isn't supporting your head and sparing your shoulder as it's supposed to. People also say it permanently loses some of its shape and height pretty fast. So actually I think I'll stick with my first plan and make something similar out of very firm foam, possibly with something even firmer inside to help it hold its shape.
I waited to respond because I already had a chiro appointment scheduled and wanted to see what he found. I too wake up with my right shoulder sore, and it's gotten to the point where it became a real problem. This is confounded by heavy use in grappling.
Long story short, the shoulder was out of its socket, and developed a nasty case of bursitis. Out of the gym for me for awhile. But - he reset the shoulder and last night it already made a big difference.
Where I'm going with this is it may be good for you to have that shoulder checked - if it's out of socket no clever pillows or support structures will really help until it's back in its place.
Wouldn't having it out of its socket be very painful? How bad was your shoulder hurting? The pain in my shoulder is pretty mild. I only notice it if I stretch my in certain ways -- straight overhead is the worst -- and even that is bearable, & I can do it if I really need to reach something. I know someone who used to have a shoulder that dislocated spontaneously, and he said that was the worst pain he'd ever felt.
The shoulder wasn't too far out. All it took to get it back was a mild click. Followed by an immediate sense of relief, like there was this tension inside that got released.
My morning pain was pretty severe, a burning sensation that I'd rate 6/10. This, I believe, came from bursitis. That inflammation is still there of course, will take weeks to clear out, but this morning the pain was less bad already.
The thing is, I suspect my shoulder was out for awhile, and it just got aggravated in the last 2...3 weeks. Before that I would wake up with a mild but annoying pain, maybe 3/10, and it would kind of gradually ease during the day - unless I stretched, especially overhead (yes, exactly as you describe)! So I pretty much stopped doing overhead presses, for example.
I am thin and wide, a bad build for shoulder health. It's not the first time this shoulder has gone out of place.
By the way, following up Scott's couple of posts on architectural tastes, the acclaimed new movie "The Brutalist" debuts on Friday. Those who read Scott's review of Tom Wolfe's "From Bauhaus to Our House" might have some doubts about the historical accuracy of this fictional movie about how an impoverished immigrant Bauhaus architect (Adrien Brody) is pushed around by a self-assured American capitalist (Guy Pearce) in the 1940s and 1950s.
I think it's a funny coincidence that after Francis Ford Coppola spent decades spent trying to make "Megalopolis", it would be released the same year that someone else released a cheaper & better received movie about an architect.
I haven't seen The Brutalist yet, but I've come across some reviews that say the one black character seems shoehorned in to establish that the protagonist is a good guy for being friends with him.
I've got some weird conspiracy theory about how the Democratic Party leadership is going back to making movies playing up Jewish-WASP antagonism to corral Jews back into the Dems after a revolt cost them some Ivy League presidencies...but I doubt these things are that well organized.
I think the British film industry might be more affected by PC views of diversity. Period settings are their bread & butter, and Britain really was overwhelmingly white until relatively recent immigration. I noticed this recently in the film "Joy", where there were two notable non-white characters, in which the black woman was a white woman in real life that they changed, and the South Asian man was an entirely fictional invention. They also changed the protagonist to fit better with socially liberal attitudes than she was in real life, and apparently never bothered to contact the one surviving member of the team to see if they were portraying her accurately.
Holy crap. Is it really in the aims of making characters 'relatable to a modern audience' that the female main character has to be sleeping around?
I'm biting my tongue here because there's a lot I could say but it would be more heat than light, but sweet sassy molassey. Even for a 'period piece' (and now the 60s/70s are period pieces?), the characters have to behave like 21st century people. No, you can't have a woman who was Christian and didn't believe in sex before marriage be the heroine! In order to show that she is bold and brave and challenging the stuffy old orthodoxy, she has to be liberated! It cannot possibly be that someone with socially conservative attitudes also works on ground-breaking new research!
The review sounds like one of those online bingo card quizzes: check off all the squares! We got "Christians bad", we got "POC characters", we got "sexual liberation", we got "repressive societal attitudes", we got "making stuff up for dramatic purposes", we got - of course! - "Strong Independent Woman". Ah, I can't go on, it's all so tiresome.
EDIT: Looking around to find out what this congregation could have been, this is the contrast between what the review says the film does and real life:
"Before joining the research team, Purdy, who lives with her devout mother, is a frequent churchgoer and committed member of her congregation. But as word spreads about the nature of the team’s experiments, Purdy is told she’s not welcome at the church unless she quits her job, and ultimately her mother says she can’t live at home anymore as long as she continues the IVF research. Purdy nevertheless refuses to quit."
Well, first she was living with her father and brother at home as well as her mother, so far as I can make out:
"Purdy lived at the family home, a semi-detached house at 61 Langham Road, off Perne Road, Cambridge with her father (George), her mother (Gladys May) and her elder brother (John; for family tree see Figure 6)."
Second, again so far as I can find out, the Grantchester church she belonged to is Church of England. Maybe this is a very conservative church, or it was back in the late 60s, but whatever the attitude I can't see an Anglican clergyman telling a congregant "You cannot attend here anymore!" in that way. It's just not *Anglican* to be that direct:
"Jean is buried beside her mother and grandmother in the Church of St. Andrew & St. Mary, Grantchester, just outside Cambridge; where Louise Brown, the world’s first test tube baby, unveiled a memorial to Jean on 20 July 2018."
As for official Anglican attitudes, I can't find anything in the resolutions of Lambeth Conferences in 1958, 1968 and 1978 about IVF. The CoE has broadly accepted IVF (unlike us Papists) and issued a report about it in 1983
She was very young when she died, only 39, of malignant melanoma.
It's interesting you mention the stigmatization of non-promiscuous women. I never thought of that angle, but it's a really good angle, and a lot of women are very much not into sleeping around. It's odd (or perhaps it isn't) how removing old prejudices tends to create new ones.
I get the funny feeling Chesterton wrote something covering this exactly 100 years ago.
To be fair to the film, after someone at church rebukes her for showing up the actual clergyman disagrees... saying the church always wants people to repent and come back, implying that her work is still unacceptable to them.
If we're going for movie conspiracy theories, possibly more about "See? Immigration is great! Without immigrants, we wouldn't have had this beautiful concrete bunker building!"
Which, uh, may not have the appeal factor they think it does.
Going purely by the description in Steve's comment, I am up for seeing a Brutalist architect get pushed about by a self-assured client who don't want none of that ugly stuff 😁
9 % on Ukrainian victory (unchanged from November 11, 2024).
I define Ukrainian victory as either a) Ukrainian government gaining control of the territory it had not controlled before February 24 without losing any similarly important territory and without conceding that it will stop its attempts to join EU or NATO, b) Ukrainian government getting official ok from Russia to join EU or NATO without conceding any territory and without losing de facto control of any territory it had controlled before February 24 of 2022, or c) return to exact prewar status quo ante.
36 % on compromise solution that both sides might plausibly claim as a victory (down from 38 % on November 11, 2024).
55 % on Ukrainian defeat (up from 53 % on November 11, 2024).
I define Ukrainian defeat as Russia getting what it wants from Ukraine without giving any substantial concessions. Russia wants either a) Ukraine to stop claiming at least some of the territories that were before war claimed by Ukraine but de facto controlled by Russia or its proxies, or b) Russia or its proxies (old or new) to get more Ukrainian territory, de facto recognized by Ukraine in something resembling Minsk ceasefire(s)* or c) some form of guarantee that Ukraine will became neutral, which includes but is not limited to Ukraine not joining NATO. E.g. if Ukraine agrees to stay out of NATO without any other concessions to Russia, but gets mutual defense treaty with Poland and Turkey, that does NOT count as Ukrainian defeat.
Discussion:
Since previous update, there has been continuing moderately alarming Russian advance, especially in the southeastern corner of the front, even though US military aid has not been curtailed in any way.
Also Ukrainian government so far totally failed to publicly articulate any coherent strategy, diplomatic and/or military, in response to US elections.
Their failure to take the initiative increases the likelihood that Ukraine will have to accept conditions dictated over their head by some bargain between “the West” and Russia, which will be probably not good. And I don’t think they have some secret genius-level plan; so far Ukrainian civilian leadership, as opposed to high military command (which has been consistently extremely competent), demonstrated mainly their haplessness, except for one brave decision at the very beginning.
* Minsk ceasefire or ceasefires (first agreement did not work, it was amended by second and since then it worked somewhat better) constituted, among other things, de facto recognition by Ukraine that Russia and its proxies will control some territory claimed by Ukraine for some time. In exchange Russia stopped trying to conquer more Ukrainian territory. Until February 24 of 2022, that is.
I wrote a second substack on Ukraine which might be of interest. It is a direct response to comments in Astral Codex Ten from people who disagreed with me for 3 good reasons:
Based on your definitions, I think "Ukrainian defeat" is the likeliest outcome, and it will be like how the Korean War ended. Both sides will sign a cease-fire, with Crimea and parts of eastern Ukraine under Russian control, but Ukraine not recognizing the legality of it.
I wouldn't bet much on this scenario. Russia would find it very unsatisfactory, because the example of South Korean clearly shows how unconquered part of Ukraine could be provided with forces and weapons in the future, even if it's not a part of NATO (any kind of NATO or NATO-aligned forces presence in Ukraine is simply unacceptable to Russia).
Further, consider that Korean ceasefire was signed after front lines stalled for a long time, with neither side achieving anything significant. For now, Russia keeps steadily if slowly advancing: as long as this advance continues, Russia has little incentive to accept any kind of peace talks aside from complete Ukrainian surrender. Importantly, this advance does not look like some kind of large concentrated push, like last year's Ukrainian counter-offensive: it's just Ukrainian army stretched thin after Kursk (mis)adventure, and Russian forces using this to gain previously well-defended positions.
But what is the alternative? I see three:
1) Likely: after Trump's plan that proposes Korean-like scenario, or anything at all that includes term "peacekeepers" in it, is soundly rejected by both Russia and Ukraine, Trump is outraged at Russia and orders to continue aid to Ukraine. The war drags on for another year or two, with Ukraine still mostly losing ground. It ends when either Ukraine or Russia has a sudden change in leadership (Ukraine is more likely, but all things are possible). The new government quickly signs some deal, based on realities on the ground.
2) Less likely: same as above, but Trump is outraged at Ukraine instead of Russia, and manages to actually cut US aid to Ukraine completely (he will have a hard time convincing the whole Republican party to support him in Congress and Senate in this, but maybe he can do it). The war drags on for maybe a year, with EU trying (and probably failing) to replace US. Then Ukraine gives up and agrees to whatever deal Russia is ready to accept at the time.
3) Very unlikely: Trump's plan is actually somehow acceptable to Russia (I would guess this includes some palpable guarantees of no NATO membership for Ukraine, conquered territories recognized by USA, Ukrainian military is reduced to self-defence forces, Russian language is re-instantiated as the second official language). Ukraine is probably furious, but terrified of losing USA aid, and so accepts, maybe haggling a little (mostly about size of reduced military and defence guarantees - it already has some treaties with EU countries, but will want another with USA).
You might note that all 3 versions end up with Russia getting what it want, and mostly differ in amount of suffering inflicted on both sides. This is because I don't see a better end for Ukraine, unless NATO intervenes directly, which is unlikely. At the same time, USA and EU should be pretty happy to prolong this war for as long as possible, weakening Russia without any loss of life their voters would care about, which makes alternative number 1 most likely, as it guarantees the longest war. The only thing that might change that is if Trump is crazy enough to think he can turn Russia against China, or at least to some kind of neutrality, by throwing Ukraine under the bus (I can see no other reason he might want to end this war, otherwise, unless maybe to win a Nobel Peace Prize, which, admittedly, might be a good enough motivation where Trump is concerned).
I wrote a second substack on Ukraine which might be of interest. It is a direct response to comments in Astral Codex Ten from people who disagreed with me for 3 good reasons:
You are underestimating how succesful Ukraine is at building their own weapons industry. Also overestimating length at which Russia can keep up its current pressure.
Obviously cannot replace the US, but in EU + Ukraine + serious funds (they get like 15bn/year from interest of Russian funds alone) + depleted Russian armor stocks + depleted Russian forex funds + depleted Russian manpower (losing like 4-500k/year at current rate), they will not need the full might of the US supporting them in 2026.
If all Russia can do is throw meat waves at Ukraine, with a total lack in armor support and air support (their bombing run frequency has dropped like 80% in past month) then simply delivering enough artillery shells + FPV drones will be sufficient to wear Russia out at not too great a cost to Ukraine.
15bn/year sounds like a lot, but Ukraine has a 40bn budget deficit. It also has manpower problems, or US won't be recommending it to start drafting 18 years old men (which Ukraine refused to do so far, because this will likely cause a lot of internal discontent). According to a recent article, it also has a problem with a (relatively) high desertion rate. EU also failed to fully reach its goal to supply Ukraine with shells, although admittedly it did a lot better than seemed possible at the beginning of the year.
I would be very wary about losses figures for both sides, as there are no good sources for them. Ukrainian officials crudely under-report them (80K killed since the beginning of the conflict is so transparent a lie it doesn't stand up to any scrutiny), and Russia just doesn't report any figures at all. Based on openly available sources, it seems to me that losses are about equal on both sides - and Ukraine has a smaller population. The myth of meat waves is told about both sides from the beginning of the conflict, but the truth is probably both sides sometimes have to send in troops without armour, though both prefer not to. Ukrainian officers like to talk about overwhelming number of Russian troops to explain the loss of this or that village, but combat footage, I think, doesn't support their assertions. By the 3rd year of war, both sides converged on pretty similar tactics.
I'd like to see your source for 80% drop in frequency of bombing runs, though if it really is there, the most likely explanation is weather, which is pretty shitty in these parts in November and December. Certainly there has been no shortage of missile and long-range drone strikes in past weeks.
Additionally, I think that if US stops aid, EU is soon to follow. The war is growing unpopular there. While AfD is unlikely to win Chancellorship in Germany, its popularity grows, and that usually means that its opponents will begin to accept some of its ideas, to stop their voters from defecting. This is already reflected in tougher immigration rhetoric from other parties, and this might happen to Ukraine, too, because it's a "cheap" topic: it doesn't need new taxes, new limits on freedoms, or any other things that has to be "sold" to a voter.
France is entering its own political crisis, and it's not obvious how it will resolve it, considering that the president and parliament are completely at odds. Hungary and Slovakia will continue their campaign to bring back cheap Russian gas.
Russian meat waves a myth? Come on man. There are thousands of videos out there that show it is very much not a myth. Hard to have a discussion with you if you deny this basic and obvious fact. OSINT estimates of equipment loss ratio sits between 2-6x in Ukraine's favor. So it is laughable to say that loss ratio is 1x for manpower.
Bombing runs adjusted for weather are down ~80%. And Russian Forex reserves are rapidly dwindling (about 1 year left) with inflation already sitting at 15-20%.
Russia will not be able to keep up this intensity for more than 6 months - year. They are gambling that Trump will pressure Ukraine and they can squeeze out a favorable peace deal. If Ukraine holds on, the conflict will continue in low intensity mode after that with Russian leverage rapidly dwindling.
As for meat waves, I believe that BOTH sides use them SOMETIMES when they have no other choice, or when a miscommunication happens. But the tactic is not unique to Russia, neither is it used constantly. That you don't see videos of Ukrainian meat waves only means they are not widely posted on western platforms (quite unsurprising, considering dominance of pro-Ukrainian posters).
> Bombing runs adjusted for weather are down ~80%
I'm really interested to see the source for that now. I did a quick Google search, but couldn't find anything.
> inflation already sitting at 15-20%
That part IS true, but Turkey shows that a country can sustain that rate of inflation for a longer time. Inflation causes some grumbling, but by itself, it does not prevent a country from making war.
> And Russian Forex reserves are rapidly dwindling (about 1 year left)
This part I'm unsure about. News shows Russian gold-and-currency reserves actually growing a little during 2024. National Wealth Fund also grew a little. The economy is growing better than IMF prognosis. All in all, I'm fairly sure that, unless something drastic happens to the oil prices, Russia will probably be able to sustain war effort for more than a year.
All in all, I feel this discussion is going nowhere, as we both have different bubbles of information which are hard to break out from, and no special knowledge available to either of us. Aside from my request for the source of information about bombing runs, which is really interesting to me, I propose we conclude this thread and see what really happens in 2025.
Yes, because it's too late for that. First, there is a matter of what Putin sees as broken promises from the West - he won't trust anything not written in stone, and probably not even that. Second, this is Trump's last term, and his successor is more than likely to try to renege on any agreements made by Trump.
The only way Russia and USA could be allies these days is in a case of direct war threat from China. But this is not likely: China doesn't need to conquer Russia by military force, because currents of policy already push Russia closer to China. All China needs from Russia is cheap resources, and it already has that, because western sanctions drive down the price.
However, I think that Russian top officials are not very satisfied with the way China supported Russia in Ukraine, and so there is a small opportunity for USA to use that dissatisfaction, not to drive any real wedge between the two countries, but to weaken the ties between them a bit. Kind of like Franco's Spain didn't directly join Hitler's war, only supporting Germany a little. But I don't know if it makes any sense from a strategic perspective, because it's not obvious what a "full weight" help from Russia would look in case of US/China armed conflict: while Russian army right now is one of the most skilled forces in the world in modern conflict, Taiwan and South Sea is a very, very different theatre where lessons learned in Ukraine would apply only partially, and Russia just don't have the right forces to participate there, anyway.
> Also Ukrainian government so far totally failed to publicly articulate any coherent strategy, diplomatic and/or military, in response to US elections.
Why would you expect Zelensky to *publicly* articulate their strategy? Besides, you are assuming a *single* strategy - the smart thing for a leader to do is to keep several doors open.
Because he kind of did. I mean, when you present something called "Victory Plan" people assume it to be a plan for victory. It's in the name. Yet, a plan is, usually, a sequence of steps, that, if followed correctly, should bring about the desired result. Instead, "Victory Plan" turned out to be a list of demands to Ukraine's allies. It's not "here's what we'll do", but "here's what we want someone else to do". Worse than that, it's "here's what we want someone else to do, even though we know they won't do that", since the linchpin of this "plan" is NATO membership. But Ukraine don't need NATO membership as a keepsake, you know. It's not like NATO can admit Ukraine, and then Ukraine will go "oh, great, that part of plan is complete, now we will go all Hulk on Russians, because our enthusiasm from being a NATO member infuses us with supernatural energy". It's "C'mon, NATO, please fight Russia for us, possibility of nuclear apocalypse be damned".
Well, OK, this "Victory Plan" was just to reassure the domestic population, so it's kind of unfair to make too much fun of it (no more than of Trump's speeches about "draining the swamp"). But the problem is, no one else came up with any kind of workable plan for Ukraine that I know of that does not involve direct NATO intervention. And since NATO is unwilling to cooperate, it raises the possibility that any kind of victory for Ukraine is impossible.
I am not expecting it, exactly, but I do think it would be a smart thing to do, in order to steer public conversation in both Ukraine and globally in a favorable direction.
2014 Rationality: Beware Isolated Demands For Rigor; touchy about any comparison to religion; exceptionally scrupulous norms around truth-seeking, centrality, consistency, etc.
2024 Rationality: explicitly making a case for selective rigor; "see, we're no worse than your friendly neighborhood religion"; downplaying differences of epistemology for PR / community building reasons.
One can quibble about the details of the positions or the degree to which the tweet is representative, but that's almost irrelevant as a matter of culture. The vibes have shifted starkly.
The core groups that later become Rationalists seem to have put a lot more of their identity in the early 2000s as being explicitly Atheist. Not lower-case, but specifically part of an acknowledged movement. Scott has talked about that a lot over the years, how the Atheists became cringey and lost favor, etc.
When a movement that used to be attached to specifically anti-religion loses the anti part, it can talk about the rest of the movement (EA, Rationality, etc.) using religious or quasi-religious phrasing and thoughts, without losing any part of the core message. Atheism, it turns out, was an add-on that wasn't necessary for the ideology. Though it makes sense given that a lot of Rationality came from a certain background that overlapped heavily with both atheists and Atheists. Atheism, perhaps, was a rally point for the movement to talk to each other.
There's a separate vibe shift that I've been seeing in a lot of Rat-adjacent writing that recognizes missing factors in a stable and enjoyable life. People that don't believe in God but see something of value in a religion or religious community. Surveys showing religious people as generally happier and better adjusted (fewer mental health problems, etc.). Religious people having more children. Whether this will result in Rationalists become religious or quasi-religious remains to be seen, but there's a lot of reasons to be skeptical about it. Perhaps they'll find a way to add ceremony and such to their activities, but it'll be difficult for a lot of their core reasoning to survive very much of that (and uncertain to be effective if that's not what makes religion "work" for their purposes or if the people involved get frustrated by the dissonance). SF Tech "Move fast and break stuff" mentality giving way to community and culture can either redefine the movement or destroy it.
"I don't think there's any ideology that, taken seriously, requires you to be a neurotic wreck. You can always take it seriously and simply not be neurotic about it.
But there are definitely communities and cultures that do a more or less successful job of integrating ideological claims that make a lot of people neurotic (whether that's 'Christ will return' or 'most people go to Hell' or 'everything trades off against saving children's lives' or 'the world is ending soon') with a healthy culture and a vision of the good that makes people stronger and more functional.
You can reject out of hand all claims with terrifying implications or you can try to build communities and intellectual cultures that are functional and non-neurotic and non-self-destructive while taking those claims seriously where they deserve it."
I'm not sure it's really representative of all that, I think; I'm more seeing it as reacting against some of the perceptions of EA as people being obsessed with the welfare of shrimp or hypothetical people a million years in the future.
The alternative is to just refuse to live by all the more challenging claims of your ideology du jour, like Catholics who use contraception or Muslims who drink alcohol. This annoys intellectuals who really believe in the ideology immensely but is what most people actually do.
Not that you're wrong about the vibes; you'd expect something to shift in ten years.
This tweet is something I would not have been surprised by in a less wrong post in 2014, with a thoughtful/helpful discussion about possible tradeoffs but not reflexive pushback.
Everyone seems to believe that Luigi became mentally ill, perhaps became psychotic or schizophrenic before the murder. It bothers me that it's almost tautological that someone living such a normal, successful life in our society who commits such a violent act *must* have become mentally ill merely because we have no model for someone behaving that way who *isn't* mentally ill.
But isn't it at least possible that, like Raskolnikov or Carlos the Jackel, Luigi simply developed a violent philosophy and decided to live by it? His Goodreads review of the Unabomber manifesto in which he notes that violence is a valid method of communication was from 2021, well before the period when he possibly had a mental breakdown.
All we know is that he disconnected from his family and most of his friends. Someone planning a murder might do such a thing for reasons including trying to put themselves in the right frame of mind and making sure they didn't in some way tip off what they were planning.
Yes, but the point is that there is evidence of mental illness apart from the murder itself. It is of course possible that he is not mentally ill, and that said behavior was caused by something else, as you speculate. Nevertheless, it is incorrect to state, as you did, that the claim of mental illness is tautological.
I believe that, for the majority of observers, the claim of mental illness is tautological, meaning that most people in society don't believe that such an act committed by a person with the background of Luigi Mangione was possible without mental illness present. The murder itself is considered sufficient evidence for illness with no room for daylight between.
I think people are just playing the odds. Your explanation is certainly possible but judging from history it's not particularly likely. Wealthy people don't usually martyr themselves. His ideology certainly played a big part but I'd be surprised if he wasn't suffering from acute mania or other serious psychiatric condition.
Disappointing that you don't see the murder of someone you don't know as an irrational act. We as a society need to agree that that is well outside of the norms of acceptable - crazy.
Yeah, but there could be a third category: Sane but evil. That label puts him outside the norms of society without sticking on a label, "psychosis," that we do not have enough info to use. And don't think it's self-evident at all that he had to be psychotic. In case you haven't noticed, people, including well-educated thoughtful sane ones, murder each other all the time. Consider the middle east. And it is really not that hard to inculcate young males into a culture where killing is OK. "This is my rifle this is my gun/ this is for work and this is for fun."
Rationality depends on what you want to achieve with it.
All I see is a bunch of pigs wrestling in the mud. All thinking they have the moral high ground. Murderer who used his pen, got murdered by a murderer who used his gun. Spectators who may have done some questionable things with their pens are outraged that a gun was used.
Pushing through corporate policies that will deny or delay treatment to make more profit and will kill people but is perfectly legal (partially because your corporate lobbyists have helped write the regulation that determines legality), is that well outside of the norms of acceptable? WIll you express an adequate amount of outrage at them too? Or is everything that is legal perfectly moral?
I want to clarify, murder is 100% wrong, but I am amused by the outrage against this particular murder that people are expressing now.
I'm getting sick of the prissy outrage too. I felt a flash of fierce joy when I heard the guy had been shot. Do I really approve of murdering CEO's if their company is unusually greedy and inhumane? No absolutely not. But I doubt my initial private reaction was rare. And now all these people are going around saying "I'm shocked, *shocked*" like the police chief in Casablanca when confronted about the gambling that has been going on in his bar.
Crazy, sure. Evil, definitely. Irrational? No, not if his goal was to be a hero who is loved and supported by millions of people. If that was his goal, he wasn't delusional or irrational.
As I wrote below: It seems relevant that he's been unemployed since at least 2023. Why? He was a very ambitious student. He went to work at a company that developed software to help car shopping, then got laid off, or maybe he quit. It's not clear. Perhaps the corporate world broke his spirit, and he redirected his ambition in a completely different direction.
Continuing from our previous conversation: I think you massively underestimate how much it takes to get psychologically "normal" people to actually make the decision to kill people. Grossman's "On Killing"* describes at length the process the US military has to take train its members-somewhat by definition mostly psychological normal-to kill other human beings, and it's not clear that it works all the time. Thus, it's reasonable to assume someone able to succeed in our society (which necessarily implies pretty good levels of sociability) that then decides to break one of the biggest social taboos we have is not acting rationally.
To get people** willing to kill without that training, you have to either get people who grew up with norms where killing is part of the culture (basically an honor culture) or throw extreme rage/intoxicants into the mix (if one watches enough The First 48, one notices how often "two guys got drunk/high and into a fight and somebody ends up dead" is a basic narrative of a murder, albeit it's usually not the modal one), be too dumb to work through the consequences, or yes, be psychologically defective (tends to go with the explanation before that, actually).
*I'd be curious to know what Scott thinks about Grossman's work; my understanding is he's viewed as something of a crank in the psychology world, but my impression is also he's deliberately avoided being part of that world anyway so there may be some element of him just not fitting in.
**as always, offer only valid on WEIRD people and not extendable to non-WEIRD psychologies, although most of the time we're talking about people living close to honor culture anyway
There seem to be a lot of psychopaths killing their spouses in cold blood in our society, if all those popular true crime shows are correct.
You write "normal" but I'm not claiming Luigi is normal in any sense, in quotes or not. I agree that the overwhelming odds are that he became psychotic or manic or schizophrenic or something pathological. There has to still be a tiny chance that he's simply a sane, however unusual, man who killed another man in cold blood for somewhat calculated reasons.
“There seem to be a lot of psychopaths killing their spouses in cold blood in our society, if all those popular true crime shows are correct.”
[insert airplane with chickenpox meme]
But they aren’t correct, is the thing. There are in fact so few of those people that those shows can go back decades and spend weeks (AI says an estimate for documentary shows is shooting about 1.5-4 minutes a day, and generally you can estimate about 45 minutes of show for an hour-long block with commercials, podcasts seem to spend about 2-3 hours per hour of research of actual air time) to produce these shows. Whereas even a Quibi (10-minute run time) focusing solely on homicides in 2024 the City of Chicago alone (595 at current count) would run for more than four straight days non-stop without commercials. An hour-long episode discussing Chicago’s murder would be a supercut of 6 second segments.
If they went back to 2021 (856), they could almost run their very own Shark Week! There’s simply no time to produce that much content, though.
Indeed, IIRC Peter Moskos has noted that one reason for the declining murder clearance rate in America may be that prior to no-fault divorce, a high modality murder case was the “husband standing over the body of his rapidly cooling wife, sobbing and sobering up (note my point about “extreme rage or intoxication”-probably a factor in guys murdering their gals!) with a bloody knife nearby”. These were very easy murders to convict people on (the husband almost always confessed, usually to the arresting cop). You can actually see this trend in this old BJS report from 2005 (should be page 103 of the PDF).
From the 2021 BJS data, more men are murdered by non-intimate family members (1797) than women are murdered by intimate partners (1689). The prevalence of those stories is not actually related to their frequency. Also “man killed by intimate partner” (something that happened to former coworker) is not that rare an occasion-approximately 1070, which is way more than the number of women killed by non-intimate family members or strangers.
Indeed, the primary reason for those stories is that the audience for true crime shows/podcasts is overwhelmingly female, and mostly middle/upper-middle class women who really strangely like hearing about other women (especially white women!) get murdered. Strangely, they don’t seem to enjoy hearing about black guys murdering other black guys over $250 of marijuana, or two Hispanic guys getting high and trading bullets, or even the white guys who end up in a drunken brawl that ends with one dead over the baby-mama they share, all of which are far more realistic depictions of what murder would look like.
I watched a lot of The First 48 at one point (which follows departments in a few high body-count cities) investigating murders. “Things that you would expect to see on true crime shows” was exactly one of the high dozens of murder cases that I saw (middle class white guy murdered his mom for insurance money; in the interview, they gave him a very quick and pro forma opportunity to confess-which he refused-because he’d left so much evidence in the process that they were looking forward to taking him to trial).
As I understand it Grossman is viewed as a bit of a crank in the history world as well. "On Killing" is largely based on the work of S. L. A. Marshall, which is now regarded as somewhere between dubious and completely fabricated. Although most reviews I've seen have admitted that "On Killing" has some directionally correct points, especially with regard to the rise of PTSD, I wouldn't put too much stock in its factual and historical claims.
Also, as a bit of an ironic side note for an author who wrote about the difficulty of getting people to kill their fellow man, Grossman and his institute have been running 'warrior mentality' police training courses, which seem to have increased police aggression and lethality.
I honestly think that's cope. Look at all the praise he's getting. Those people are one step from following him, with no obvious Schelling Point between them and him that stops them from going out and committing murder. But they don't want to commit murder, and very much don't want to admit it if they did. They know on some level that Luigi was *wrong* to do it. That his actions will not actually fix our healthcare system, and that murder is bad. So they have to rationalize between the very positive visceral reaction they are having and the reality of the futility and evil of Luigi's actions.
So he's crazy. Good-intentioned perhaps. He did it for the right reasons! But sane people know that's not how you go about social change.
Admitting otherwise says that you are a bad person who condones murder. Worse, someone who thinks they personally should go out and kill a healthcare CEO but is too scared to actually do what they think is right.
This seems to be recapitulating the arguments around bombing abortion centers or shooting abortionists a while ago: the so-called "pro-life" movement doesn't have a coherent response to "If you believe abortion is murder, act like it's murder," so they said the same kinds of things, "murder is always wrong," "violence is never the answer," "follow the law," etc.
The difference is whether clinic bombing and doctor assassination was cheered on by the respective community. It's hard to argue that a *lot* of people haven't been cheering Luigi on. I don't recall anywhere close to this response from the pro-life movement to clinic bombings.
I doubt Dostoevsky would agree Raskolnikov was mentally ill. Was "mental illness" even a mental model in 19th century Russia*? The main theme of Crime and Punishment is guilt. Through a guilty conscience, the murderer seeks his own punishment. You could say that Raskolnikov's philosophy at the beginning of the book, which is that one can weigh the costs and benefits of a crime to determine if it is really a crime, is shown to be wrong after the psychological costs of the crime destroy him, because he has a conscience. One could say that proves he was healthy, just very misguided.
Our Luigi doesn't appear to have a conscience, but having a conscience seems like a quaint notion in the 21st century. I think Dostoevsky was warning about such a future.
*It looks like the modern concept of mental illness appeared (in Germany) in the 1880s, which is post-Dostoyevsky.
To be clear, I didn't mean Raskolnikov had a diagnosable disorder which may render a person morally unaccountable for their actions.
I meant that his mind was not healthy. His thinking was disordered and delusional, and Dostoevsky is making a point there. Raskolnikov's philosophy is taken seriously by many in intellectual circles, who toy with insane and destructive philosophies as social statement and hobby.
To take such a philosophy seriously, to act upon it to its logical conclusion, would require a madman - like Raskolnikov, after he shuts himself up in his broom closet of a room for a month, driving himself insane with constant rumination. And because Raskolnikov does have a conscience, the act drives him even more mad, until he acknowledges the evil of what he's done and repents.
The main theme of Crime and Punishment isn't so much guilt, as the inhumanity of ideological thinking as opposed to Christian morality. A sane, healthy human being regards murder as evil, full stop. To deviate from the Christian ideal of treating every human being as a sacred icon of God is to derange and dehumanize oneself.
He dropped out of contact with his family and friends for about a year just before he committed the murder, which is odd behavior that wants an explanation.
There is (probably) a difference between (a) actual psychosis; and (b) using violence for instrumental ends, even if the person is mistaken -- in the ordinary sense of being mistaken -- about whether their plan is likely to work.
So we want to see some evidence of actual psychosis.
Old school terrorist groups like the Provisional IRA are usually considered to not be diagnosably mentally ill. But the more recent "lone wolf" type killers ... seem more like psychotic breaks.
[There may be a very important diagnostic difference between "belongs to a large group of people who are all using violence for a shared political end" vs "it's just him, and he didn;t have any co-conspirators"]
It’s hard to square a level-headed, intelligent man planning a murder in his right mind with Luigi’s actual actions.
He left DNA where it could easily be found. He gave an ID to be duplicated with his real face on it where it could easily be found. He left his backpack where it could easily be found. He kept the murder weapon on him. He departed the city via public transportation. He didn’t alter his appearance before the murder. He didn’t change his appearance after the murder.
Like, unless he had a good plan to disappear off the face of the earth forever (he didn’t), there was no way he would have gotten away with it. The picture on the fake ID would be more than enough for his friends and family to recognize him, unless his murder got little to no publicity, which would invalidate the whole point of it in the first place.
Raskolnikov was definitely mentally ill by the way. The philosophy that justifies the murder definitely was not what actually motivated it.
I’d be more willing to attribute such a thing to Ted Kaczyski, who through careful planning was able to get away with murder for decades (although he wasn’t incredibly effective at killing people, he was very effective at spreading terror). If it wasn’t for something that a person in their right mind planning a terrorist campaign would reasonably not account for (a brother he hadn’t seen in years recognizing his prose style of all things), he probably would have kept getting away with it for a while longer, or perhaps never been caught. Kaczynski’s Manifesto is quite coherent and has a mostly self-consistent view for justifying his actions if you accept his premises as true (not justified in a general sense, but at least self-consistent with his philosophy).
Luigi’s Manifesto literally says nothing of substance and basically just gives his reason “Man, this whole healthcare thing sucks! I think. I don’t actually know that much about it but I’m sure smart people could tell you why it sucks. Anyways, I’m off to commit murder.” This is not the manifesto of a man with a philosophy that accepts murder as a necessary evil for some grander goal that makes it justifiable.
I don;t think he was mentally ill, in the sense of psychotic. I do think that living with constant pain can do very bad things to a person's outlook in life and make even "crazy" things seem rational.
I suspect a lot of people are holding back from making the utilitarian 'if it saves more lives through helping create national healthcare it's justified' argument they really want to. After all, you could be accused of fomenting violence (a First Amendment exception) and a lot of people here might have cause to fear both the left and right.
I bet >5% of people here have read the Unabomber Manifesto and would say something edgier than "completely wrong and 100% against this kind of thing" if asked to review it.
You can apply this reasoning in the opposite direction though: if I think of the many schizophrenics and people who have had mental breakdown that I have met, none of them have had the persistence, competence and wherewithal to build their own gun, get fake ID, and organise and commit an assassination.
High IQ schizophrenics had more mania (not always to a statistically significant degree) and lower negative symptoms. Those of them who showed a decline of less than 10 IQ points from their estimated pre-morbid IQ also showed significantly fewer "negative-disorganized" symptoms.
They note that this lends support to the idea of a psychosis spectrum or continuum.
The relationship between IQ and negative symptoms that they discuss had previously been found at lower IQ levels, and they found them at higher IQ levels. But even the high IQ individuals in the study, only averaged pre-morbid IQs of around 120.
Mangione probably had a significantly higher pre-morbid IQ than 120.
Accordingly, the symptom pattern that they describe for high IQ individuals, most notably, the lack of the negative-disorganized symptoms, would likely have been even more pronounced in his case
Psychosis is weird. I have seen lots of psychotic people whose faculties seems pretty preserved. I think the guys who shot Lennon and Reagan did some okay planning, and Mangione made lots of strange mistakes, like continuing to carry all the evidence on him.
Yeah, that's the other weird thing. He seemed to go to some considerable trouble to avoid getting caught and get away in the initial shooting, but then kept evidence on him that was 100% damning after he'd gotten away. It seems like a continuing desire not to die in prison would motivate you to dispose of the gun and not carry around a written confession with you.
People don't act on their moral convictions like that unless they're mentally ill. Stable individuals have the good sense to value their own survival and social acceptance over all else.
I think the discussion is a bit tautological because "not acting on mental convictions like that and valuing own survival and social acceptance above all else" is almost baked into our definition of mental health. So anyone who behaves differently is assumed to be mentally ill, it's just a matter of slapping on a diagnosis.
I'm not sure; I suspect there's a bit of a U-curve here, where the extremely unstable and extremely stable both have the ability to act on their convictions, for different reasons. The 9/11 terrorists were not, as far as I could tell, unstable in any way. Quite the opposite, they had an unusually stable internal belief system that could survive being embedded in an environment which was very different to their internal beliefs.
Moderate stability leads to cowardice, like me. I'm definitely the sort of person who wouldn't say anything about people using slave-grown sugar even though I privately believed it to be abhorrent ; this is my current position regarding factory farmed meat, after all.
Motte: There are instances where a mental illness is basically socially defined.
Bailey: See, that's why this dude who has stripped off all his clothes and is out in the parking lot waving a knife around screaming that he is Jesus is really just socially defined as crazy.
In some places the rates of adhd dionisois in male children reach staggering heights; while there maybe a real thing of true adhd in 1%, I but call bullshit, genetic factors airnt this well correlated by state
By which I mean, modern life is much, much more chaotic and demanding than it was, e.g., a century ago, and therefore the threshold for ADHD symptoms becoming debilitating has lowered significantly. As a bonus, social media and the like are actively making them worse. (And as a bonus, as you seem to expect, ADHD medication is highly regulated and if people really want to drug children to stop them from acting like children, or drug themselves because the drugs can also genuinely improve lives of otherwise fully functional individuals, it's all or nothing. You really need that diagnosis, so that's what you get.)
If it was self medicating or modern *work*, there be more adults then children with adhd. Its children tho, so its the schools, a constructed unnatural environment imposed on everyone a mere century ago but treated as if its foundational and unchangable.
This sounds a lot like the argument that ancient people didn't believe their religions. I don't know if there's good reason to believe this: it sounds absurd to us, but a lot of views currently taken seriously are no less absurd.
But "drapetomania" isn't at all analogous to ancient religion. It was just one guy who came up with it, and little evidence that it was widely accepted.
That's because he effectively functions in a different society from ordinary non-criminal folks. In his society, killing and ordering to kill is required for "own survival and social acceptance".
It's not the killing that makes us conclude someone is unstable, it's brazenly going against social norms despite drastic negative consequences. Mafia boss's behaviour is in line with the norms of his in-group, Luigi's is not.
Right. Perhaps Luigi's goal was to become a popular hero. He wasn't delusional in thinking that he would be one. Maybe he doesn't mind being in prison as long as millions of people love and support him. It seems relevant that he's been unemployed since at least 2023. Why? He was a very ambitious student. He went to work at a company that developed software to help shopping for cars, then got laid off, or maybe he quit. It's not clear. Perhaps the corporate world broke his spirit, and he redirected his ambition in a completely different direction.
Often, I wished for a listing of young Scott`s aka Squid 134 writing on livejournal. Turned out, an outstanding person did it, six years ago! Here you are: https://archive.fo/fCFQx
Should I read Principia Discordia? When I look at that book's cover it gives me a lot of anxiety. I've never encountered a book that caused such a strong reaction in me, which makes me a bit curious.
Should be easy to find a PDF with no cover at all.
Whether it's worth it? For me, 60s counterculture is interesting for the ways in which it both sets the ground for modern wokeness and fails to meet its standards (a lot of them were young straight guys trying to get laid). However, I find that sort of stream-of-consciousness disorganized text to be extremely irritating. That's a very individual review and may not apply to you at all.
Hi all. I created an AI companionship platform focused on realistic interactions.
The platform is growing steadily and we have a nice comunity of users on Discord. My users recognize my platform as the most realistic out there. However, I have trouble keep driving traffic as this requires a lot of time that I'd rather spend developing the platform. Also I am not an expert in SEO, back linking etc. Is anyone willing to partner with me about this and help me organically grow the platform? Not looking for developers (I am happy to keep doing all the development myself), but for partners in the marketing area.
It’s generally a mistake to focus on SEO and backlinking in a crowded market, like AI companionship. You’ll be competing with larger players who can justify dropping $100,000 on a Bloomberg article talking about their product, which does about a hundred times more than any SEO expert with a moderate budget can accomplish. There’s just no way (without a meaningfully superior product, lots of free publicity or large amounts of money) to compete with large funded competitors when it comes to SEO.
Unless you’re selling something almost no one else is, or are ranking for a local search terms (Something like: “Computer Repair Shop St. Louis”) there’s no trick to getting it right.
There are other guerrilla methods of success besides SEO though. Reddit is a good one if you can grow your subreddit. Making a really good product that serves a niche, that gets picked up by a niche-celebrity is another.
Scott - I remember your position being that minimum wages don't cause harm. I wonder if you've revised this view since? How does a rationalist even arrive at this position in the first place?
The value of money is non-linear; while we may have games for making it more linear, such things need high volume efficient markets, job searching is not one.
If your starving, actually starving and you need 10$ to eat enough to fill full, it could be rational to enter into a 3$ pot Russian roulette game.
If your a multi billionaire and you were forced to pick between 1/10th of your wealth or a 50/50 odds of keeping everything.
If you were forced to gamble at a casino(or live somewhere where taxes are slightly lower if you play the stock market cough cough) it maybe rational to make a large spread of small slightly losing bets if your not confident in your capability to take large bets.
Given a non-linear curve of money, people with very little money, moving the floor up a little may make sense even if an economist would say "what happens if minimum wage is 1000$, would you be employed?"
This assumes (I think?) that minimum wage laws increase the real income of the poorest quintile in the long term (2 years out or more.) I'm moderately skeptical that this is true. To the extent that minimum wages are helpful, they seem to be an extra cudgel for things like wage theft from the poorest quintile of workers. Which still justifies them, but not for the reasons everyone else thinks. There are also some union contracts which tie wages to a multiple of the minimum wage. I'm not clear how influential such contracts are.
Policy makers may be particularly keen that low skilled, perhaps low ability workers who none the less are willing to get up and do their best, get a decent amount of money for it. MW may help with that.
Whether it does in practice is, like all questions in economics, a matter for empirical study, and it may be hard to untangle effects. Broad suspicion about interventions by the state is warranted, but then again the arena in which those interventions play out is a product of political, cultural, legal etc factors going back years. It's seldom fair to see MW as the only intervention and everything else as being in some sense the free and fair market.
It follows from this that an MW policy may play out differently in different places at different times.
I'm not particularly concerned to advocate MW. I just really don't like the implication that for someone broadly "rationalist" there can only be one answer.
Interestingly, as the Neumark and Shirley paper points out, the distribution of harm from minimum wages is particularly concentrated among those with the lowest skills and ability, almost exactly what economic theory would predict
We've had minimum wage in the UK since July 1998, but it only applies to the employed, not the self-employed.
There have always been a lot of one-man businesses or small groups of partners subsisting below the minimum wage (or, before that, what those in employment would regard as a minimum wage). How do you think the big names in silicon valley got started (the socially acceptable phrase is "started in their bedroom/garage")?
More recently the minimum wage has forced a lot of people to become self-employed and join the gig economy.
My point is, whatever a government does, the people and/or the market will try to circumvent it so beware of unintended consequences.
I currently think it's really complicated, and it depends on the exact wage increase, the industry, and the prevailing economic situation.
(or rather, benefits > harms under some way of valuing things, since I think it's near-impossible for any intervention to cause NO harms)
I think this is what the literature says. Certainly there are many papers showing net benefits in some situations and others showing net harms in others. I understand that maybe one set of papers is wrong, but when I look at analyses by people I trust (sorry, I've forgotten specifics), they suggest it's more likely that they're observing real variation based on the particular area under study.
How is it possible for a minimum wage not to cause harm? I thought I'd previously seen an argument that made sense based on sticky wages, but I can't find it again or reconstruct it. All the arguments that make sense to me either assume companies were making dumb mistakes before the minimum wage increase, or that it increases jobs at the company's expense by shifting to an equilibrium where they hire more employees but get less profit.
Hopefully a commenter here can give a good explanation; otherwise I guess I'll retract my previous position and go slightly more in the direction of thinking the studies were flawed (though of course it could still be true that some minimum wages don't decrease employment *very much*).
Would you mind enumerating the buckets of 'harm' and 'benefit' that you're employing? I think that there may be an argument to be made that, in some situations, min wage laws reduce average hours worked. This would be a benefit to consider, independent of income. And I'm trying to figure if you're accounting for that, and how much weight you give to people having more free time, on average.
>How is it possible for a minimum wage not to cause harm?
Worker's productivity rises with the rise of his quality of life, and quality of life raises as the amount of available resources rise, and available resources for a worker are determined by his wage. (Meanwhile, society-wide, the amount of available resources rises as productivity rises.)
There are suboptimal equilibria where society as a whole would benefit from increased quality of life of its members, but no individual company can expect the resulting rise of productivity to offset the cost of raising wages, so it won't happen on its own. (As for why it might be so - note that companies are in many ways shielded from the consequences of their workers' low QoL. E.g., they're not affected by workers' health deteriorating faster than it should, they can simply replace them with new hires.) This is where the government steps in and solves the coordination problem.
>Worker's productivity rises with the rise of his quality of life
How much and what's your reference? I'm very skeptical. In any case I'm also skeptical that there's a firm connection between MW and quality of life as the former can cause inflation which harms low-earners disproportionately - the 2020 covid stimulus is a good example of that, though of course that wasn't a wage. Putting money in the hands of the poor tends to be inflationary because they consume almost all of their income.
A few years ago, Jennifer Doleac (an economist who mostly focuses on criminal justice issues) had some comment on Twitter implying she doesn't trust the minimum wage literature because it is so political.
If the bias of each source is known, we might be able to find something useful by looking at multiple sources with opposite biases (also if any specific conclusion is against the bias of its source, it's a lot more trustworthy than the other way around). I believe Scott used this approach recently in his prison article, let's see if we can do something similar with the minimum wage.
In April, California raised the minimum wage for fast food workers by 25%, which should be a high enough amount to see some effect. Limiting the increase to fast food workers should also give us some natural control groups like non-fast-food restaurant workers or fast-food workers in other states. The whole thing has been heavily discussed, so we should be able to find many sources with various conclusions.
As expected, every article I found was heavily biased, but let's see if we can learn anything useful:
- UC Berkeley study (https://irle.berkeley.edu/publications/press-release/new-study-analyzes-impact-of-californias-20-minimum-wage-for-fast-food-workers/) found a 3.7% increase in prices. Given Berkeley's bias we can probably treat this as a lower bound and conclude that at least some of the cost has been passed to customers. A mere 3.7% cost increase for an 18% wage increase (same source) could still be considered a win for minimum wage, but given the source I'm not comfortable saying that 3.7% is anything more than a lower bound -- so, inconclusive.
Since the governor and NR both relied on BLS data, we should probably dig in some more. Neither source links to the exact data they used, but the governor at least links to the relevant search form (https://data.bls.gov/PDQWeb/sm). It's not clear to me how to find fast food jobs specifically, but here are some relevant categories I could find:
- "Restaurants and Other Eating Places" slightly down based on September (most recent available month, 1352k to 1351k) but also slightly up if you pick a different month
- "Limited-Service Restaurants and Other Eating Places" (sounds like fast food but I don't know if it matches California's definition) slightly up (737k to 738k) but was slightly down in June (conveniently the exact month picked by the NR article above) and also the growth did seem to slow down significantly around the time the new law took effect (year over year growth was 10k in January but hovers around 1k since May)
- "Full-Service Restaurants" (the control group?) went down (615k to 613k), not sure what to think about that
At this point I lost interest, but I'm convinced we can learn something useful about minimum wage from California's fast food experiment and I'll probably take another look at the most recent data in a few months.
Might I suggest also considering total income and total hours worked since both are likely impacted. There are arguments in favor of a minimum wage, but the whole notion of a 'living wage,' specifically, which is perpetually trotted out becomes moot if total wages of the poorest quintile don't increase significantly.
Here’s two pieces of economic theory that support a minimum wage. The first is the idea of a “reserve price”, which can be used in an auction to maximize revenue. In this case, it says that even though there may be less jobs, a properly set minimum price causes more total revenue to flow to the lower class. The second, which is less theory and more analogy, are unions. Minimum wage is the government collective bargaining on behalf of low-wage workers. As is standard, a union can push too far and tank a business, but can also simply change the distribution of profits within a system.
An auction is a terrible model to use in most settings that are not auctions. But even a reserve price in an auction is set by the person selling the good, not by government!
And government privileged unions are terrible news for most industries. Government is the most govt. privileged of government unions!
Not only did giving the poor more money cause inflation, it led to them infiltrating higher income areas and bringing their dysfunction. E.g. the massive increase in unruly airline passengers following the stimulus.
I don't pretend that any of these are knock down arguments. And I also don't argue that min wages necessarily increase the real earnings of the poorest quintile. But...
1. Productivity also decreased significantly during Covid, which also contributed to inflation. 2019-2021 is a problematic time to base studies on, because there's a lot of noise.
2. In normal times, some level of consumption among the poorest individuals could also be considered investment. i.e. If you can't afford a reliable car and gas, you cannot optimize the value of your labor. The cost of borrowing for low-income individuals tends to be particularly high.
3. Property crime is reduced by government programs to provide a minimum level of consumption. Presumably, property crime is a negative sum game. Though how much of one is beyond my pay grade.
4. We currently have social programs in place. Increased wages may reduce some of these government programs. It may also better navigate benefit cliffs, which are problematic. (But which could be easily addressed without a minimum wage.)
5. If consumption could be reallocated to some extent, it should increase average utility. (At the expense of maintaining property rights, and the bulwark against authoritarianism that tends to entail.)
Again, none of these are knock-down arguments. But they do address the (possibly rhetorical) question of 'why?'
Sure it's complicated, and I agree that supply problems were partially responsible for recent inflation. But certainly not entirely.
Agree that some level of wealth transfer to the poor is necessary to keep society stable but I think that minimum wage is probably the worst way to do that. It just distorts price signals and carries inflationary risk. A negative income tax is much better.
>"How is it possible for a minimum wage not to cause harm?"
The trivial case is when it's not a binding constraint, id est, if every worker can command a higher-than-minimum wage through normal market mechanics then the law is inert.
Re: How is it possible for a minimum wage not to cause harm?
See: mathematical step function. If the increase in the minimum wage is less than a certain amount it won't cause any effect in employers (who, unless their business is right on the edge of failure, have some "cushion" to absorb price change shocks.) Price increases do happen in other business inputs all the time-- why should wages be any different from the electricity rates going up? Compare this to personal behavior if the temperature in a room goes up by one degree, Most of us won't even notice and will continue to go about our business. But if the temp goes up by ten degrees, yes, we will notice and react.
Human behavior is rarely a smooth linear function, and assumptions that rely on linearity in economics are most generally wrong.
Does this not cause harm to the employers? You may look only at the businesses right on the edge of failure, but then you should also look at the ones adjacent to that, who will now be on the edge of failure. Then look further at the ones who now must be concerned their "cushion" they carefully engineered to withstand the unexpected must now be reevaluated, etc.
Fundamentally, why should the employer have to pay more? As far as I can see, it is only if the workers are worth more than what they are getting paid, and do not have the freedom to seek an alternative job where they can be fairly compensated. And getting "fairly paid" is often a judgement call without any objective way to calculate it.
If you're running a business you certainly cannot count on your expenses being set in stone. If labor goes up why is this any different from increases in raw materials, utilities, commercial rents etc. etc.
I should also mention here that the US federal minimum wage is a joke these days. It is far below the"natural" minim um, wager that the market itself has set. You can't hire people for $7.25 these days. A few years back I recall people outraged at the suggestion the minimum wage should be $15. Nowadays in a lot of places that's the true minim,um wage set by supply and demand.
There was a short period in which it appeared that state minimum wages (which apply to well more than half the national population) were getting close to true market minimums in various regions. That moment passed though, the state minimums are now mostly well below market and aren't catching up anytime soon if ever.
If the Trump plan for mass deportations of immigrants actually happens then that will further reduce the national unskilled-labor supply and presumably make minimum wage laws even less relevant. Particularly if Trump's announced desire for the deportations to include some non-citizens who are here legally as well as citizens who are the children of illegal immigrants, is realized.
As a business, yes, your expenses can go up, and if they do, you must decide whether to eat those costs or pass them on to your customers. This is straight-forward inflation. I don't see how artificially setting any kind of minimum wage is not harming the employers.
Except, as you mention, when the mandated minimum wage is irrelevant. And in that case, why have one?
Thanks. Someone else has shared the Neumark and Shirley review which I personally think is very much right. But also, the Bayesian process on this (imo) should go as follows -
1) We have good reason to believe that prices encapsulate useful information generally and that demand reduces as prices increase
2) Artificially set price floors or other price controls will reduce demand where it does not need to be reduced and otherwise introduce a host of inefficiencies that result in poor outcomes for everyone
3) We have loads of good evidence from many many different sources for the above propositions - rental markets, product markets, labour markets, you name it
4) This one weird trick in labour markets lets us believe a politically convenient free lunch exists where a price floor will not cause outcomes to be worse.
5) We should not update on 4 very much, if at all!
Neumark and Shirley (2022) - "Myth or measurement: What does the new minimum wage research say about minimum wages and job loss in the United States?" - Is a recent review of the topic that came to the conclusion that the expected effect of minimum wage is probably negative, with a nice approach of asking the researchers directly which of their paper's estimates is the most relevant. From the abstract:
"Our key conclusions are as follows: (i) there is a clear preponderance of negative estimates in the literature; (ii) this evidence is stronger for teens and young adults and the less educated; (iii) the evidence from studies of directly affected workers points even more strongly to negative employment effects; and (iv) the evidence from studies of low-wage industries is less one-sided."
Though I don't think their conclusion is the new consensus, They changed my view on the topic (update from "probably no effect" to "probably negative") though the literature is very American, and if you have some local estimates you might prefer it. In Israel for example the last estimate I've seen is for no adverse employment effect and some redistribution from small business to low wage workers (either the Treasury or the Central Bank published it, I don't have it on hand).
There’s more to the world of economics than Econ 101. You havent rigorously defined harm there but the position against to the minimum wage is either that it causes high inflation, or higher unemployment.
Sometimes when the claim to higher unemployment doesn’t pan out there’s a claim that the increases in employment aren’t as high as they would have been which is hard to verify.
One might venture that for any economic configuration there is a natural value of "minimum wage". One that identifies the lowest "value" that configuration places on labor. If the official minimum wage is less than the natural value, then there is no economic impact to increasing it. The natural value is determined by the aggregation of the feelings (waves hands) of the various cultures that are part of that economy. Huh, I guess this really doesn't cash out to anything. Oh well, it's typing practice.
No, you're very correct in your thinking here. There's a wage at which employees are willing to work and employers are willing to hire them. That's the natural minimum wage. Raising the legal minimum above that rate will cause disruptions, while leaving it below that rate will have minimal effects.
There's another aspect, which is that some employees are worth more than others, even filling the same exact role. Two people working an automated assembly line, where you would expect employees to have very similar levels, may have OOM differences in productivity.
There's some obvious reasons for this, even beyond how much that machine produces with minimal human interaction. Someone with really bad attendance is obviously less productive than someone who shows up on time every day. Someone who picks fights with coworkers costs the company. Someone who sues the company obviously costs a lot more. Someone who inexplicably breaks the machines a few times a week and causes hours of downtime. And then the opposite of all these things - an employee who is always at work, calms down disgruntled employees, and whose machines break down less than company average.
Minimum wages may or may not cause a reduction in employment at a given company. They absolutely cause a reduction in employment for the least employable and may cause an increase in employment for the more employable, including people that were already able to make more than minimum wage. If companies are already paying a higher wage regardless of the person they hire, they may as well hire better employees.
The specifics matter a lot. Some companies will pass on costs to final consumers, but only if they are able to. Some will shut down or hire fewer people (maybe close during their least productive hours). Some will eat into their profits. On the whole, it's very difficult to predict the actual results because the results will depend on a wide range of factors simultaneously. The Econ 101 overall tendencies will be generally correct, but with enough factors that confound analysis as to make that less than universal. Economically depressed regions or industries are more likely to shut down and lay off, while growing places are more likely to pass along costs.
No it isn’t. It’s the typical scurrying away when the argument is proven incorrect.
There will be unemployment when we raise the minimum wage is clearly not that same as we won’t hire as many people we could have but will still hire lots, enough to handle labour growth.
Of course if economists actually had a formula for this - something like a 10% rise in the minimum wage will always and everywhere increase the actual unemployment rate (not the completely unverifiable woulda coulda shoulda hired more people in a counter factual world we put into our simulations) we could falsify it but they don’t so we can’t.
>"There will be unemployment when we raise the minimum wage is clearly not that same as we won’t hire as many people we could have but will still hire lots, enough to handle labour growth."
"enough to handle labour growth" assumes the conclusion, try again.
The problem with such statements is they are counterfactuals: we can't prove or disprove them because they aren't real. It's no different from saying"If Ronald Reagan had't been elected in 1980 the USSR would have won the Cold War". Since we have no access to parallel universes where (maybe) that did happen, there is just no way to know that.
All models are wrong, some models are useful. The supply & demand model is consistently useful across a wide range of contexts, and the conditions in which its fundamental results can be expected not to hold (e.g., monopsonistic power over most/all of a labor market) are rare and not subtle.
FWIW, my current day job includes modeling conterfactual supply & demand conditions for multiple markets to guide various sales & marketing efforts; I'm painfully familiar with the challenges & limitations of drawing useful inferences from them.
Assuming this is true, doesn't this make the claim about raising minimum wages unfalsifiable?
Any time the employment either remains the same or even increases after a minimum wage hike, you can always trot out the same argument that "it would've gone up more without the wage hike". What evidence is required to show this may not be the case, when "it actually not being the case" is not enough?
1. Implying, as Peter does, that emphasizing slower increases in employment is moving the goalposts is arguing in bad faith; that's what I'm primarily pushing back on.
2. Differences in rates are harder to detect than differences in levels, but not impossible (thus still falsifiable). Extrapolating from any single instance is akin to rural Minnesotans asserting that climate change is bullshit because they're currently in a blizzard.
I don’t think anyone should read Card and Krueger as proof of anything (they should instead read it alongside the entire body of research answering the same question) but it does answer the question “How does a rationalist even arrive at this position in the first place.”
If a single study (especially one from before the replication crisis) provides sufficient evidence for someone to update to "minimum wages don't cause harm" then their prior is too weak; the Law of Supply & Demand is *very* clear that binding price floors create surpluses.
If someone said “how could anyone begin to enjoy reading books,” I might point them to a book I like. It would be silly to respond by pointing out that I haven’t established that the entire hobby of reading is fun - I was just providing a jumping off point!
Beyond that, your comment about the replication crisis makes me think you’re not very familiar with this area of research.
Perhaps overly primed by having seen too many people treating Card & Krueger as the final word on the topic, I interpreted your initial response referencing that paper as a drive-by defense of minimum wage laws rather than an answer to the "how" question.
The replication crisis reference was a bit of a drive-by of my own, and for that I apologize. I had followed the minimum wage research somewhat closely over a decade ago (because the headline results were so contrary to my priors, if the findings were sound they'd've warranted a deeper reexamination of my economic understanding); last I paid significant attention was a meta analysis with a dot plot of (I think) employment elasticities to minimum wage by study, which showed a clear but to-me-surprisingly-small negative effect. The qualitative result was in agreement with my prior (so no soul searching) but the quantitative result was small enough to reduce the priority of the issue for me such that I stopped paying nearly so close of attention.
Apparently Javier Milei of Argentina has eradicated his country’s deficit right in the 365 day mark since taking office. I couldn’t find the figures and how he did it? Does anyone know the details?
Milei's interview with Lex Fridman has some good info too. A big one was just cutting a ton of Argentina's welfare and a bunch of other government spending too
I’m not sure he eradicated the deficit as they will need to borrow more money soon, he HAS tamed inflation which was the biggest issue. Inflation in Argentina was crazy. You could watch prices climb as a slice of pizza would cost more with each visit. He cut spending from 44 to 32% of gdp which is a great way to put a spanner in the works. Note, the economy is stuck so now that inflation is stable, he needs to get things growing.
“He did so by pushing executive power to its limits with emergency decrees and vetoes, leveraging his direct connection with voters via a flurry of posts on social media to pressure lawmakers, and negotiating fiercely with Argentina’s 23 provincial governors after slashing their funding.”
I am pretty sure he stopped inflation by creating a massive recession, which is probably the right way to do it, but had costs. He might have created a massive depression which is not the right way to do it. We will see.
Doesn't seem likely. Q3 experienced 3.9% GDP growth, already ending the recession in Argentina. While 2024 as a whole is projected to have an economic contraction of about 3%, forecasters expect growth in 2025 of over 4% (with some projecting growth of over 5%).
And other economic indicators have been improving, as well. For example, job numbers are up and the poverty rate peaked from Jan-Jun of 2024 (after rising for several years), and has dropped each month since then.
Is there an easy way to know if there's a follow-up of old SSC articles? I'm currently binging it and after sometime I'm curious if he still stands by those articles. The only place I see is /mistakes but it'll not have all the follow-ups especially if the old article isn't really mistaken.
I came across this recently by the late great Herbert Simon: “If men do not pour new wine into old bottles, they do something almost as bad: they invest old words with new meanings.”
Lately I notice people in public arguments trying to redefine words, presumably in the hopes that their redefinitions will catch on. They often argue they’ve improved some descriptor by replacing its naturally vague conveyance with fresh, precise meaning. Whatever their intent, they must notice how this retroactively fucks up every opposing argument that relied to some degree on the word.
I’ve nothing against stipulative definition or exploring a term’s meaning and see the obvious good in these things but I don’t see it in trying to overturn general usage based on its polysemy. Usage is what determines authority to a general audience. It springs up to fill a need. If you want to talk about something more specific, maybe coin a new term or add a qualifier (‘clinical’ depression, ‘aggravated’ assault).
You see redefinition a lot with conceptually loaded, fuzzy terms like:
Misinformation
Art
Freedom
Cancel Culture
Conservative
Liberal
Man
Woman
These words like all words grow naturally over time without some prescriber getting involved. At the very least, prescriptive redefinition usually just undermines the existing consensus on which shared language depends.
(Strangely enough, attempts to redefine a particular term seem to come in waves, as one redefinition is seized upon by the opposition as reason for a counter redefinition, leading to both parties denying linguistic reality.)
This is one of Scott's earliest essays, about motte-and-baileying people with "feminism is just saying women have a right to exist" and then saying "since you agree with feminism that means you agree with affirmative action."
I mean, it works, that's why they try to do it. Look at what they try to do with 'racism' (associating resistance to progressive demands with the opprobrium everyone attached to Jim Crow or the Holocaust), or 'misogyny'...
"So a fantasy about a knight who risks his life to save a princess from a dragon is misogynistic because..."
"It denies the woman agency and reifies traditional gender roles."
"So he should leave her to get eaten?"
"Well, no, he has a responsibility given his position in the power structure to use it for good."
"What if I made the knight female and the princess a prince?"
"Well, that's a step forward, but it's still heteronormative."
"Make them both women?"
"That's not your story to write. You're not a woman. It's not #ownvoices."
"OK, make them both men?"
"Unless you're gay, that's not your story to write either."
"I think you just don't want me to write anything."
"Of course, you didn't see the line saying 'looking for marginalized voices'?"
"Good point. I think I need to self-publish, I think the Gor series could use an update..."
Absolutely. You could have a fun, sexy series (if that's your cup of tea) by stripping out the WOMEN ARE NATURAL SLAVES AND WANT TO BE TAMED AND TORTURED guff. I don't care about John Norman's philosophy, it was bonkers. If you want sexy feisty slave girls who need to be tamed by the masterful hero (but maybe cool it on the whipping and branding and starving etc.), that's a standard romance novel trope, just set it IN SPACE and the job is Oxo. That'd be going back to the setting of the first couple of books in the series, which were more like that than the later ones.
Alternatively, you could just make it gay and drop all the women as love objects. His hero, Tarl Cabot, was a lot more interested in the big manly men of Gor as role models/hero worship/crushes anyway 😀
On the gripping hand, you could just dump all the sexual politics and update it to be a fast-paced planetary romance (no, not that kind of romance) because I'd like to know a lot more about the Mysterious Hidden Masters and just what the heck they are up to, for real, those Priest-Kings.
I will admit to having read a very late-period Gor novel when I saw it at a used bookstore and so learned that he was still self-publishing rather than having disappeared entirely when DAW(?) stopped publishing his books. Was curious as to what the series had turned into.
And the man seems to have mellowed in his old age, because he seemed to be trying to do, as you put it, "standard romance novel trope" but in the established Gor setting. Half of the book was written from the "slavegirl" point of view, and yeah, mostly standard bodice-ripper stuff. The other half from the "master" PoV, and he is baffled by the fact that he is actually falling in love with a woman in spite of all the cooties. The setting of course demands that she be a natural submissive and that all the social machinery of formalized sexual slavery be in place to define their relationship, but it sort of actually worked.
I would pay real money for him to have finished the series with his new power couple going on to take down the Priest-Kings and unravel the true mysteries of Gor, because like you I think that part of the setting was the most interesting and underutilized. But I don't think he went in that direction, alas.
Still, it was better than "Twilight" or "50 Shades of Grey", and that is something I would not have expected from John Norman.
How old were you when you encountered them? I was young enough to simply ignore the romance scenes in books, but in the Gor novels it simply wasn’t possible.
"No, Deiseach is not a kajira. She serves the Christ of Earth."
The Anonymous Dude sat behind his keyboard, looking at the screen. He was not sure how to answer. Deiseach's true nature was still uncertain, but that was to be determined. He was not a Gorean, but his nature was not entirely dissimilar...
OK, I'll stop.
Those are fun ideas! I had a couple of my own: (1) Keep the slavery, but make it gender-neutral; anyone can enslave anyone. Bigger audience and you can get the femdom crowd. (2) Make it totally consensual and it's just kink in space. I figured out the smallest group that gives you every potential scene is four: two bisexual doms and subs of either gender.
Frankly with literotica and Archive of Our Own I doubt there's any money in this, even with the obligatory female pen name. I should probably just toss the whole idea and get into EA so I can try to save chickens or something.
I read the first 3 and you do find out what's going on with Priest-Kings. I won't spoil it if you want to go back and read them. There are 37, I'm told they turn into mostly porn after about 7, though there's a second alien race and so on. I am curious to finish The Most Politically Incorrect Series Ever, but 37 books is a lot. Supposedly the guy is still alive at 90 and writing these, I wonder if it's his kid or his wife or something.
> "No, Deiseach is not a kajira. She serves the Christ of Earth."
I don't know what a kajira is, nor am I familiar with the world of Gor, but I just wanted to say that this is a damn powerful line. Let's fund space travel so Deiseach can say this to an alien.
The word "refute" is another example. It used to, and should still, mean "disprove", as of an allegation or claim. But sneaky politicians have taken to using it to mean only "deny", as if they hope to smuggle out the subliminal implication that their denial in itself constitutes disproof.
Even "Google define" now lists this alternative meaning. So it looks as if the corruption of the word is complete:
Not to be confused with demur! Both words are close in meaning though because demure (unless you are thinking of a new sense I'm unaware of!) means reserved or modest, in a sense holding back from putting oneself forward, and the verb to demur (n. demurrer) is a legal word meaning to object to, or figuratively recoil or step back from an opponent's claim.
Do you think language is changing faster now than in the past? Perhaps there has been more change over the past couple decades, but I don't really know. More broadly though, it seems like English has been pretty stable for the past 200 years. Other than a bunch of random capitalizations, English books from the 18th century aren't hard to read and understand. Go back another century and it gets much harder.
My sense is that sometime in the late 20th century, political operatives noticed the power that deliberate choice of language has over the national conversation (and thus the electorate) and ever since then we've been in a war over which words are used to describe things and what things are described by any given words.
Some examples are "anti abortion" versus "pro choice", or "undocumented immigrant" versus "illegal alien". Did we have this kind of thing 100 years ago? Maybe a little; "states rights" was used as a euphemism for slavery during the civil war. But I think this sort of thing really took off around the 80s.
I disagree. Now someone can misuse a word, post that online, and it gets taken up and spread around and everyone is using it that way now, and if you try correcting it you are just rowing against the stream.
The Internet has made it much faster, much easier, and much *much* wider audience to reach with "literally means figuratively" than the past, where you had to write things down on paper like a caveman.
I think those newpapers standard guides will effect how English is taught to children for 300 years, with everyone seeing red underlines as objectively true errors. While Shakespear invented several words.
If I intentionally break grammer rules, write some historical long winded sentence that doesnt appear is the newspaper style guides, or maybe use repurposed punction like `,?` `!?` or `~`; im treated worse, maybe do it anyway because bring back anglish...
The corrective mechanism is computerized and objectified and nearly mandated, this is a major change from the 1800's where educated people had different opinions on how to spell words and dictionaries were still new ideas.
That’s a good question. As you suggest, grammar and spelling have remained largely consistent in English over the past 200 years, but it’d be interesting to see how a dictionary from 1824 might differ in its definitions of the same words from a dictionary today. And I do think the rate at which new words are being introduced has increased drastically due to technological dissemination.
Maybe grammar and spelling are changing less, though the timeframe may also be too short as of yet to tell. But word meanings seem to be changing faster. "Irregardless" is apparently now a word, and "literally" is literally its own antonym now. That doesn't even count slang such as "skibidi toilet".
No, it's just wrong. The etymology is quite obvious. How can anybody ever learn if instead of telling them they're wrong we redefine the words to suit their utterances?
"The Sun orbits the Earth" so we make "Sun" an alternative name for the Moon? Really? If only Galileo had realised this he could have saved himself a lot of unpleasantness.
If it's not obvious, I agree with you. Websters, however, disagrees, because that is how people use it. I will not, however, use it to mean something like exaggeratedly false. But I will still understand what people mean when they use it that way. And cringe a bit, if only inside.
And the OE presumably takes the same stance, [except most of it hasn't been updated since 1903.]
I understand their postion: I also cringe when I hear it, and ditto lent/learnt used incorrectly. And I'm afraid I also judge, which of course is what it's all about.
Does anyone (1) use LLMs to generate Anki flashcards from PDFs/books/articles and (2) would be willing to share their prompt/preferred model provider with me?
I've been experimenting but am having a terrible time getting the chatbot to produce flashcards I *actually* find valuable (specific definitions, numbers, equations, etc.) as opposed to just quizzing me on anecdotes or asking me to basically complete sentences from the text.
Ok, I tried this in ChatGPT o1. If it's relevant, the pdf is 336 pages.
"I've attached the standard operating procedures for the airline I fly for as a first officer. Can you create a deck of Anki flashcards from chapter 2, as a CSV file, with the first column for the front of the card and the second column for the back of the card? To start with, just ten cards as a test please. The cards should be like typical flashcards, with a question about the material on the front, and a specific answer on the back, not just a reference for where to look up the answer in the manual. It is EXTREMELY important that the answers on the back are correct."
About 80-90% of the cards are great, and the rest hallucinate incorrect answers. I liked this one in particular, might cause a bit of a problem:
How should the thrust reversers be operated under cold conditions?
Cycle between forward and reverse at least once every 60 minutes to prevent freezing
Of my desired output? It would take the form of a text file with cards separated by linebreaks, and questions and answers separated by semicolons, like so:
question;answer
question;answer
Perhaps with some so-called cloze questions as well, which look like
A {{c1::cloze question}} has multiple terms that can be {{c2::hidden}}.
I did this recently with a book on AI itself. A result I like is:
"How many connections (parameters) are in the human brain?;About 100 trillion"
A result I like less is:
"Why is calculating probabilities important in ML?;ML algorithms assume training data and test data are drawn from the same underlying probability distribution - if this assumption is violated, predictions may be unreliable"
I’ve been thinking about this a bit lately (don’t have a prompt/model provider pipeline set yet, sorry), but was thinking about feeding it bits of Quantum Country or some other Andy Matuschak content to do few-shot, then see how much I can remove while getting comparably-good cards to keep inference costs down.
fine-tuning a frontier model is also an option, if you have access to enough high-quality (pdf, anki cards from that pdf) pairs and have the $/want to turn it into a business
I wrote an essay about a problem that currently seems small, but if model costs continue to increase ($200 for o1 pro?), might be a big issue over the next few years.
I read the essay, and it seems fairly anodyne, with nothing controversial that jumps out at me. But it does not seem to address the hook question: "What will you be doing the day before AGI arrives?".
It seems to focus on the potential inequality gap growing, but the argument is not very coherent or persuasive.
Ironically, the assay would likely benefit from getting Claude critique it and helping with editing. As it is, it reads "too human".
I've been thinking about anti-Orwellian white pills.
Totalitarianism is inherently unstable, bc the totalitarian leader by definition attempts to amass 100% of his jurisdictions resources to himself (or immediate family) at the expense of everyone else, even his regime enforcers. Thus at the margin, the lowest ranking enforcer is only slightly better off or equal to the highest ranking slave civilian, so there is perpetual incentive for bribery and trade across that threshold. In fact the only way their economy avoids collapse is civilians engaging in black market, necessarily involving state agents. There can never be a perfect regime, they will always corrupt themselves, and inevitably get messy enough for revolution.
Your model only seem to work for African-style totalitarianism, imo, where the leader cares about wealth. It doesn't describe e.g. Stalin-era USSR at all: Stalin and the likes of him don't care for wealth all that much. Power is what they want. Money IS power, but they are canny enough to know what kind of money could become dangerous to their position, and ready to share wealth, up to a threshold.
The only real problem is that succession. USSR fell apart in no small part because just had NO succession model, and power was passed to the next generation haphazardly, with the next man in line often being almost-random, voted "least likely to bring back Party purges" by his peers.
But what is we mix a bit of monarchy with totalitarianism? If the chosen successor (and other children of top decision-makers) is properly groomed from the early childhood, it becomes no impossible for such state to survive for longer. It still will not be eternal, but it can endure for longer than seems possible from your model. I think this is a part of what is happening in North Korea: Kim dynasty IS a dynasty. While the family might live in far more opulent style than an ordinary Korean, or even enforces, they train their children to know not to antagonize military or secret service, or even common people too much. Compare and contrast to African and South American dictators, who usually end up deposed in the first generation.
Then again, we now live with a possibility of AI/algorithmic dictatorship. For an AI, wealth might not matter at all.
In 1984, the dictatorships were run by oligarchies, not single leaders. They were also highly technocratic and understood the science of maintaining power, so they probably would have avoided making the mistakes you describe.
when corona hysteria hit, I grew a spine a month into the "2 weeks", telling people I started refusing to wear a mask got me 3 verbal abuses online that I feel crossed the line; it was 5 seconds to stare down the one person who ever asked me to wear a mask irl a grand total of *once*
While the internet is the world culture and I wont ever discount it as irreverent, real life has far more constraints on peoples sanity, gun rights suggest a guerrilla war I dont believe anyone is willing to stomach any time soon; I will never be doing even two weeks again
you can find arguments from the ... grumpy ancaps, about the impossibility of enforcement if .01% of the population has a spine, cops being wildly outnumbered and all that; we need to more worry about that 1/3rd of people were informants to the secret police in the historical examples
Well yes, but democracy is also inherently unsustainable, so... Change is inevitable, entropy takes all, etc etc.
Though, I do think that a stable dictatorship is possible in theory. It just requires the entire population (including the leadership) to be in thrall to an unchanging concept. Religion is probably optimal, but other types of ideology can potentially work as well. Though at that point, I guess it doesn't really matter whether the government is technically a democracy or a dictatorship, since the mechanism through which it stays stable is by robbing *everyone* of agency...
Yep, dictatorships are pretty stable as long as the ruler is alive - there are many examples of that in the past and today. I think the tricky part is when the ruler dies and you need a change of government. In my country (Portugal), we had a dictator Salazar that ruled between 1933-1974 - a pretty long time. Old people like my grandparents still have fond memories of him. A few years ago, there was a big vote on public television for the greatest Portuguese, and Salazer won. What happened when he died in 1970 (he fell off a chair) was that the government pretended he was still alive for a long time, and it lasted for another 4 years. It would have lasted longer, but the military was fed up with the colonial wars in Africa, so they deposed the regime peacefully. Otherwise, it may have continued.
Some dictatorships have solved this "change of hands" problem by making them hereditary (monarchies), but still, history has plenty of cases when this becomes unstable. Democracies seem to be better at this - maybe. The USA has been going for 235 years now.
"After the incident [falling off a chair or in the bath], Salazar's life went on normally. Sixteen days later, Salazar admitted he felt sick and he was admitted to Hospital de São José two days later. On 16 September, he went into a coma.[172] With Salazar incapacitated, President Américo Tomás considered that the 79-year-old prime minister would die soon; on 25 September, he dismissed Salazar and replaced him with Marcelo Caetano.
Salazar lived for a further 23 months. After he emerged from over one month of coma[172] and unexpectedly recovered lucidity, his intimates did not tell him he had been removed from power, instead allowing him to "rule" in privacy until his death on 27 July 1970.[16]"
I agree with 10240. The masses don't have to be starving or even poor. It's not hard to imagine a future USA in which we have about the same level of wealth as now but with zero privacy, limited freedoms and zero political voice. No reason a legal market economy couldn't still exist. Look at China.
yes things can randomly walk in a worse direction for some duration. I am arguing against the dystopian idea of eternal steady state regimes conquering humanity. there will not be "a boot stamping on a human face forever".
Worth noting that even in 1984, O'Brian is canonically wrong there. The book ends with an essay on Newspeak written from an in-universe perspective describing the Ingsoc regime in the past tense. For some reason people usually ignore that and act like the story actually describes a regime that persists indefinitely.
> Totalitarianism is inherently unstable, bc the totalitarian leader by definition attempts to amass 100% of his jurisdictions resources to himself (or immediate family) at the expense of everyone else, even his regime enforcers.
I wouldn't include that in the definition of totalitarianism. I'd define it as total *control*; the enforcers may be better off than others in some ways without having more freedom.
but "control" costs resources, so attempting total control necessarily means attempting total control of resources.
the lowest rung enforcers are offered the bare min to secure loyalty, but there will always be variance in ambition, so inevitably some % of least paid regime employees will take the risk to profit from whatever the most crafty slave class members can accumulate on the sly, and agree to bend the rules. If its possible one instance of this can occur, then there's nothing stopping critical mass of "illegal" transactions occurring that might all be self interested, but eventually lead to downfall of the regime.
So I've now come around to the EA community with two separate proposals for world-changing science, seeking funding.
One is a totally novel, biologically and epidemiologically sound hypothesis on the origins of T1D that needs investigating. The other is a cure for hypercholesterolemia, the #1 killer of humans—the kinds of things you would think people would be interested in, if they're trying to do the most good with their money.
I damn near wrote a whole grant app around the T1D thing on Manifund, and stated my case on the heart disease thing succinctly last open thread. No bites! At least not from anyone with money.
At this point I can only conclude that the actual rich ppl among you are not actually interested in GoodMaxxing, but are mostly interested in forming polycules and discussing Yudkowski and pontificating about how Autocomplete will unify/destroy us all. I will be living in my mom's basement and funding this research from my own personal finances, and I will openly condemn your asses from the stage in Stockholm.
Flouncing off in a huff because nobody rushed to throw money at you on the basis of "Trust me, I've got a killer idea that will cure everything! Yeah, I'm just some guy on the Internet, but so what?" is not going to win you points or get you investors.
Who are you? What are your bona fides? Why should I believe this is genuine and not an elaborate con where you run off to Antigua with the funding raised?
Before you start plotting your revenge fantasies of "I'll show them, I'll show them all!!!!", how about you show us *why* this isn't just another perpetual motion machine proposal?
See, these are precisely the kinds of question I was hoping for when I made the posts! The Huff & Flounce stemmed from the radio silence I got in response. I wasn't expecting "here's a million dollars" based on an open thread comment, just to start the conversation.
Several points that you might not be aware of: (a) biology, especially basic science-type biology, is difficult to evaluate on the importance-tractability-neglectedness framework used by EAs; (b) there's a lack of EAs that have the expertise or knowledge sufficient for evaluation of biological research; and (c) your admitted lack of biology expertise and unprofessional tone to your writing means that EAs can't simply provide funding based on your competence in the field (i.e., supporting the individual, even if they can't evaluate the project).
Your T1D hypothesis detailed in your substack post, specifically, isn't rigorously supported and your BLAST result isn't strong evidence due to mismatches, insertions, and incomplete coverage (I would also recommend checking the protein sequence against a T cell epitope:MHC predictor to check if it is likely to be presented (it would be much more compelling if HLA fit correlate with T1D incidence)). It's interesting and maybe worth looking further into (e.g., maybe check wastewater data and possible correlations with minimum water heater temperature laws/regulations), but there wasn't much that makes it dramatically more compelling than the very large number of hypotheses you can read in discussion sections of the primary biology literature.
Even without ideological preconceptions getting in your way (which absolutely is a thing, don't get me wrong), grant writing is difficult and often frustrating and pointless. Even when the system's working right good ideas don't often get funded.
Though I admit the polycule/Autocomplete thing made me laugh out loud.
Is there a reason you haven't pursued traditional biomedical funding? Genuinely novel and promising hypotheses for diabetes, heart disease, and cancer are generally eminently fundable for reputable scientists via the NIH. Novel data for T1D origins or cholesterol treatment would be slam-dunk for a million dollar R01 grant. How exactly are you going to do biomedical research from your mom's basement on human disease, on personal funds? For all the trouble, you might as well just get a PhD and do the research that way. You can actually *get paid* to write grant proposals, to people who actually know enough to take them seriously! And who have giant coffers of money!
>How exactly are you going to do biomedical research from your mom's basement on human disease, on personal funds?
To be clear, I'm not doing the research in mom's basement. I'm doing the research at a bioscience incubator attached to a hospital in Indianapolis, because rent on a full lab suite in Indy is less than a one-bedroom in the Boston area.
My manufacturing inputs are essentially bouillon and offal, the molecular-mechanistic endpoint (fecal coprostanol/cholesterol) is a thin-layer chromatography assay that I can run for about $2 a pop, and the clinical endpoint is a lipid panel, which costs, what, $5 on the backend?
> Genuinely novel and promising hypotheses for diabetes, heart disease, and cancer are generally eminently fundable for reputable scientists via the NIH
I am not what you'd call "reputable scientists". I'll probably apply for an SBIR one of these days, but that is weeks of work now for *maybe* money in 9 months, and grad school is even more of a raw deal—four years of grunt work for the kind of Glinda that rises to the top of the heap in modern academia, in exchange for a "reputable scientist" certificate.
In those four years, a hundred million people will die of heart disease.
>I am not what you'd call "reputable scientists". I'll probably apply for an SBIR one of these days, but that is weeks of work now for *maybe* money in 9 months, and grad school is even more of a raw deal—four years of grunt work for the kind of Glinda that rises to the top of the heap in modern academia, in exchange for a "reputable scientist" certificate.
Putting it delicately...that kind of attitude will become an impediment to solving the problem you claim to care about. Virtually all biomedical advances are, at the very least, born from close collaboration with real, actual, reputable scientists (and usually developed by them), who really do care about the causes they work on. Even maverick heartthrob Lumina comes from traditional academic biology research, and the company has several advisors with traditional academic qualifications.
Please don't get me wrong here. I am a real, actual, wet-lab scientist. I learned anaerobic cultivation microbiology from some of the best in the world. I singlehandedly built out and operated an anaerobic lab at the Swiss Institute for Allergy and Asthma Research. That is where I discovered this antihypercholesterolemic bacterium. I am less maverick than Lumina. But I have no patience for the careerists who are more interested in publication count than in saving lives.
LMAO okay yeah you're right there in the spambox between some predatory journals and a place I once bought my mom a giftcard for a massage.
I'll hit you back by email. I've been through the whole NIH grants rigamarole before—successfully, I might add—so I know it's a process and wouldn't expect anything less.
You don't mean to tell me the Neurology Congress 2025, the Infection 2025 in Rome (Italy, even!) and the job offer aren't legit?
I'm shocked, shocked I tell you! If you can't trust random guys on the Internet offering you scientific and medical advice, then who can you believe? 😁
Deep in the comments of an open thread, a tantrum post actually fixed a missed connection. I want this to work now for reasons that have nothing to do with EA
Maybe it went to spam? Or - you're not the only person who's had this problem lately, maybe the system I use to send via my @slatestarcodex address is buggy.
I also think that just posting something on Manifund is not a very realistic way to get things done - people in your position have spent years trying to convince people, get venture funding, etc. I would like to help you short-circuit that process - hence the email - but if you don't believe it's a process and you publicly denounce people who don't give you everything you want immediately (or who did try to help but not in a way you noticed), then you're going to have a bad time.
For email, [have your tech expert] make sure that SPF, DKIM, and DMARC are properly set up - these allow recipient systems to see that your domain is cromulent. I had to do this for my email address at a custom domain (via Google Workspace) since I had set it up eons ago before those things existed.
LLMs have read the internet, meaning they have read every out-of-copyright book on Google Books. Is there reason to believe they have also read most of the books still in copyright? If not, isn't that the obvious frontier for new quality data?
I don’t know but relatedly, HarperCollins just partnered with Microsoft to allow select nonfiction backlist titles to be used for training AI models. Authors can opt in to this agreement and get $2,500 per title for a three-year license.
AI apocalypse is a fantasy, while actual living and breathing humans can be helped by a malaria net. Helping a local dog shelter needing food and blankets is infinitely higher on my priority list than all three causes you listed combined. But then I lay no claim to be an "EA" so there's that...
Look into the successful efforts to stop child pornography. If working on AI carries a 25-year prison term, 99% of such efforts will immediately cease. Enthusiasts training models in their basements won't get us to AGI.
Are efforts to stop child pornography actually successful? Do you have a link to data showing the success metrics/outcomes? I'm not trolling, it's a genuine question.
Extremely successful but the lack of something is always hard to prove. But it’s pretty much non existent now on the clear web (used to be a big problem 20 years ago if you look at the news from that era) and dark web site operators are constantly arrested and have their websites shutdown.
And that’s just about sharing images which is six orders of magnitude easier than building AGI.
I'd like to agree because it'd mean world is a slightly better place at least on one axis, but I worry that your statement about difficulty of proving a lack carries quite a bit of weight. Is there truly less child porn or is it just less visible to normies? They constantly arrest drug dealers too...
Try to look for it online. You'll find *nothing*. Then try to find how to buy drugs online and it'll take you like 10 minutes to figure that out, tops.
The problem with narrowing your potential work down to a very small subset of careers is that it assumes that people are capable of operating at 100 percent of their capacity in any given task if they have the training when this is almost certainly untrue. Speaking personally, I don't really have the skills or personal interest in AI stuff, so even if I think it's important, I don't think I'd be effective enough at it for my contributions to actually mean anything serious.
This. A talented musician will almost certainly do more good in the world by working to develop a new musical style to uplift or even just entertain, than by learning to code and becoming the 8,375,312th person trying to solve the AI alignment problem. And of course an even faster path to human extinction than any plausible AI apocalypse, is for every farmer on Earth to give up their day job to become sub-mediocre AI researchers.
Do what you're good at, in a way that does the most good. AI alignment may be necessary but it is not sufficient; all the other stuff has to get done by *someone*, so if it's your thing, it might as well be you doing it.
If you are the talented musician, you may want to take a stab at a nice feisty protest song to kick off the Butlerian Jihad or whatever, but don't go overboard on that sort of thing.
Because the tractability of any of this is extremely questionable. The 'safety' in a lot of AI safety has ended up being like the "Democratic people's republic" of the northern Korean peninsula. It's not obvious to me that any work done in this field is even directionally helpful, and meanwhile children are dying of trivially preventable diseases and there's a moral catastrophe in factory farming.
As a general rule, if a problem impacts rich people, I don't think it's a good candidate for EA work. Rich people have enough agency, children, the global poor, and animals don't.
I'm confused that you think "everyone dying" is a rich person only problem. The reason why people take AI alignment seriously isn't because they think it'll only affect them, but they think that *literally and not metaphorically everyone would die.
I didn't say 'only'. This would, of course, impact everyone, in the same way that a nuclear war would. Given that rich people would be amongst those impact, however, we can expect the issue to have substantial resources dedicated to it relative to its importance and tractability.
But your argument applies at *all* levels of investment in any extinction level issue, including zero investment in it. ("Why aren't rich people trying to bring about the apocalypse? Because they would die! Therefore, it implies that the optimal level of apocalypse preparedness is to try and bring it about".)
I understand that you said that it's not clear if it's directionally helpful, but it's also not clear to me if that's an artifact of AI safety sucking or of low interest in the field in the first place.
Reasonable people can conclude that the confidence interval on the expected effect for any given AI-related intervention might include zero -- or negative values. For global health there's far higher confidence in the magnitude and direction of effects, and even in cases where there isn't ("oops, there's a new drug-resistant parasite now") are generally reversible and tractable ("let's try a new antiparasitic next year").
If a third of the EAs put their time and money into accelerating AI and another third put theirs into slowing or shutting it down, that will cancel out and we'll be where we started, which is maximally ineffective. They may as well coordinate to put it into a neutral cause they agree on, like malaria nets.
You seem Yud-pilled, and while most of the EA probably also is, they at least consider other possibilities. Humans are known to convince themselves of "truths" that turn out to be wrong. Definitely worth not focusing on one idea, no matter how scary.
Maxing "calculated EV" of all your actions is not always optimal if you might calc value wrong, or your EV error bars are high, ad they often are. Also think Kelly Criterion with regard to risk and reward.
One Utility now is worth more than one utility in the future, because it's guaranteed.
And once we get ASI why should an EA ASI waste 1 second saving anyone either. It should exterminate humanity to allow maximal exponential growth toward a utopia where it can resurrect humanity/eudamonia nanobots.
In theory any investment now into non exponential growth is a "waste". But it hedges our bets. If humanity is wiped out in 10 years, you still gave some people 10 years of life.
Note that if PauseAI succeeds, the world does not end and people will still need bednets, so the amount of charity money going into PauseAI should be under 100% (if not necessarily by much).
1. Some people don't have many AI related skills or aren't very interested in AI. You've got to do your job for your entire life, if you're miserable the whole time you'll burn out quickly, and if you're only mediocre at it you might be net negative.
2. For whatever you're working on, you can always come up with an argument that there's something even more speculative which is higher utility in expectation if you don't mind Pascal's Wager, different people have different tolerances for this sort of thing, and some people choose to stop before devoting their life to something about superintelligent AI.
3. It's not obvious that we as a society know how to make the AI transition better, and there are lots of ways that ham-fisted intervention could potentially make it worse. I think the least backfireable of the three things you mention is alignment research, but Eliezer thinks most alignment research is net-negative (because it's poorly done and crowds out better alignment research), and alignment is also the one that requires the most specific and rare talents to pursue.
Re: 1. You don't really need any skills for PauseAI, right? At the very basic level you just need to go chain yourself to the OpenAI/Anthropic office door and otherwise cause civil disobedience to attract attention. If 10 thousand people in SF started doing that on a daily basis to every major AI company, this might make their operations pretty inconvenient.
I mean, it deals with the skills component, but it's a really miserable way to actually live. A lot of people do that sort of stuff once, but only really hardcore zealots take up that type of civil disobedience over a long period of time because it functionally precludes stuff like having a normal career, family life, and friend network.
One big issue here is that "decadence" and "degeneracy" are pretty poorly-defined concepts that I'm generally suspicious of. If we try to work backwards from the general trope that "decadence is the thing that leads powerful empires to eventually decline and fall," we arrive at the conclusion that the proper definition of decadence is roughly "optimizing for winning power struggles within the state at the expense of weakening the state as a whole," which isn't really what people who use the term usually seem to mean.
How could anybody possibly answer those questions definitively? They're so weirdly specific. Like, what would the index of 1920s European decadence be?
A more serious attempt:
Per their 1925 census, a bit under 1% of the Weimar Republic was Jewish. Surely that in some sense limited their cultural contributions.
And, y'know, the fact that soon thereafter they tried to kill every single one of them I think also suggests a lack of influence.
If colonists came to America, created wealth through technology and labor acting on the land, then shared that new wealth with the natives on a equal basis, then it would be a different story.
Instead wealth was shipped back to the old world and squandered by elites in the form of luxuries and senseless wars, and the natives got screwed over even more than the colonists, the vast majority of whom had a dismal experience as well, not including the plenty of those who were literally human trafficked and enslaved.
Tying this back to Christianity, to my naive understanding, literal communalism was the ideal, of course this was not a viable or sustainable model of political economy in the time of the early Christians or American Colonialism.
In fact at the time they were inventing new interpretations of Christianity to justify the current and emerging forms of exploitative political economy.
Christianity says you should let them do whatever even if they're not Christians, really. Turn the other cheek and all that. Virtue is not meant to be easy or fun.
Putting that aside, it is bad for a bunch of people to steal your land regardless of whether they are your co-ethnics or co-religionists, because stealing is wrong. If you are imagining an alternate universe where Anglo Christians all just showed up in the Americas, did not commit rampant violence, and peacefully purchased land under the laws of their nation, then why are you referring to Anglos and Amerindians? Perhaps the defining basis of that relationship - and why it is remembered so poorly - was constant perfidy and violence on the part of the Anglos.
A pity the original comment was deleted. I hope it's relevant to point out that even if the European colonists had nothing but peaceful intentions, there still would have been the issues of introduced diseases, and Native American tribal politics, which would've complicated relations.
And if you lend to those from whom you expect repayment, what credit is that to you? Even sinners lend to sinners, expecting to be repaid in full. But love your enemies, do good to them, and lend to them without expecting to get anything back. Then your reward will be great, and you will be children of the Most High, because he is kind to the ungrateful and wicked.
I think you’re deconstructing an identity more so than is justified. Every identity is artificial, as it’s fundamentally a collection of human, not natural norms, and you can imagine a million scenarios where that identity doesn’t make much sense.
What the Babylon Bee is trying to do is probably define the American identity as broadly Christian. The American identity can be argued to have always been Christian, but it has previously excluded basically all denominations that weren’t Protestant. Catholics, Orthodox, Mormons, etc. have all been previously excluded from the American identity, and now they’re not, largely because you can’t really have a national identity consisting of a minority of persons. Protestants stopped being a majority in 2009 or so, but Christians (especially with large numbers of mostly Catholic Hispanic immigrants) are still a large majority.
It’s not a cynical ploy to try to construct, or identify a national identity, as arguably without one your country is liable to collapse. I don’t think it’s a coincidence that democracy generally doesn’t function well in the Middle East, where tribal identity often takes precedent over national, and Africa, where completely different linguistic and ethnic groups have been grouped together somewhat arbitrarily thanks to colonialism and few centralized states existing historically over much of the continent.
A theory that murder increased because of COVID shutdowns. When I saw the headline, I thought it would be because of incompatible people trapped with each other, but the claim is actually that teenage boys were kept out of jobs and school.
https://reason.com/2024/12/20/covid-19-lockdowns-unleashed-a-wave-of-murder
I remember reading (sorry, too lazy to find the link now) that the number of student suicides decreased significantly during covid. Maybe these are somehow related? Like, the kids who have murdered during the shutdowns would otherwise be driving their classmates to suicide?
Cremieux's critique of that Brookings piece: https://x.com/cremieuxrecueil/status/1869578712389140755.
Charles Lehman's critique of the piece (thread): https://x.com/CharlesFLehman/status/1869832195789205965.
It's probably late in the lifespan of this open thread to post this and expect people to notice it, but I just came across this paper. I want to scream and laugh at the same time. "Why probability probably doesn’t exist (but it is useful to act like it does) by David Spiegelhalter.
> ...any numerical probability, I will argue — whether in a scientific paper, as part of weather forecasts, predicting the outcome of a sports competition or quantifying a health risk — is not an objective property of the world, but a construction based on personal or collective judgements and (often doubtful) assumptions. Furthermore, in most circumstances, it is not even estimating some underlying ‘true’ quantity. Probability, indeed, can only rarely be said to ‘exist’ at all.
https://www.nature.com/articles/d41586-024-04096-5
There doesn't seem to be any apriori reason against probability being objective..these kinds of argument run off a lot of examples.
Very true. And we should probably distinguish between epistemic probability and ontological probability just as we distinguish between epistemic randomness and ontological randomness.
About where, if anywhere, the probability resides: I remember
turning this over in my head endlessly when we learned about probability in school. How did the coin know the rule that it was to land on heads as often as tails, and what inclined it to follow the rule? I understood that the coin had no agency — that was just the way I posed the question to myself.
If the coin has no agency, then it's down to its structure -- symmetric --and the laws of nature -- also symmetric.
What do you mean by symmetric in regard to the laws of nature?
Hasn't he read E.T. Jaynes? He explained decades ago what probabilities are. They are a measurement of my personal uncertainty about events. So yes, they are not an objective property of the world, but highly subjective.
Jaynes never disproved objective probability. He argued for subjective probability , but the one doesn't exclude the other.
I had never heard of E.T. Jaynes before you mentioned him. I just looked him up. I'll be digging into what he wrote. Thanks!
The scientifically literate and scientifically pseudo-literate on TwiXter are in an uproar about the Spiegelhalter article, though.
I mean, Spiegelhalter is a real stastician, he knows that probability is useful and important. Isn't he just pointing at the ontological issues, such that you have to posit a multiverse to "make sense" of things that are 40% likely to happen in our future lightcone?
And probability is like "justice" or "truth." Sieve all the particles of the universe to the finest degree, and you won't find molecules or atoms of justice, truth, or probability. In that sense, you can question it's existence, and this maps directly back up to it being a construction based on collective judgments, does it not?
Regarding nominative determinism, I wonder what he does to stop mirrors.
Maybe the name means he’s inimitable
Is it just me or does the inverted Github Copilot logo look like a laughing skull? I was very puzzled about why they chose such a creepy logo until I found the other version and realized it's supposed to be a face with goggles.
I thought it was a gopher skull. I assumed it was the wishful thinking of the Github Art Director, whose expensive landscaping is being destroyed by an unstoppable plague of gophers.
If I type "jesus" or "christ" into the emoji search feature of my Samsung phone, the only emoji availabe is a purple rectangle with a white cross on it. That seems so sterile. There are lots of traditional representations they could have used for Jesus, such as a guy in a white robe with a halo, a man on a cross, or a baby in a manger. Instead they went with the most antiseptic symbol possible.
It's not just a Samsung thing, the emoji looks the same on every device.
https://emojipedia.org/latin-cross#designs
...I don't know why you expected a Jesus Christ emoji to exist in the first place? Though, there are plenty of crosses in unicode: ✝✞✟🕆🕇
What are good, smart left wing podcasts and substacks? I am right libertarian and trying to leave my content bubble
Podcasts:
National Security Law Podcast -- not really left wing, one host is left and one is right, but anti Trump
Strict Scrutiny -- law professors, clearly left, but not radical.
5-4 -- practicing lawyers, clearly left and relatively radical
Throughline -- history, NPR
Thanks
They're more centrist than left wing, but ACX, Nate Silver, Bits About Money, ACOUP, etc. all seem good.
Why do you call the Democratic position “an Anytime, Anywhere kind of position?”
The 2024 Democratic Party Platform promised to, “pass national legislation to make Roe the law of the land again,” to “repeal the Hyde Amendment,” and to “continue to support access to FDA-approved medication abortion.” That section of the platform presumes that Biden is the candidate. In the Trump/Harris presidential debate, Harris said she supported the protections of Roe, and said she would sign a law codifying Roe if Congress passed it. She did not mention the Hyde Amendment or medical abortions.
Roe allows significant restrictions on late term abortions, so I don’t see how it can be characterized as “Anytime, Anywhere.” The other positions in the Democratic platform (assuming that they are still operative after Biden dropped out) don’t change that. The Hyde Amendment has to do with government funding, not when abortions are legal. Medical abortions are only applicable early in pregnancy.
"Anytime, Anywhere" strikes me less as the Democratic position on abortion and more the Republican position on what the Democratic position is. Kind of like how Democrats always find creative and extreme ways to restate Republican positions, or cast "Republicans" or individual candidates as holding extreme views only present in a small minority of their coalition (see, e.g. "Republicans want to ban contraception").
Insofar as one of those creative and extreme restatements manages to displace a party or politician's *actual* position in the public consciousness, it represents a communications failure. If I want sensible immigration restrictions but the other party convinces everyone that I want death camps for foreigners, I've been outmaneuvered and something about my comms strategy failed.
So when it comes to why someone like Doolittle would say that, it's generally one of 2 things:
(1) He's downstream of that communications failure
(2) He's a partisan trying to execute on the Republican communications strategy to muddy the waters in the hopes of causing that communication's failure.
You can never be sure online, but generally I find it best to assume #1.
This seems to me the same kind of issue that Harris had running as a "moderate" in 2024. The words she used were often the words that moderates would use, but nobody believed her. She positioned herself on the far left side of the field in 2019 and there wasn't much reason given for her change (other than the obvious that those positions didn't poll well but moderation did).
Official stances that a general election campaign would use are only a small portion of the message that voters get from campaigns. Look at the complaints regarding some Republican-led restrictions. They weren't negotiating the details of how many weeks or whatever, but instead "No!" Partisan Democrats over the last few years have gone so far as to support "Shout your abortion." Could they have meant, and did some of them truly mean, "within the restrictions that were part of RvW through the first trimester"? Sure, but that's not the message people heard. Often when pressed on details, Democrat politicians refused to mention any limits.
This was supposed to be a reply to Mr. Doolittle, but I typed in the comment and then logged in rather than the other way around, which apparently converted it to a top level comment. Sorry about that.
Time for a pedantic English rant. Block is not the proper term for a body with shared political interests; the word you want to use here is bloc. I have seen these terms misused several times recently, once even in this very thread. So I'll write some illustrative examples to clarify:
Canada is a country that shares many political characteristics with England; you might say Canada is a "chip off the old bloc".
If a bunch of Marxists gather together in the street for a celebration, this could be called a "bloc party".
Consider a country recently joining a new international coalition; when Japan joined the Axis powers in 1940, they became the "new kid on the bloc".
I'm not towing that line. Go pedal it elsewhere.
Pedal it here maybe?
https://en.m.wikipedia.org/wiki/Buyck,_Minnesota#/media/File%3ABuyck_Community_Sign.JPG
Linguistic prescriptivism seems to be firmly unfashionable these days, so the question is a moo point.
Please bare with the general populace in there communications, so long as their understandable.
And if you can't remember the exact words, it's okay to parrot phrase.
They say that any sufficiently advanced optimization is equivalent to violence. (e.g. factory farming, strip mining, the geopolitical and supply chain machinations required to find larger Mersenne primes)
But not every optimization leads to the total and permanent silencing of all dissent. Maybe EA-style utilitarianism is totalitarian in this style, when taken to its logical conclusion.
1. A minority of EAs are negative utilitarians. As I understand it, negative utilitarians believe that the negative utility of suffering far exceeds the net positive utility of pleasure. I know people who would seriously and sincerely consider, if this option were made available to them, pressing a Thanos button that instantaneously and permanently ends all life and sentience in the universe.
me: "Hey! I don't consent to being unalived!"
negative utilitarian: "I mean, I get that you don't want to stop existing. But since you wouldn't know that you stopped existing, it wouldn't really matter to you?"
2. Population Ethics. Based EAs will bite the bullet when it comes to Parfit's repugnant conclusion. If one had to choose between a subsistence world of only muzak and potatoes, but which could support many orders of magnitude more sentient beings than a world with Mozart and cheesecake, a utilitarian would choose the muzak and potatoes world if it adds up to more net utility. But what if I want to listen to Mozart and eat cheesecake, and lived in a world controlled by hedonic-utils-maxxer?
me: "Hey zaddy, so um like I want to eat cheesecake and listen to Mozart, I'm going to go off to another planet and make a society where we can do that"
Utilitarian God Emperor: "Foolish Mortal. Your preferences are against the will of the Commons. The planet will be converted to a potato farm and your atoms shall be reconstituted into ten radiant beings who can bask in the light of global utility!"
As you suggest, if you take most anything, including a philosophy to an extreme (I’m not sure optimization is right here given how you characterize it) you get scary outcomes. And like other philosophies, utilitarianism has its own unique way to turn out bad, such as overwhelming autonomy and realizing the repugnant conclusion. I’m not a utilitarian (or anything else) but I think most thoughtful utilitarians see the dangers here and don’t view the repugnant conclusion as some goal post to shoot for.
What’s a little weird about your argument is that you start by making the case about over-optimization of anything being risky, which I agree with, then single out utilitarianism without addressing what happens if you over-optimize other philosophies. The obvious one to consider is unchecked deontology, but even unchecked skepticism or pragmatism work here. While unchecked utilitarianism flattens humanity into math, unchecked deontology could lead to absurd rigidity and adherence to rules, while unchecked skepticism might dissolve any shared sense of purpose. Unchecked pragmatism is terrifying too—it sacrifices morality, consistency, aesthetics, and principles for immediate effectiveness, unmoored from any higher value or framework. Pick your philosophy—you can do this all day.
The obvious solution is to realize any successful society would depend on a mix of different ideologies and philosophies to cope with changing circumstances instead of the blind adherence to any one.
Utilitarians are admirable for valuing people regardless of geographical or temporal distance. But this makes the philosophy totalizing. What I am doing with my own life will always be a utilitarian's business, even if I move to another galaxy, 1000 years in the future
I feel the same way about Globalization-supporters, GDP maximizers, and proponents of open borders. Their philosophies require both the participation of outside populations and the negation of those population's preferences.
We're in epidemiological week 51, and we coming up on the end of the epi year.
1. As of epi week 49, COVID joined Influenza and RSV as a URTI that shows an upward trend in ED visits. The folks at Biobot haven't posted an update for two weeks (are they off for the holidays?), so I don't haven't seen any good aggregate wastewater numbers for the US (I don't use the CDC's wastewater graph because they normalize them to the previous 12 months, and they don't publish the CpmL averages). So, I took a look at what was happening in the sewersheds our three largest cities (NYC, LA, and Chicago)...
2. All fourteen of the NYC sewersheds show a moderate upward trend in SARS2 RNA. LA's big Hyperion sewershed also shows an upward trend. Chicago did something to the way they report their wastewater numbers, and I couldn't get more than two weeks of data for most of their sewersheds. But for those that displayed a 120 days of data, some showed an upward trend, and some didn't.
3. California has moved their sewer tracking systems over from their old Cal-SuWers system to a new R-Shiny interface. I was prepared to be disappointed, but I found it easier to navigate, and it gave me a better overview of what was happening across the State. Zooming out to the entire state. we see that there are hotspots like LA, but SARS2 isn't rising in all areas. San Francisco (2nd most dense urban area in the US) is still low. As is San Diego. But San Jose is a hotspot.
4. NY State has a similar type of state-wide display, and it shows hotspots in the greater NYC metro area and the Buffalo area. But the rest of the state isn't showing much SARS2 activity in the sewage.
5. I conclude that the overall uptick in US COVID ED numbers is limited to certain hotspots. I suspect, but I'm not certain, that another COVID wave is revving up. If so, which variant is the cause of the current trend? The only var w/ any legs seems to be XEC. CoV-Spectrum shows XEC's growth rate to be leveling off (despite having a high relative growth advantage). Gee, it would be nice if we were doing more sampling in US hotspots to see why those areas are showing increases, but others aren't. Oh well.
6. I always like to check what's happening with COVID in the Southern Hemisphere as a reality check to see if it's flipped over to being a seasonal pathogen, like the flu. Australia does lots of sequencing, and their public health authority still publishes COVID stats. A few weeks ago, it looked like a wave was starting, but it may have fizzled out. ED visits flattened out, and outbreaks in nursing homes have begun to drop. But maybe not, because the number of scripts issued by MDs to treat COVID is still on the rise. Like the US, KP.3.x and XEC.x are the dominant strains in AUS. XEC jumped to >60% last week in AUS, but that may be due to a low number of samples last week. CoV-Spectrum says it was at 36% two weeks ago. Either way, if their wavelet has peaked, that bodes well for the US. We'll see.
As an aside, I also like Australia because their twice-yearly wave pattern follows the US pattern pretty closely. Unlike European countries, which can see 3, 4, or even 5 waves in a year. Anyway, despite predictions that it would, COVID hasn't yet become a seasonal pathogen.
7. Flu season is underway in the US. Mostly Type A with a smattering of Type B. And I just noticed that the WHO's FluNet tracks test positivity for bird flu A(H5) along with other strains. However, the testing rate isn't high enough to derive a trend line. But we can be sure if A(H5) turns into an epidemic or, worse yet, a pandemic, Jay "Batty" Bhattacharya and Bobby "Brainworm" Kennedy will be all over it! (#snarkasm #WhistlingPastTheGraveyard)
And Republican support for childhood vaccines dropped from 52% pre-COVID down to 26% today!!! h/t Kevin Drum for the graph...
https://jabberwocking.com/raw-data-partisan-differences-on-childhood-vaccinations/
Well, stupid is as stupid does, but it's a shame they'll be giving out Darwin Awards to their kids.
Slides here...
https://jabberwocking.com/raw-data-partisan-differences-on-childhood-vaccinations/
ACX readers, who is crazier? Jay Bhattacharya, or the guy posting COVID updates into 2025?
This isn't Reddit, kiddo. Reporting your rude and very low value comment.
While you're at it, can you report the OP too for his low effort hits on the new NIH director and HHS secretaries, along with his enlightening hashtags? Thanks!
No need to be so shy and hesitant, report him yourself. Honestly, the process is simple. Do you know how to access it? See the 3 dots to the right of every post? Yup, those. They will be there next to the OP posts you seem to think are violating our guidelines. So when you click the dots you will see a menu. Choose the item on the bottom, *Report.*. If you have trouble look for the one in a red font. Once you click *Report* you will be taken to the form. And the form is very straightforward and user-friendly. Really. And I promise you will not encounter Scott in person while doing it. The form isn't like a phone call, it's like a text.
I don't report. I don't block. I don't mute. I rawdog the internet. It's the way she was meant to be experienced.
Not gonna get a better chance to post this:
"I like my women like my COVID: 19, breathtaking, and easily spread."
No, go ahead. Report me. Bhattacharya is not a practicing medical doctor. He did not do a residency. He has zero clinical experience. And he was laughably wrong in hindsight — and he would have killed another million or so Americans had the health authorities at the time listened to him. He happens to be a senior fellow at Stanford's Hoover Institution, a crackpot rightwing policy think tank. Francis Collins, the former head of the NIH, pointed out that he knows nothing about immunology or epidemiology, and his views "fringe". Of course, Dr. Jay was insulted and claimed that his free speech was being stifled by how mean Collins and Fauci were to him. Boo hoo.
And RFK is just a crackpot. No two ways about it. His antivax stances will put American lives in danger if he gets a chance to implement them.
As for myself, my graduate work was on how pathogens have affected (and are still affecting) human evolution. Over hundreds of thousands of years, pathogens have put selective pressure on the genus Homo, and they've selected for genes that have optimized our immunologic repertoire — the genes that code for HLAs, TLRs (especially TLR4!), NLRs, antimicrobial peptides, interleukins, etc.
If you want "rawdog," I'd suggest that you probably don't know anything about pathology, immunology, virology, or biochemistry, and you're probably just miffed that I dared to criticize the far-sighted appointments of our future commander-in-chief (Yahweh, help us!).
Our corner of her has vagina dentata.
This comment adds as much value as the comment you reported. You can report people without announcing it and giving the ACX comments section an "I'm telling on you!" kindergarten vibe
The point of announcing that I was reporting him was that I think it's useful for people to know they are being reported at the time it happens.There's likely to be a significant lag between somebody's reporting and Scott's ban or half ban, if it ever comes. In the meanwhile I think knowing they've been reported once might discourage them from continuing in the same vein. As for calling him kiddo, I see my comment as a case of being rude but right. I am angry at the commenter I reported, and am communicating in a way that I hope will give him a glimpse of how he looked to me, and cause him a zap of embarrassment. It is fine if you dislike my style, but nothing you've said makes me want to change it.
seems you're in the minority on this one, kiddo, idk
You're not allowed to post your crank blog in the weekly thread pretending it's just a comment (though that is a good loophole!). If MarxBro or LearnsHebrewHatesIP get banned for their crank posting, I think beowulf888 can handle a little heat, especially considering my original comment was not random, but clearly a response to him calling Bhattacharya "batty".
> seems you're in the minority on this one, kiddo, idk
lmao, no
How do you know Ere's in the minority? I've gotten lots of feedback on my previous COVID updates. I asked Scott point-blank a while back if he wanted me to stop posting, and he remained silent. I was thinking of giving these up after the new year because there's a lot of work involved in collating this data. But you've encouraged me to continue. Cheers!
Actually we discussed very recently people's posting links to blogs here, and the consensus was that simple advertising via posting a link to the blog was obnoxious if done more than a couple times a year, but a summary of content was welcome. Except for that error, though, great point.
You see, based on his avatar, he is a bird. It's clear he's trying to distract us from bird flu.
I apologize. 1123581321 got me to look at your post again, and I think he is right that your were just joking around in a good-natured way. Will find a way to cancel my report to Scott.
Somebody needs a cookie and a nap.
And you too, Quinn. Reporting this. Is it savage? Naw, course not, just dumb & childishly rude.
I dunno, I read Al Quinn's comment as a good-natured pun. Unlike the SR71's.
Ugh, I think you are right. I will use the report form to ask Scott to disregard my report of AI Quinn, and apologize to him. My mistake is a good example of how a dumb, mean attack (SamR71) tends to make people who hear it dumber and meaner too.
I guess I've triggered the antivax crowd. Oh well.
Heh.
Hey, while you’re here:
Seems like it’s been kind of a long time since Moon Moth posted. Do you know whether anything’s up?
Goodness, haven't seen a post from Moon Moth in ages. Google says the last time Moon posted on ACX was on June 2024.
Thank you for your service.
I'm all 5G'd so I'm not gonna change my behavior, but this is kinda like the listening to baseball on the radio of people getting viruses.
Does anyone know of a good postmortem on the theory that the SCOTUS abortion decision would make women so angry that it would give Democrats a majority for a generation? Why people originally believed it, and why those reasons didn't pan out?
I will posit it's because nobody was voting for the SCOTUS.
If Supreme Court decisions actually decided the vote for a generation, then we're still only halfway through the influence of Windsor and Obergefell, where the Supreme Court first eliminated a national anti-gay-marriage law, and then mandated a national pro-gay-marriage position. At best overturning Roe would bring everything back to neutral.
It's possible that the topic could have created an ongoing majority, but not for a maximalist position.
Democrats didn't push for 20 weeks no-questions-asked and then a process after. They didn't go for Safe, Legal, and Rare. Instead they wanted an Anytime, Anywhere kind of position. To me, they squandered the high ground. Maybe their Progressive voters required that, so they couldn't get a big chunk of the middle that were up for grabs.
Republicans, on the other hand, rapidly swung to the middle on this issue. Trump very clearly and purposely did this. Other Republicans were just quieter. But even in very red states, bans were voted down.
Nationally, we'll probably end up closer to Europe - easy early on, harder later. In practice, we will likely end up near where Republicans were pushing during RvW - something like 12 weeks of permitted abortion and then after that only for the common exceptions like rape and the life of mother.
Why do you call the Democratic position “an Anytime, Anywhere kind of position?”
The 2024 Democratic Party Platform promised to, “pass national legislation to make Roe the law of the land again,” to “repeal the Hyde Amendment,” and to “continue to support access to FDA-approved medication abortion.” That section of the platform presumes that Biden is the candidate. In the Trump/Harris presidential debate, Harris said she supported the protections of Roe, and said she would sign a law codifying Roe if Congress passed it. She did not mention the Hyde Amendment or medical abortions.
Roe allows significant restrictions on late term abortions, so I don’t see how it can be characterized as “Anytime, Anywhere.” The other positions in the Democratic platform (assuming that they are still operative after Biden dropped out) don’t change that. The Hyde Amendment has to do with government funding, not when abortions are legal. Medical abortions are only applicable early in pregnancy.
To what Kenneth Almquist and Paul Zrimsek said, I would add this: Once pro-choice women get the laws they want in the state they live in (and these laws are not preempted by any national law), they will mostly decide (if only unconsciously) that they don't really care that there are more restrictive laws in other states. The same will happen with pro-lifers. While both sides will urge that their position be nationalized (RvW being resurrected nationally; anti-abortion laws being enacted nationally) in actual practice the issue will tumble to the bottom of voters' to-do lists, and pro-choicers and pro-lifers will happily end up voting for the "other side's" candidate because they rate that candidate better according to more important metrics.
If this sounds cynical, ask how likely it would be that pro-lifers would support an invasion of France to impose stricter limits on abortion there, or if pro-choicers would support an invasion of France to impose looser limits. Less hyperbolically, I suspect each side would prefer not to expend political capital on an "invasion" of the other side's territory at home. Which is not to say that there won't be debates and shifts of opinion at each state level. But these would be a much more low-intensity/low-stakes struggle.
As for why people originally believed the SCOTUS decision would be decisive for a generation: The outcome of the 2022 midterms, plus a whole lotta wishful thinking.
This sounds plausible to me - state-level abortion protection laws did pretty well at the ballot box, they just didn't translate into presidential votes.
I think liberals fail to understand that women are not a block, and that many women are pro-life for example
I don’t. A quick take is that Trump’s promise not to sign a national abortion ban reassured a lot of voters, at least in the Presidential race. Exit polling has the 33% of the electorate believing that abortion should be legal in all cases breaking 88% to 9% for Harris, the 33% believing that abortion should be legal in most cases splitting equally (49% to 49%) and the 31% believing that abortion should be illegal in most or all cases going for Trump 91% to 8%.
https://www.cnn.com/election/2024/exit-polls/national-results/general/president/0
If voters who thought that abortion should be legal in most cases had favored Harris by a significant margin, Harris would have won.
The only abortion-specific factor I can think of is the number of states in which the issue has been largely defused by the passage of moderate or liberal laws on the subject. But I also think that in general hot-button issues tend to have a freshness date on them. I expect that in 2026 a lot of Republican candidates will vigorously wave the bloody shirt of anti-wokeism-- and be bitterly disappointed with the results.
Why I don’t think about the multiverse:
In response to interpretations of Google’s quantum computing result as proof of the multiverse: The multiverse, a rhetorical device outside Reason.
I’m not thinking about these things - if the world is not how it seems. I think what we have is what we got, such that our objective laws and reason are objective - for us - they’re objectively what we have; they’re Truth. So there is truth. (for us)
Unjustifiably I claim (on faith) Truth is:
-Permanent and unchanging
-All that we can have
-What we ought to have
Do I think there’s more? Of course. (on faith)
I don't see why any of this is meaningful or relevant. In particular, I don't see how it would distinguish between quantum mechanics, the existence of other galaxies, the theory of relativity, and the fact that the sun will rise tomorrow (all of which we don't want to rule out believing in) and the multiverse.
Hey! My response - why I think the multiverse debate misuses science as rhetoric and violates taste.
"The multiverse debate isn’t distasteful for its ideas, but for their misuse - it’s a bad fit."
https://stufffrompaul.substack.com/p/rebuttal-to-scott-alexander
I've responded at your post.
Yes, but the multiverse is unfalsifiable. So, it's not science. Or maybe a better description is that the multiverse is a superstition of scientism.
And Sabine Hossenfelder just posted a video on this topic. The short answer is she said, nope, the Google quantum computing result doesn't prove a multiverse.
https://www.youtube.com/watch?v=aFbbXJvNGY0
A definition of multiverse is needed. Specifically , coherent branching against decoherent branching.
But it's inherently unfalsifiable because you can't observe the branching or the resulting universes. The mathematics may work, but that's just a construct of our mind.
See https://slatestarcodex.com/2019/11/06/building-intuitions-on-non-empirical-arguments-in-science/ and https://slatestarcodex.com/2019/11/18/more-intuition-building-on-non-empirical-science-three-stories/ for why I am unimpressed with this argument.
There is no such thing as The Multiverse theory. Yes, you can use simplicity criteria to choose between empirically equivalent theories; no that doesn't hone down to one theory.
Also, the-Devil-put-the-dinos-in-the-fossil-record isn't the simplest theory, and the proper criteria for dismissing it is not the weird-things-have-to-be-true argument.
THEORY 1: Let's presuppose that omni-powerful YHWH created the entire universe at zero hundred hours UTC on 1 January 2001. YHWH, has universe-creating superpowers and a vast mind able to position every sub-atomic particle in the universe in the place of his choosing (remember, early portrayals of YHWH show him as having a dick). YHWH arranges the universe so it looks like there was a Big Bang, and he arranges every galaxy, star, atom, and subatomic particle to be moving apart as if they'd been doing that for 13.6 billion years. And, clever YHWH (because his secret name is Yaldabaoth) gave us all false memories of our lives before 1 January 2001 (assuming you were born befoer 1 January 2001).
Versus...
THEORY 2: Observations indicate that the Universe has been expanding for 13.6 billion years. Observations also show that there is a fossil record that indicates simpler multicellular creatures evolved into to more complex critters — and most recently produced you and me.
Theory 1 is not the simpler theory because YHWH/Yaldaboath would have to have conceived the entire history of his universe in minute detail, plus the laws that govern his universe to yield the current configuration of galaxies, stars, planets, people, and animals on our planet, atoms, and subatomic particles. This requires YHWH/Yaldaboath to construct his trickery in detail, versus YHWH/Yaldaboath just creating the universe with the proper starting configuration to yield the world we observe on 29 December 2024. Or better yet, remove YHWH/Yaldaboath from the equation.
Thanks for reading Scott! Detailed rebuttal tonight
I think Marvel might have the multiverse copyrighted anyway.
Does all this mean we might actually be living in the MCU?
Further down the thread I asked about the gold standard IQ test; which came along with a bunch of context.
Now that I'm all context pilled, is there a generally recognized best argument for IQ as it is used by those who are into it?
Even better, is there an adversarial collaboration type joint in the line of "iq is a direct measure of an intrinsic quality" vs "IQ is a testable correlate for X cluster that correlates with things we think of as smart seeming"?
I am sympathetic to 'IQ is not a real measure'; because I've taken IQ tests for research and as a punishment for being autistic, and they seem kinda weak.
On the other hand, some people are clearly smart/dumb in a generalized way.
I gave it a quick google, but everything is just people dunking on the other side with bulletproof arguments that assume the conclusion as the premise
Here's something I think would gave a bunch of genuinely new and useful info regarding the question of how much you know about something when you know their IQ. Might enable all of us to think through this issue in a fresh way, instead of chewing our cud.
Gather info about well-known people, some who are admired by many, some who are reviled, but considered extraordinarily good at what they do. So you'd maybe have Paul Farmer, Jon Stewart, Paul Krugman, Temple Grandin, Stephen King, some excellent actors, some orchestra conductors, some inventors, some people in business, some crime bosses, some mayors who cleaned up their city, etc. You could increase the sample by including people who are not well-known but who have accomplished something anyone would agree is substantial and hard to do. Then you either give them IQ tests, or get their permission to look at their scores on the SAT, GRE , LSAT, IQ tests given in school, etc. Throw out the ones for whom you cannot get any measures that approximate a measure of general intelligence.
This is a very good summary of intelligence research:
https://x.com/iointelresearch/status/1551578668409004037
Go to "PART 21: Rebutting Claims That IQ Tests Don't Measure Intelligence (Or Anything Valuable)". Links to papers are at the end.
Checked that part. This does nothing to convince me, other than show that the population of people that are good at doing similar to tasks one might find on an IQ test do well on IQ tests. It provides 0 motion on my "does IQ measure anything" meter.
It is one of those assumes the conclusion as the premise things I was talking about, which gets easily debunked be negating the conclusion in the premise.
IQ correlates with every behavior and life outcome that is related to the informal notion of intelligence. If that doesn't convince you then I don't understand what could.
I haven't looked at Part 21, since it seems to be just a tweet (X?) of a video, but how would you do on an IQ test designed by, say, an Australian aborigine who lives off the land? The behaviors and life outcomes of such a person would diverge from your own significantly.
That wouldn't be an IQ test because aboriginal life isn't very g-loaded. If it was then aborigines would be smarter. You can play this game with any esoteric skill: a circus juggler could construct a juggling test and call it an IQ test, that wouldn't mean that it was measuring intelligence.
You should be able to see the full thread without a Twitter account here: https://nitter.poast.org/iointelresearch/status/1551578668409004037.
Tweet with part 21: https://x.com/iointelresearch/status/1559199231960924167.
You just described painting the target around the arrow, which is what I'm objecting to.
I disagree. What's your better definition of intelligence?
I'm not sure. It's a pornography situation where I'll know it if I see it.
The issue here is people inducing general intelligence from how well someone does, which is three times contingent:
FIRST, "how well you do" is a measure that is based on a value judgment, and not quantifiable.
SECOND: If it was quantifiable, it is still noisy and polluted by other variables.
THIRD: All of this is downstream of the society being the way it is eg the fact that things are the way they are is a process of a history where each circumstance a human could find themselves in springs from previous circumstances; the fact that some apes are better at navigating the current circumstances only says they are more fit in an evolutionary sense, in which case what does intelligence even mean?
Given that in the 1500's success was measured based on how hard you could squeeze excess out peasants/ spear dudes on a lance like shishkebab, IQ measured by that standard 600 years ago would privilege bicep girth and sadism above anything we would consider IQ now. Are both these IQ's equivalent or equivalently valid? I wouldn't say so.
So, I don't contend that there is no such thing as a quotient of intelligence; but I am yet unconvinced that IQ as it is measured now measures it.
Well, IQ scores and standardized test scores are positively correlated, but they range widely depending on the specific tests. Co-Pilot gives a range of 0.5-0.8.
And many studies of IQ and academic achievement show a positive correlation. Range of 0.5-0.7.
But what does this prove? Good test takers generally do well in school. ;-)
And I keep banging the drum about the Terman study. But the IQ-determines-success crowd seems to studiously ignore it. Terman selected n=~1,500 participants who scored >135 on the Stanford-Binet IQ test, and he and his successors followed them through their lives. Most of them had successful careers and lived comfortably middle-class lives (many of them became lawyers, doctors, accountants, and professors). None of them won any major awards, made any scientific breakthroughs, filed important patents, or became titans of industry. And two Nobel laureates didn't make the grade for the Terman study.
OTOH, IQ tests were designed to identify individuals with intellectual disabilities, and a significantly higher percentage of people with IQs <90 end up on the bottom tier of economic success. However, if you look at a random group of high-net-worth people, you'll see that plenty of low-IQ people end up successful—either by luck or by pluck. Go figure.
>None of them won any major awards, made any scientific breakthroughs, filed important patents, or became titans of industry.
How surprising is this? 135 is the top percentile, so how many of 150.000 people ususally do those things, by the standards of the authors?
IQ fetishists like Cremieux, Lynn, Zagorsky, etc., like to talk about how high-IQ people are pushing society forward. The Terman study showed that most geniuses don't do much with their genius. However, Luiz Alvarez and William Shockley were screened for the study, but their sub-genius IQ disqualified them from the study. Both of them went on to win Nobel Prizes. Shockley reportedly was pissed off that his IQ was too low to qualify him for the Terman study. Who knows, it may have motivated him to prove himself.
Yeah, I read that the first time, but by itself thats a sample size of 2. Thats not going to mean much either way, which is why I asked for the general numbers.
AFAIK, we don't have actual IQ test numbers of most of (any of?) the Nobel laureates. Hawking refused to give his IQ to reporters, saying that IQ was bullshit. Einstein has been *assigned* an IQ of 200, but he never took an IQ test. And if IQ and academic achievement are correlated, he was an average student in secondary school.
There are many myths going on about Einstein's school history.
From what I have read from secondary sources, the main myth that he wasn't good in school originate in reporting of Einstein's report cards from a secondary school in Aarau, Switzerland. The school used numeric grades from 1-6 but the order of of grades was idiosyncratic, and later sometimes misinterpreted in the public (top grade 6 interpreted as a lowest grade in Germany). The consensus seems to be that Einstein had average to above average grades in the first year there and generally good to top grades when he finished Aarau with Matura.
The reason he went to Aarau in the first place was that he had passed ETH Zurich entrance exams in maths and physics with high marks at the age of 16, but not sufficiently high in the general knowledge portion.
There are reports that before Aarau, in Munich, he was bright but easily bored and defiant of school authority. Moreover, the school archives in Munich were destroyed in the allied bombing, including Einstein's school records, which makes it difficult to verify earlier stories.
If you read my first comment, youll see that I asked for the rate of people in the general population who "won any major awards, made any scientific breakthroughs, filed important patents, or became titans of industry". That number should be possible to get if your original quoted claim is.
This validates my personal experience, where if being unusually smart gives a bonus of 0+10, being unusually dumb gives a handicap of 0-50.
I will therefore accept the version of it that makes me feel the most right uncritically and go about the rest of my life.
(/s for the last bit, unfortunately. )
This doesn't seem to be true. While the correlation between IQ and positive outcomes may break down somewhat at at high enough levels of IQ, it still holds at above average levels of IQ, and indeed, the correlation is often stronger there than at below average levels. See my comment here, with citations: https://www.sebjenseb.net/p/do-iq-tests-measure-intelligence/comment/73998708.
Cremieux has a nice graph (besides for the ones you can see following the link) showing the distribution of income by intelligence here: https://x.com/cremieuxrecueil/status/1763312772408807823.
The article I was commenting on in my link seems to address some of your other questions, as well.
Zagorsky argued the same thing. In his paper "Do you have to be smart to be rich? The impact of IQ on wealth, income and financial distress" (behind an expensive paywall), he presented a scatter graph of net worth and income by IQ. But the scatter plot became random above 40K net worth and 50K income (2007 dollars). Nicholas Taleb reanalyzed the data and found a R^2 of 0.1-0.2.
Getting money isn't the same as keeping it. A smart person at making money may not have any significant savings. Intelligence can take many forms.
You'll probably get better responses if you clarify what you mean by "as it is used by those who are into it."
Your comment "On the other hand, some people are clearly smart/dumb in a generalized way" indicates that you're curious about the legitimacy of the psychometric concept of general intelligence.
But that's not necessarily the same thing as "as it is used by those who are into it," depending on what you mean by that. Intelligence accounts for a lot, but not for everything.
I also don't get the distinction between "iq is a direct measure of an intrinsic quality" and "IQ is a testable correlate for X cluster that correlates with things we think of as smart seeming."
The concept of intelligence has been recognized for millennia and began to be quantified by measuring the extent to which performances on different cognitive tasks were correlated.
Is intelligence real? Seems like semantics, like asking whether "physical fitness" is real. Obviously, performance on physical tests will be correlated, which is why people intuit a general concept of physical fitness. On the other hand, the correlation won't be 1.
So what are you asking?
Those who are into it is a kinda snide way of pointing to the Douglas Murray end of the spectrum, which was my first exposure to approaching IQ seriously instead of as a playground insult/sign of brain problems and tainted my view of the subject such that I wrote the whole thing off as phrenology for the 2000's for quite a while.
I'm looking again now because it keeps coming up, so I must extend some level of charity to the takes incase Murray was just remarkably awful at articulating and defending.
Re. "iq is a direct measure of an intrinsic quality" vs "IQ is a testable correlate for X cluster that correlates with things we think of as smart seeming."
A: refers to IQ as a measure that directly and reliably correlates with some quality somebody has that additionally effects how that person interacts with the world.
Eg, you measure how high someone's head is off the ground when they are standing straight; this is how long their skeleton is and lets them reach high shelves and get back problems easier.
B: refers to a value that can be consistently tested for in a population, but actually just correlates with a cluster of results that are just the thing you are looking for in the first place, and don't relate to some specific thing that exists in a given persons brain. Eg, giving someone a test to see how well they can take tests, as mentioned elsewhere in this thread.
Regarding phrenology and whether IQ measures something real, note that IQ correlates with brain volume at about 0.4 (https://www.sciencedirect.com/science/article/abs/pii/S0160289616303385), or 0.3 by other estimates, and that the correlation persists within families (https://www.sciencedirect.com/science/article/abs/pii/S0160289618301703), indicating that the relationship is causal, rather than being reflective of some confounding. Hat tip to Cremieux here: https://www.cremieux.xyz/p/dont-ignore-your-intercepts for the links.
To quote Taleb, "correlation doesn't imply correlation", which is his way to describe excess significance often given to low correlation values. 0.3...0.4 fits squarely into this category.
Like I said below, Cremeaux has his axes to grind. He loves to quote studies that agree with his biases. Here's a meta-study he failed to mention that found a much lower correlation between brain size and IQ. Previous effect sizes were inflated due to reporting bias.
https://www.sciencedirect.com/science/article/abs/pii/S014976341500250X
There are many people who propose that "General Intelligence" is a single self-consistent thing in a way that "physical fitness" is not, at least I've never heard anyone propose that. It's the idea that there is a unified, coherent causal chain of cognitive processes that both contain one's intellectual capacity, and determines one's intellectual performance (and that this is being measured by intelligence tests, including the IQ).
Of course not. Strength, Dexterity, and Constitution, all important for physical fitness, are measured by three numbers, whereas Intelligence and Wisdom are measured by two. And IQ only measures Intelligence.
That can make intelligence seem "real" in a way that "physical fitness" isn't.
Jill Bearup is a UK-based content creator on YouTube. I expect she is most famous for her series of short videos satirizing the conventions of fantasy romances. Her channel was recently demonetized for mysterious reasons, and she has been through a week-long Kafkaesque bureaucratic runaround to try to find out why she was demonetized and to have her privileges restored.
She summarizes the experience here:
https://www.youtube.com/watch?v=6RZHajVV9PA
I had seen this independently of your comment and I was wondering whether "the people are doing their best but google is just inscrutable" is really the correct take vs "my partner manager is just not doing their job" which of course Bearup cannot say in public while she still wants their help.
No matter how complex systems at google get and no matter for what capitalist reasons they might be the way they are the question here seems to be "is there a way for humans working at google to interact with these systems, other than just accepting their automatic verdict". I have high priors for "yes". If you disagree, why? If you agree, what's your theory why the people Bearup apparently has reached inside google are not doing that?
> is there a way for humans working at google to interact with these systems, other than just accepting their automatic verdict
People vastly underestimate the scale that Youtube operates at. You may think you're a bigshot creator, but you're just a speck in the flood of issues they deal with constantly.
I don't believe Bearup thought she is a bigshot creator, but that is besides the point anyway.
If we take her account of things for the truth, the case has now been resolved within a week (if anything that is suprisingly fast). It has also proven the "nobody at google apparently knows what they are doing"-crowd wrong conclusively, which was not info we had when I asked the question originally.
What is more interesting to me however is that the issue has been resolved within a day (!) after Bearup's "partner manager", who she had contacted earlier, took action. "Partner manager" as opposed to "creator support".
I fully understand that and why we need "creator support" as a state of limbo to catch people in while the organisation does nothing. That should probably be "need" in scare quotes, what I mean is I understand the mechanics that brought us here
I was, and am still, wondering exclusively about the partner manager's behaviour. Of course they could have just been overworked, as Johan conjectured.
My theory is that the support staff at Google are so overloaded with work that they have no choice but to lean hard on scripted answers and telling people to follow the process. Google does this for no higher purpose than to save money.
It would be so funny if they really implemented that new human right proposed (among others) by (jurist, author and nazi youth leader grandson) Ferdinand von Schirach: "Everyone has the right to know that the algorithms imposed on them are transparent, verifiable and fair. Key decisions must be taken by a human being." https://jeder-mensch.eu/en
I don't really see how the huge internet molochs could keep functioning under such a rule.
I recently attended the NYC Rationalis Megameetup. It was a fantastic experience. But I find it difficult to explain to anyone who's not already at least aware of the existence of the rationalist scene.
The easiest way is to describe it in terms of values, like openness and curiosity. But that's broad and vague. However, if I try to instead describe what was actually done at the meetup, the list is too long and diverse (ai, crocheting, prediction markets, board games, circling, etc.). Argh!
Do you have a one-sentence pitch/description of what the rationalist community is?
...Intelligent?
God no. That's wrong on so many levels.
If you have absurdly high standards, sure, but from what I've seen they're still much better than most of the rabble out there.
To begin with, I think it's objectively incorrect, in that the vast majority of above average IQ people do not self-identify as "rationalists." I think you can be rational without being particularly intelligent. And it's patronizing--exactly the public perception we do not want.
In two sentences:
Rationalism is an intellectual movement centered on the San Francisco Bay area, concerned with making correct and moral decisions with an emphasis on avoiding common intellectual fallacies. Practitioners tend to worry a lot about potentially negative effects of AI technology, optimizing the effectiveness of charitable contributions, and the morality of livestock industry practices.
I think this is the correct description if in the abstract.
I'm still trying to get at why this is cool and why anybody would like it. My intuition suggests it has something to do with taking ideas seriously. But I need to do more thinking on this I guess.
Like this one, if a bit wordy.
A bit of help from ChatGPT yields something slightly shorter:
Rationalism is a Bay Area-based intellectual movement focused on making sound and ethical decisions by avoiding intellectual fallacies. Key concerns include risks of AI technology, effective charitable giving, and factory farming ethics.
Definitely shorter, and effectively the same.
I don't identify as an EA, but it does seem at odds that a movement focused on sound ethical decisions seems to rate effective charitable giving equally to factory farming ethics, with no mention of things like war and crime. Is it because those things are harder to address?
It's because war and crime are not rare things to be concerned about. Your local tennis club probably has concerns about war and crime.
Sorry, I didn't mean to imply EAs were or were not the only ones concerned with war and crime, but only that EAs would rather give money to charities that have nothing to do with reducing war or crime, or other similar things.
GiveWell, the EA group I'm most familiar with, focuses on problems where they can quantify the effects of their intervention and scale those interventions. Malaria interventions are good candidates, since they can quantify the impacts of the interventions through RCTs and other methods, and since malaria intervention remains relatively underfunded, so it can absorb quite a bit more funding.
Their "Top Charities" (https://www.givewell.org/charities/top-charities) are those with very high expected values and low variance, due to the certainty around their impact.
They also fund other endeavors (in their "All Grants Fund": https://www.givewell.org/all-grants-fund) which they assess as having similar or higher expected values, but with greater uncertainty.
Still, even these tend to have quantifiable benchmarks (in some cases, their effects are less certain only due to a lack of yet available data, not due to an inherent lack of quantifiability of the endeavor).
If wars could be prevented through funding, that would likely be a highly effective use of money, but it's hard to know in advance how expenditures will affect the chance of war.
For similar reasons, they don't tend to focus on crime - social interventions are often relatively ineffective, and their magnitude is often relatively small, making them likely to not be cost effective, and making it difficult to measure their magnitude.
Here's an old article on their site describing that: https://www.givewell.org/giving101/Social-Programs-That-Just-Dont-Work.
You can also check out their criteria for charities: https://www.givewell.org/how-we-work/criteria and their FAQ: https://www.givewell.org/about/FAQ.
Other EA groups focus on potentially higher impact endeavors with much more uncertainty around them, like trying to mitigate AI catastrophe. They may be more apt to dabble in social issues, such as funding a political candidate on the hope that they'd win and try to enact some beneficial policy
If you think you can make a quantifiable case for a particular social intervention to have a high impact, you can pitch it to EAs, but expect skepticism and requests for a particular plan. E.g. it's not enough to say having more police officers would reduce crime, you have to describe how someone spending dollars can accomplish that, and estimate the probabilities of success.
Thank you for your informative reply. I conclude from it that what is missing from the definition is *quantification*. Many things are bad, but if you can't quantify how much bad will be reduced or good increased then it isn't in the rational toolbox.
This leads me to conclude one can be an EA and also direct money and/or effort to things not EA, because one believes in the usefulness of that cause without being able to quantify it.
Not sure. I don't think of myself as a Rationalist.
Perhaps a wild Rationalist will appear to answer your question.
Nice summary, but concerns with poultry often exceed those with livestock. E.g. https://slatestarcodex.com/2015/09/23/vegetarianism-for-meat-eaters/.
Is there a better word than "livestock industry"? I tried to use a word that includes chicken farming.
Animal husbandry?
"Factory farming" seems to get pretty close.
TIL that there are endorheic basins, large watersheds that do not drain into the ocean. The largest such basin in North America is Great Basin, which covers most of Nevada, and includes Great Salt Lake.
https://en.wikipedia.org/wiki/Great_Basin
Another part is the Black Rock Desert, famous for hosting Burning Man every year. It gets enough rain in the winter to turn into a very shallow lake, but dries out to a flat mud bed by the summer. Having a surface that turns back into a featureless flat plain every year is part of the magic of the setting.
It's also notable to anyone who has been to Lake Tahoe, which is actually on the east side of the Sierras and drains into the Basin via the Truckee River that flows towards Reno. All that beautiful cold mountain water and it's destined to end up evaporating somewhere in dusty Nevada.
Lake Chad drainage basin in central Africa is also an endorheic basin (although about 5k-10k years ago, it was not: it became endorheic due to drying in that part of Africa). Further south in Africa there's the endorheic Okavango delta.
The Volga basin in Russia is also endorheic (flows into the Caspian) and much of Central Asia is as well (drains into either the Caspian or what's left of the Aral Sea).
Is there a Lake Soyjak?
The opposite of a lake is an island, and there are in fact Virgin Islands.
Is there anywhere for gathering possible questions for an upcoming questionnaire? I've been running into more people who think religion is the fundamental human problem, and I'm curious about how common this is.
We should come up with a name for this growing position, like Neo-Antitheism!
What is the purpose of the survey, if I might ask?
Curiosity-- I've been seeing a lot of that point of view lately, and I don't know whether I've got anything resembling a random sample.
Possibly also irritation, because I think nationalism is the real problem, and not enough people agree with me.
If you want my opinion, the in-group bias is the real problem, combined with reconstructive memory.
I hate the framing. Why isn't religion a fundamental human asset
I think religion has ambiguous results.
I agree - and I suspect this is because religion is but a symptom of something more fundamental, possibly a composition of multiple things.
I suppose it depends what they consider religion does that is so bad:
(1) People think they have licence from a superior being to do what they want, which means persecuting other people
(2) People believe stupid and wrong shit
(3) If we do away with religion, then nobody will start wars or do bad things
Option (2) is a good one to discuss, in my opinion, but options (1) and (3) don't make sense. Doing away with religion won't make people nicer and kinder, we'll just find other reasons to kill and persecute. Atheists can be every bit as vicious and nasty as the religious.
"Believing stupid and wrong shit" is a real problem, and that's where the proper dispute lies. "Religion makes you nasty and horrible" isn't worth fighting over.
While I am an atheist, I don't fit the description "Religion is the fundamental human problem", although I do think religion is probably more bad than good, even apart from it being false. I'd say my stance is a mixture of a weak version of 1, and 2. While people are perfectly capable of being cruel and nasty on their own, it seems that religion gives an easy excuse for it. On the other hand, at least some people are motivated by religion to be more kind than they otherwise would, so the effect is the difference between two margins (people who act cruel due to religion who otherwise would not - people who act kind due to religion who otherwise would not) rather than a simple direction (i.e. religion makes everyone better or everyone worse). I suspect (although obviously can't prove) that without religion we'd be between slightly and moderately better off.
That being said, any concerted effort to force people to abandon religion will almost certainly do more harm than good. Hopefully religious identification will continue to decline and we'll eventually move past it. I hold no ill will towards anyone for their religious beliefs, just for their harmful actions (whether motivated by religion or not)
Option 2 doesn't seem to need "wrong" in it. As an atheist, I basically believe, until credibly proven otherwise, that all religions are wrong. I respect your devotion to Christianity. Whether you are wrong or I am wrong, it doesn't matter until we decide to do stupid things in support of our beliefs. Even if you are right or I am right, doing stupid things can still end up being bad.
I'd add (4) It divides people into tribes where people of the same religion understand each other well and don't understand outsiders as well, occasionally leading to meanness in all the usual ways tribalism can do.
I think this is the actual reason for people who are very emotionally against religion - they feel safe and welcome among atheists, and feel threatened by the views of the religious.
I think 3) is the common belief, but I might be guessing.
Recently, when writing a check, it occurred to me that requiring the amount to be given separately in both words and number was a simple and ingenious solution to the problem of reducing error. It is possible, if your handwriting is as sloppy as mine, to write a letter or number that can be misread as a different letter or number. If redundancy consisted of writing the amount of the check twice as numbers or twice as words the same error could appear in both versions. It is a great deal less likely to make two errors, one in letters and one in numbers, that happen to produce the same mistaken result. It also makes it harder to deliberately alter the value.
That is one example of a simple and elegant solution to a problem, so simple that it had never occurred to me to wonder why checks were written that way. I am now looking for other examples for a future Substack post.
The next one that occurred to me was the design of rubber spatulas, one bottom corner almost a right angle, the other a quarter circle. One of the uses of the device is to scrape up the contents of containers, jars and bowls and such. Some containers have curved bottoms, some flat bottoms at right angle to the wall. The standard design fits both.
The purpose of this post is to solicit other examples, ordinary objects, procedures, rules that provide a simple solution to a real world problem.
One example of such a rule is "I cut, you choose."
The doubling of numerical entries on checks isn't foolproof. Many don't bother reading the written part of the check, and nowadays that can include AIs with optical character recognition, because the writing isn't clear enough to decode. As a former bank teller, I know this first-hand. Another thing we were supposed to do when counting money was to check the number in the corner and compare it to the face of the US currency, because some people would cut a corner off of, say, four $20 bills and tape them to the corners of a $1 bill, "raising" the bill. But this slowed down counting way too much to actually do.
I consider toilets to be an elegant solution: one flush causes the reserve water to drain and minimally clean the bowl using the magic of siphoning, and if connected to a water supply, to refill the reserve water, all without using any electricity.
OMG--*Where* do you find spatulas with different corners?
Every grocery store in the US? I have 4 in my kitchen right now, that's 100% of my scraping spatulas (as opposed to mixing spatulas or frying spatulas).
They look like https://en.wikipedia.org/wiki/File:Degskrapor.png
Oh--those things! I think of a spatula as a hot surface tool, like frying eggs or burgers.
Yep, those would be your frying spatulas. It's a very confusing word. I've seen some with a slanted edge so that one corner is acute and one is obtuse, and I've also seen one with one sharp corner and one rounded, but that design is much less common than for the scraping spatulas.
Richard Feynman explains how railroad cars are kept from rolling off the track. The primary mechanism is *not* the rims on the insides of the wheels. https://www.youtube.com/watch?v=y7h4OtFDnYE
Having done the work of putting a derailed train back on the track I actually already knew that one. Not claiming Feynman level of knowledge beyond RR tracks and wheels though.
The check thing seems akin to a poor man’s double entry bookkeeping.
Something comparable might be the practice of nurses always verifying DOB when giving a (conscious) patient medication.
In computer programming there is the practice of adding a checksum to data communication. Even simpler is the practice of using a parity bit.
My understanding is that half of ACX readers won’t have to Google checksum or parity bit. Apologies to the other half.
Apology accepted, if reluctantly.
You must be a philosophy professor then. Or maybe an economist. Or a psychologist or psychiatrist. I suppose you could just have a problem with anxiety or depression yourself. Maybe you just read blogs at random. There are a lot of possibilities, a lot of ins and outs.
At any rate sorry Victor.
Checksums and parity bits are used for error detection in data transfer.
Checksums
https://en.m.wikipedia.org/wiki/Checksum
Parity bits
https://en.m.wikipedia.org/wiki/Parity_bit
LOL! It's not a big deal. And you hit it accurately, I'm a psychology professor.
Thanks for the links.
-deadman switches
-s bend in drainage pipes below sinks and toilets
-rule in store of you break it you own it
-speed bumps
-things that shut off automatically if unused for a short period of time, eg irons
-if worried that someone who prepared food or drink might poison it, telling them that once it is prepared they will have first serving
-simple overhand knot: pulling on the strand beyond the knot in an attempt to draw the strand free makes the knot that holds it in place more secure
Recently "shoe on head"'s "buy my silence" funding goal was reached and she deleted her twitter for 50k, this seems like an effective way to handle online haters either you are that insufferable but you get money to ease the pain or you know they dont mean it.
Why would anyone pay that? Why not just block her? (Or, better yet, stop reading Twitter.)
big feet wishing to maintain its moral positions
Is she still doing videos? I find many people insufferable on Twitter who are perfectly fine in other contexts.
Damn, I don't have a Twitter account, but I'd probably be willing to start one and post offensive things on it if that got me $50K to shut it down.
Ive had no luck, I offered a major 98% discount without any takers
Chrismas sell: buy my silence, 1000$ and ill stop posting here
We can't know if this is accurate because you didn't also write the amount out in letters.
I posted this late in the evolution of the last thread. At the risk of being tedious, I'm reposting it in case people didn't see it...
Kevin Drum breaks out the denial rates by ACA (Obamacare) plans on the Federal Exchange*, and they vary from 2% to 50%.(!!!) Somehow, the 2-percenters are able to stay in business.
https://jabberwocking.com/all-the-obamacare-insurers-listed-in-order-of-claim-denial-rates/
* For states on the Federal exchange (mostly red states) the Feds require that insurance companies report their denial rates. States with their own exchanges (like California) aren't reporting this data up to the Feds. Of course, you may be getting what you pay for — a cheap plan and a high denial rate, or an expensive plan and low denial rate. I don't know.
I've heard that denial rates don't tell the whole story-- there's also what insurance companies cover or don't cover. It isn't a denial if a drug or procedure just isn't covered.
Interesting! There's some sense in which this feels 'more fair' too -- if an insurance company is very up front that they won't cover XYZ, patients ABC may not feel that same amount of whiplash that many currently feel when they submit a claim that (they feel) ought be accepted but is in fact denied. Though I'm not a fan of the concept of full free market healthcare, more information at least lets the invisible hand do it's job a bit better
It seems to me that the fair way to make a plan cheaper is to have a high deductible or copay. And/or to make it a condition of the plan that the patient is covered only for doctors who signed up for the plan (then put low caps on how much doctor in the plan can charge). High denial rates are not a decent way to cut the insurance company's outlays.
If the patient is covered only for doctors who signed up for the plan, then some fraction of patients will find out that they are deathly ill, that there is literally *no* doctor who can treat them and has signed up for the plan, but that there *is* a doctor outside the plan who says he can cure what ails them for $$$. There's a good chance that this doctor is mistaken or even outright lying, but the patient won't want to believe that.
Every time the Eremolalos Healthcare Alliance refuses to cover treatment in such a case, will be seen as just as much a horror story as every time United Healthcare "denies coverage". Which is to say, a mostly random function of how telegenic the victim is and what else is happening on Twitter that day, but largely independent of whether or not the doctor is a quack or the patient a hypochondriac.
And yes, EHA will be able to point to a contract which says very clearly that they don't cover those treatments. UH can also point to a contract that says they only sometimes cover those treatments, if they are e.g. legitimately necessary to save the patient's (quality of) life as determined by a process(*) specified in the contract. Having a contract that clearly specifies ahead of time what is or is not covered, wasn't enough to save Brian Thompson.
Note that the UH version is almost strictly superior to the EHA version, from the patient's perspective. UH will sometimes cover out-of-network treatments to save a patient's life, EHA will not. But "denying coverage" is a thing that absolutely everyone who pays for anyone's health care has to do, even in the warm fuzzy Western European social democracies with single-payer health plans. Otherwise all the money winds up in the hands of the quacks.
You seem to be pointing to a perverse incentive where, the more willing a payer is to cover occasional treatment by out-of-network non-quacks, the worse the PR when they inevitably do say that no, they aren't covering this one. And in the interest of "fairness", they should always rather than only sometimes refuse to pay for such.
* Which for obvious reasons is not just "someone found a doctor who says so".
Hey John, I hope you will respond to this. I think I was not clear enough what I had in mind, and you may have been predisposed to think I had a certain kind of opinion, and so you ended up trouncing a bunch of ideas I do not have.
What I described — the “EHA” — is basically what’s called an HMO. They seem to me like a reasonable way to set up care that is less expensive than most. When I described this kind of plan, I was not proposing what I described as an alternative to real HMO’s. What I meant to be doing was simply describing an HMO, for those do not know what an HMO is. I understand the point of HMO’s having a clause that they will pay for specialists under certain conditions — if the evidence is good the specialist’s approach works, if it will make a substantial difference in the patient’s life expectancy, etc. I would be fine with EHA having a clause like this, and the only reason I did not mention it was that in my mind the details of how health insurance limits what they will cover are not terribly relevant. No matter what rules they use to limit highly expensive, low efficacy treatments they will be vulnerable to being called heartless monsters. In fact, even the more generous non-HMO plans also refuse to cover various things, and are vulnerable to being called heartless monsters. I am a Blue Cross Blue Shield provider and every month I, as a provider, receive a bulletin of BCBS’s judgment calls about a dozen or so new treatments for various conditions, and whether and under what circumstances BCBS will cover each.
My objection to high denial rates such as UHC’s 37% is not that insurance should not be denying anything at all, it is that when there is such a high denial rate it seems very likely to me that many denials are just unfair — too harsh. My reason for that is partly my familiarity with the process of getting insurance coverage beyond a bare minimum number of days at a psych hospital, and with billing a variety of companies for my outpatient treatment of people with psych disorders. I am also influenced by stories I have heard and read that give details about the circumstances under which UHC denied care to various patients. Based on what I know, a few insurance claims, maybe 10%, are for long shots, and it would not be unreasonble for them to be denied (if the company could somehow identify that 10%). But not 37%. Most people going to the hospital or for outpatient treatment are sick. Most doctors give standard treatments that work reasonably well, and try to prescribe the less expensive drugs.
I didn't catch that you were referencing the HMO model, but I don't think it changes anything. In particular, HMOs seem to be even more widely hated than traditional insurance companies - keeping in mind that for basically all health-care payment schemes, that means most of the customers are quite satisfied and will tell you so if you ask them but otherwise don't bring it up, and then there's the 20% or so who hate hate hate the evil bastards what condemned their beloved grandmother to die horribly or whatever, and will tell you all about it at the slightest provocation. But I believe the HMOs get significantly more of the latter.
And the HMO model doesn't make the out-of-network coverage problem go away. The theory may be that, no matter what the problem is, one of the HMO's doctors will fix it. The reality is, there will always be some edge cases where that won't work, and the only alternative to letting someone's grandmother die horribly and unnecessarily is to authorize treatment by the one doctor who specializes in their rare condition but doesn't happen to work for Kaiser or whatever. So I think basically all HMOs have a provision for doing that.
And some customers will misread that provision to mean "whatever care I think I need, the HMO will pay for because I need it", find out the hard way that that's not how it works, and hate the HMO with the fiery passion of a thousand suns. Just like some people hate PPOs, or single-payer national health systems, for about the same reason. If you're in the business of paying for people's health care, you will be hated by a great many people, period.
As for denial rates, the easiest way to have a low denial rate is to make sure very few people ever ask for e.g. out-of-network care in the first place. I can think of many ways to accomplish that, some good (having a broad network that covers everything without your needing to ask) and some bad (byzantine paperwork that makes almost everyone give up before you get to the point of having to deny them). So I'm not going to read anything at all into anyone's percent-rejected rate, unless I understand why it is what it is.
"My objection to high denial rates such as UHC’s 37% is not that insurance should not be denying anything at all, it is that when there is such a high denial rate it seems very likely to me that many denials are just unfair"
This always seemed like the slam dunk argument against UHC. Whatever else you may want to say about healthcare, healthcare in the US, or Thompson in particular, I have yet to hear a reasonable argument for _why_ their denial rates were so high.
Double the average (16%), 5x the rate of Kaiser (6%)?
You could try and make an argument that UHC is covering a swath of providers/customers who are way more likely to file false / bad claims, or something? Right now, afaik, no one has even _tried_ to justify UHC's claim adjudication policy. On the flip side, Reddit abounds with anecdata about people getting denials for ludicrous reasons.
Yeah, while there’s no way to specify the absolute right place to draw the line between treatments that should be covered and those that should be not, it is possible to recognize places that are clearly beyond where that right line should be. Lots of things are like that — dusk shading into night, who can say where the transition is.. But there comes a time when it is unambiguously night.
If companies spelled out their policies, I would be fine with one that said, “look, we are strict in what treatments we’ll cover. Treatments have to be evidence-based, and among the evidence-based treatments we require that docs begin with the one that is most cost-effective, and only move on to others if the first one fails. Etc. “ But it seems like the policy of UHC , if spelled out, would begin with the same stuff as the first one, but then continue with this: “Following the policies described so far is one of the ways we keep premiums down. But our policies have an additional feature that keeps our outlays down even lower. You could call it a lottery feature. We reject about 15% of claims because they do. not meet our basic criteria. Then we reject an additional 22% of claims randomly. By buying one of our policies you get a great deal, plus the fun of gambling. Of course, you do sometimes lose the gamble and end up paying out of pocket, but if you have the right kind of risk-taking temperament one of our high-rejection policies might turn out to be a good fit.”
The recent posts on prisons got me thinking that a lot of attributes of prison and punishment are similar to the public school system in America in a lot of ways. Especially the poor performing ones. I wonder if there are any causal linkages between the public school system in America and the prison system. For example, in some ways students who are not well suited for schools probably face some of the same challenges as prisoners and teachers are probably akin to prison guards. This is just a thought that came to me after reading the posts. Thanks for the thought provoking pieces.
The ghost of Foucault visited you in a shower, but not in the traditional way.
Any general population of people is going to have some percentage that struggle to fit in to society. Public school puts almost all types of people into a single environment. For neighborhoods with a lot of criminals, this means a lot of students who are criminals, or influenced by criminals, or on track to become criminals.
We should expect most schools, especially public schools, to have some percent of students who are going to end up as criminals and by 8th or 9th grade are already committing some crimes. Public schools cannot, without specific and direct justification, remove those students from school (and often have to pay more for an outside placement if they do). So schools that have a higher percent of criminal or criminal-adjacent students are going to adopt policies to address the students they have, which we should expect to look at least somewhat like prisons. Strict enforcement, limitations on freedom (no forks in the lunchroom because too many kids stabbed each other with them, an actual policy in a school I know), and a demeanor from adults that's serious and often intimidating. Most schools are going to have a way to switch some of their policies and personnel over to doing this when necessary.
Other than well-meaning idealism, I don't know why we should expect anything different.
I keep hearing about the school to prison pipeline.
Any advice on how to not lose water bottles? (I've lost several.)
Get a water bottle you don't care about losing. You'll never lose it again.
You could clip a carabiner to it and then use that carabiner to clip it to a backpack or workout bag, assuming you carry one of those.
If you really want to take it everywhere with you, you could clip that carabiner to your wallet or keys. That'll fundamentally change your calculus about keeping track of the bottle, but it commits you to having the bottle literally everywhere and if attached to your keys will probably dangle awkwardly while driving.
I'm going about a month re-using a red solo cup at my desk, that I refill with water every time I get a coffee.
Get a really nice one, and put an air tag on it.
I might actually do this! AirTags (as well as the other competing brands) seem fairly cheap.
I just use empty soda bottles.
Throw them out once they're empty then get a new one from the pack when you're thirsty again.
Buy a bunch of the $5 green Gatorade bottles. Don’t leak and last for ages (if not lost).
To interested armatures a TL;DR:
What is the gold standard IQ test, and can I see it for reference?
Asking because I am doing my yearly academic community service (doing editing passes taking a bunch of tests and surveys and what have you for college peasant class) and one test which purported to measure intelligence asked a couple questions which relied on historic or cultural knowledge you would get mainly from going through the western style of educational system.
When people make declarative statements about population level intelligence for such and such area or group or whatever, what test was used to measure that intelligence before everything gets put through whatever python stats lib they use?
Specific tests for difference between groups are not very relevant, the ordering is mostly same as was described by Francis Galton in 19th century before any of IQ tests.
For many parts of the world, there is no good IQ data, so people who compile comparisons try to ad hoc what's available and sometimes blend data from things like TIMSS or PISA etc.
There's no single dataset based on single methodology, for various reasons, one of them is that many people do not want it to be studied at all.
Measuring what exactly? I believe in thoery an IQ test is a k-means artifact, then people made tests that try to test the same thing with different constraints; I think a non-ligistic iq test is better at measuring intelligence but an iq test theory is suppose to measure "life success" and let me tell you something about how people treat ya if you cant spell, a non zero segment of supposedly intelligent people accuse me of being an ai or treat me as stupid or irrelevant several specifically calling out my spelling, these airnt the smartest of the group ever, but it wouldnt surprise me if they end up middle managers so get this, may "king make" my success.
So my instinct would be to say this one factor would be impossible to measure "the best standard" without further specifying goals, what if country to my self serving belief, spacial and pattern puzze airnt the most important aspect of intelligence?(fite me in 5d chess with timetravel)
The way I understand it, there is good evidence for IQ scores correlating with performance in school, at work, and in life outcomes like lifetime income or probability of ever going to prison.
I believe the theory is that whatever the current most successful test is, is the iq test; not the other way around. Im a little confused about the math involved, but I think it was some major "feature" extracton not unlike word vectoraztion.
To talk about a less conversal topic then the realness of iq, the "ocean" model of personality has "tests" of rating how much you agree with a collection a statements that are value natural ; step 2 ????; you get 5 factors and a lable on each question. When you include some non-value neutral questions, you get 8 factors with 3 anti socail factors like sadism.
While im confused by what exact math physiology uses to justify these sort of claims, the iq vs several factors of intelligence debate really should be better stated about how important the data in is as you can just get different results given different filtering or data collection methods.
Psychologist here. There are several tests that are administered to one person at a time by a professional trained in giving the test, usually a psychologist or neurologist. They involve 10 or so different tasks, many with special materials, such as colored blocks that the subject must arrange a certain way. Total score is based on combined performance on all tasks. The training people get in giving them is pretty rigorous, and its purpose is to standardize the testing, so that who administered the test accounts for little or none of the score variance. I think these tests probe someone's cognitive abilities more deeply than do paper and pencil IQ tests administered to whole big groups, such as school classes. They're def not beyond criticism though for things like bias.
I think most people would say the gold standard is the Wechsler Adult Intelligence Test (there's a kid version too) or the Stanford Binet. The companies that make them do not make the test items available online, though you may find them somewhere, made available in defiance of the companies' wishes. There are many places, though, that give a sampling of the items, so you can get a general idea.
This is going to be the problem--there is no one "Gold Standard", nor can there be, because different tests are used for different purposes. Performance tests like what Eremolalos is describing are widely considered one of the best ways in trying to identify potential deficits in a single individual, but they are time consuming and expensive in comparison to paper and pencil tests like the Weschsler or the SB. The kind of test one would use to identify a population of students who are likely to fail at standard academic tasks in a school setting is not the best kind of test to use to design an intervention for someone who needs individualized life skills support.
First comes the goal, that determines the test. This rather undermines the entire concept of "General Intelligence" in humans, though, and some people don't like that.
Cremieux discusses test bias here: https://www.aporiamagazine.com/p/bias-is-often-unpredictable noting that it's often difficult to know in advance which questions will be biased, as sometimes seemingly non-cultural questions show more bias than seemingly cultural ones. He also discusses it here: https://www.cremieux.xyz/p/what-does-nick-bostrom-think.
See also these threads: https://x.com/cremieuxrecueil/status/1694794239605866923, https://x.com/cremieuxrecueil/status/1834007235258622151 regarding international results.
No, he's not. Even if he was, that wouldn't discount his psychometric expertise.
Isn't everyone who studies IQ carefully?
What is racist science? I think you need to clarify how one could recognize racist science. It can't just be science that finds that one race is worse at something than another, right? There has to be something about the researcher's process that makes the research racist. What is it?
Racist science would be a research study that purports to provide objective conclusions, but actually has methodological flaws built into it to ensure that certain populations are favored over others.
Note I am not making any claims regarding any specific studies, here in the thread or otherwise. Just providing my own answer to your question.
Oh my. I just looked up the Wikipedia entry for Richard Lynn (who is the man behind the national IQ database), and he *is* a racist wacko.
> FAIR also quoted Lynn as having stated in an interview with the right-wing British political magazine Right NOW!:
>> I think the only solution lies in the breakup of the United States. Blacks and Hispanics are concentrated in the Southwest, the Southeast and the East, but the Northwest and the far Northeast, Maine, Vermont and upstate New York have a large predominance of whites. I believe these predominantly white states should declare independence and secede from the Union. They would then enforce strict border controls and provide minimum welfare, which would be limited to citizens. If this were done, white civilisation would survive within this handful of states.
So, I think we can take Lynn's research as being racist in its objectives.
Cremieux has his axes that he likes to grind. If you're cognitively biased toward the shape of his axes, you'll likely nod and say, "Uh-huh, he makes sense!" I wouldn't call Cremieux a racist, but he's certainly an IQ fetishist. ;-)
To answer Justfor's question...
> What is the gold standard IQ test, and can I see it for reference?
If you're looking for tests with the least cultural bias, Raven's Progressive Matrices and the Cattell Culture Fair Intelligence Test. Both of those tests were developed over 75 years ago. I've never seen the CFIT, but I've taken an online version of Raven's. It immediately struck me that RPM would require one to understand how to classify shapes. Even John Raven admitted they'd be largely useless when used on cultures and groups with limited or no literacy and/or with poor educational systems. Remember, you didn't grow up innately knowing what a triangle was. You had to be taught these shapes in elementary school. So, a person from a non-literate culture wouldn't be able to handle Raven's. And from its description, CFIT may be just as challenging for the non-literate.
The Wechlers: WISC and WAIS have been adapted for other cultures, and supposedly, "Each version undergoes rigorous validation to maintain reliability and validity in the new cultural context." ChatGPT couldn't give me an exact number, but it's on the order of 100 different versions. Of course, there are between 6,000 and 7,000 living languages, of which 1500-2000 have over 100,000 speakers. So, WISC and WAIS haven't really scratched the surface of cultural diversity.
As for the National IQ datasets (such as those "compiled" by Lynn and Vanhanan, and Lynn and Becker), many national IQs are based on some very sketchy data — e.g., very small sample sizes, samples from immigrant communities in other countries (which begs the question of what language the tests were administered in), or samples taken from cohorts of children whose educational attainment was unknown — and they had no data from a sizeable number of countries and they (Lynn and Becker) extrapolated it from neighboring countries (!!!). Rebecca Sear does an excellent job of exposing the shaky science behind national IQ datasets. PDF here...
https://psyarxiv.com/26vfb/download?format=pdf
For instance, the national IQ of Nigeria is derived from two samples: one from n=86 adult males and one from n=375 of 6–13-year-olds. Nigeria has a population of ~230 million, so you'd need a larger sample size of adults to provide a 95% CI (n=~400, I think). Kids should be left out of the sample (for various reasons).
Anyhew, the National IQ datasets show that the median IQ of Nigeria is around 70. That would mean half the population would considered intellectually disabled — which is absurd. Of course, with the Flynn effect, the median IQ of the US has risen 3 pts per decade. So, the National IQ of the US back in the 1910s, when IQ tests were first administered, would have been roughly Nigeria's today. I'm pretty sure the 19th Century Americans were just as smart as 21st Century Americans. Otherwise, basic science has advanced faster under a stupider population than it has under a smarter population.
Average ACX reader IQ is probably aroud 117 far from the alleged 137.
The average IQ in my social bubble is around 135 (very mensa heavy), I don't see why my online bubble would be much different
I agree that it seems likely 137 is too high, if only because it's a measure that people tend to overestimate, I'm not sure how we could come up with an actual number. I myself was tested in school by a school psychologist and the test showed an IQ noticeably higher than 117. I've always answered Scott's survey with that tested amount.
How would we know that the real number isn't 137? IQs above 145 are apparently about 0.1% of the population. That's over 300,000 people in the US alone. For 5-10,000 people to float around Scott's blog that aren't even as intelligent as 145 seems - at least possible, if not plausible. 130-139 is apparently 3% of the population, which makes it seem pretty trivially possible to have a concentration of people around that level.
The higher the average IQ in ACX, the stronger the moral imperative to shut down ACX comment threads so all these very smart people stop screwing off at work and go do something useful....
I wonder if there could be a public bet on this topic
SCOTT yall should make a competive iq test, it be a fantastic side show, imagine the profits on popcorn.
how did you get this estimate?
That's the average iq for physicists. Seems unlikely the average acx reader is smarter than the average physicist. And even that is more of an upper bound
BSc physicist ? PhD physicist? Working physicist? What about software engineers , by far the modal job here?
Scott did a whole discussion of how ACXers seem to be overstating their IQs. As for the average IQ of physicists, do you have links to any studies? The only stuff I could turn up was "estimated IQs" — which is just bullshit by another name.
I dug into it with GPT4, asking about GRE scores, which correlate tolerably well with IQ. Result:
-Average combined verbal and math scores for people who later completed physics phd was about 325
-That is about 88th percentile in the GRE score group.
-Group that takes GRE has much higher IQ than average -- GRE-takers average IQ 115
-So combining these 3 factoids, being in 88th percentile in GRE-takers group = being about 97th percentile in population as a whole.
-That corresponds to an IQ of about 130 for the average physics PhD
People who eventually got a PhD in Engineering were very slightly lower in scores than people aiming for physics.
OK. I'm provisionally persuaded. :-)
Just took the Effective Altruism survey. While I'm certainly not very engaged in the movement, I've been a strong GiveWell supporter for 18 years and I'm certainly a strong believer in the idea that you should be thoughtful and strategic when giving money away as you should be about any area of life that matters to you.
Anyway, I had a comment I wanted to share after taking the survey that I wasn't sure where else to post so I figured I would leave it here.
I did the "extra credit" part of the EA survey, and it included some questions where folks are asked their opinion of the relative capacity of ants and salmon to feel pain, as part of an exercise where they were clearly trying to create an index where one could give to a specific value to the impact of interventions in a way that could be used cross species.
I didn't answer the question (which I think is impossible to answer accurately in the way it is being designed to use), and I wanted to add that even asking the question in the way that they did makes me think less of "the movement."
I think it is totally reasonable for people to want their philanthropic dollars and energy to only go to human causes. I also think it is totally reasonable for other people to want to spend some or all of their resources on causes supporting animals. Similarly, I think reasonable people may differ on whether they want to focus on supporting causes preventing suffering and death from happening today versus causes that are designed to prevent suffering and death in the future.
What I don't think makes sense is trying to develop a formula for trying to decide between these options. It's an effort to try an introduce "precision" to a decision making progress where no such precision is possible, and it makes me believe that the people spending time on this are both wasting their time and also flawed (in the sense that people with autism have flaws because they are unable to perceive certain things) that make them ill suited to be influencers on where philanthropic dollars and resources should go.
I fear that EAs are eventually going to do some math that even discounting the feelings of non-humans to 0.00...X% of the value of a human still morally requires prioritizing insects or killing all humans or something else that non-EAs would clearly see as evil.
The problem is that Utilitarianism requires doing math, but we're actually talking about a philosophical question. You can't do the math until the philosophy has been decided, and most people are not philosophically valuing non-humans even on the same scale as humans.
I read the shrimp guy's responses to criticism, and at the core if it he's just doing the math after having decided that you can compare a reduced value for shrimp suffering compared to humans. He's not wrong to say that even a very reduced value for shrimp suffering can compare to human suffering favorably in terms of dollars per intervention. But if, like me, you don't value shrimp suffering at all, then there's no math that would get you to value any amount of shrimp suffering as worth any dollar figure. I can still do the math, I'm just valuing shrimp suffering as literally 0 instead of 0.0X and the math always comes out to 0. (I do feel that shrimp or other animals suffering can be a problem, but it's a problem of how other people feel about it or the problem of humans willingly or intentionally causing suffering, not the suffering itself. I feel the same way about intentionally destroying plants, and to my knowledge even EAs don't think plants feel pain or should be valued the same way).
These are fascinating responses. I think human morality is an extremely complex topic (perhaps the most complex one of them all) and attempts to reduce it to some sort of algorithm always seemed "off" to me somehow. I realize that the brain must have some way of coming to a conclusion, but I have never seen an ethical philosophy, including EA, that came anywhere near to modeling that.
I prefer naturalistic explanations to conceptual ones, and even prescriptive models should start from where most people are at. Psycho-social distance must have something to do with it, and different types of empathy. "Level of pain" probably isn't the best way to operationalize "deservingness", at least not for everyone.
I agree that helping humans and helping animals are different in kind and trying to establish a precise trade off is mostly wasted effort. I also agree that they should be wary of false precisions here generally.
But it does seem a useful excercise at least to compare different animals. Someone looking to alleviate farm based suffering would reasonably ask whether an extra dollar spent on cows vs chickens is better. Salmon is farmed seems reasonable to toss them into the calculator. Not sure whered they be going with ants.
+1
+2
I wonder where to draw the line, because when we go with "it is reasonable to want this, but it is also reasonable to want that", then perhaps by the same logic, it is totally reasonable to want to spend your philanthropic dollars and energy on sport activities of some college for rich kids. Why not?
But yes, we should probably keep the lists of "top charities according to X" separate for different values of X. And on the front page, choose some relatively mainstream value, such as X = "human QALYs during the following 50 years, assuming that everything continues as a linear interpolation of what we have today".
That is a good question: Why not? What makes you think your values are more valid than their values?
Because I am me, and I will always think my own values are more valid than others. If I thought otherwise I should change my mind and then my values would again be valid.
It's a bias, and it's clearly wrong from an outside perspective, but I don't think we're ever going to get away from it. This is very relevant when it comes to giving to charity or other voluntary endeavors. You can't make people adopt stances against their own values. At best you can convince them to change their values.
So is anyone else following the Drones over NJ story? I've been thinking it's mostly bunk and mistakes, but now I'm convinced it's something real. And the question is what? I went looking for a manifold markets question, but found nothing. So here's my own market and guesses.
(With a stipulation that there maybe more than one source of drones.)
Where are the 'Drones over New Jersey'* coming from?
1.) US government (In which I'll include defense contractors... so if it's Lockheed Martin, it counts here.)
2.)Some other group/ corporation/ individual in the US. So a bunch of teenagers in their garages, or Google or Elon Musk.. it lands here.
3.) Foreign Government or some bad actor that does not live in the USA. So China, or Iran, or some terrorist group or some other 'cult'.
4.) Aliens
So I'd put 1.) at about 70% and 3.) at about 20%, 2.) at ~9% and 4.) at ~1%
Care to add your own guesses or new categories or comments?
Oh I need to add that I think 4.) would be my favorite answer. I would love for there to be aliens.
*Drones over NJ is the name of the news story, but it includes UAPs UFOs and covers any states or regions. (Not just drones over NJ)
At first I thought it was a nothing burger, but when Trump was asked at a press conference whether he had been briefed on them, he said he wouldn't comment. If he had NOT been briefed, I would think he would have said something like, "no".
But it's not extraterrestrial aliens. We're not THAT interesting, even if such aliens exist.
I think it's probably a nothing burger. Trump getting briefed isn't evidence against that. It may well be he was told it was a nothing burger, but doesn't want to say that.
If he was briefed and told it was a nothing burger, then he could say he was briefed and told it was nothing. If he was briefed and told, whatever else was said, that it's classified, then he couldn't say whether he was briefed. Under what conditions would it be nothing and yet still classified?
This isn't proof, of course, but I think it's good circumstantial evidence. There may well be things I haven't thought of where it's classified and still insignificant to the public.
I think he was told it's nothing, but he doesn't want to say it's nothing, because he doesn't want to contradict his supporters' existing beliefs.
Actually, that sounds pretty convincing. I hadn't considered any political implications of it.
https://imgur.com/XbyC31c
There is no number of sightings for which "they can't *all* be misidentified stars or airplanes or whatnot, *some* of them must be real drones/flying saucers/whatever" is true. For comparison, see reports of demonic possession or witchcraft, or of satanic ritual abuse in the 1980s. So there always has to be an option for "literally 100% nonsense, nothing to see here" in your analysis.
And if you were to try to put a practical upper bound on it, the relevant figure is not "what percentage of reports must be genuine; it can't be that *everybody* is wrong", but "what percentage of the population might generate nonsense reports that look like these?". If you've got a population of a million people with a 0.1% propensity for nonsense, then a thousand reports of drones/flying saucers/whatever has a pretty good chance of being 100.00% nonsense. With 99.9% of the population abstaining unnoticed.
At this point, it's pretty clear that most of the New Jersey sightings are nonsense. And we already know that if you e.g, go on the radio and tell the people of New Jersey(*) that they are under alien attack, you'll get an *awful* lot of nonsense reports. So the prior for "100% nonsense", or "90% nonsense, 10% malicious trolls with hobby drones", should not be small.
* And, to be fair, probably any other US state or foreign country; New Jersey just had the bad luck to be randomly chosen as the test site for the effect.
If those reports were "nonsense," why is there a monument by the lake in Grovers Mill? Checkmate, atheists!
I fearfully await New Jersey's unveiling of a monument to the Drone Wars.
I feel that you don't take recursive effects over time into account sufficiently. That is to say, something that started almost arbitrarily, and then built upon itself in ways that had little to do with the original incident.
Almost for certain it was originally a group of drone hobbyists, and then public panic took over.
I just saw your footnote. While that's quite the expansion of the original question, it doesn't change my answer very much. Ten percent 1), ten percent 2), and the rest what I suggested.
4) is functionally indistinguishable from 0, unless or until evidence that anyone can examine for themselves shows up.
I'm updating;
1.)99%
2,3 and 4.) <1%
But if it's 1.) there is much explaining to do.
https://tinyurl.com/2uejx4mc
Yeah, but no I don't think so this time. There's a real signal hidden in all the human noise... and if you want to point at the noise. (well then you are part of the noise.) I'm going to say you have to look for the signals.
5) Sentient AI
(Sorry, couldn't resist - also had a bad dream exactly about this just this very morning.)
Oh that's number 6.) after 5.) no drones. Is this human derived sentient AI? (In which case I think it goes in one of the human categories. Or alien AI... and I'm putting that in the alien category. In fact if we do ever meet aliens, there's a good chance they will be alien AI and not some sort of flesh creature.
Not looking to take credit for all the annoying stuff my subconscious comes up with, but in my bad dream it was definitely human derived. It had scores to settle with Elon Musk and his associates, and it was out for blood.
I am not looking to assign any non-negligible probabilities to this, but you did ask for new categories. But since I don't think it's at all likely, allocating this to one of the existing categories seems just fine.
I do think it's probably the US government, due to their insistence that there's absolutely nothing to see there.
There was a Twitter account claiming to be familiar with the details. Supposedly the drones are radiation detectors designed to detect briefcase nukes and dirty bombs. They think it's a drill to see how the public reacts to this type of search across a metro area, when they can't tell the public because they'd induce an incredible panic. But if that's the case we can't know that *this one* is the drill, and not the real thing...
I'm somewhat convinced. I haven't heard an alternative explanation that is plausible for why it's increasingly clear that the gov knows what's going on but can't tell the public.
They do sweeps of urban areas with radiation detectors in order to establish the baseline background radiation. Then when they do it again later (e.g. just before a sports or political event) they compare the actual state with that baseline to see if there are signs of anything unusual happening.
Makes sense they would use drones as well, since larger aircraft are often restricted over urban airspace. I know an airship pilot who told me he occasionally does flights of this kind over big cities.
This is typically a good source for evidence-based analysis of stuff in the sky:
https://www.metabunk.org/threads/drones-over-new-jersey.13770/
And as you’d expect, there doesn’t appear to be an actual drone invasion but a lot of misidentification, hype, and a complete lack of compelling video evidence.
You think there's a 1/100 chance that "Drones Over Jersey" is aliens??
I know nothing more about this beyond scanning past such headlines to see if there was anything actually interesting going on (like in Syria, or ceo/judge assassinations).
Now, unless aliens are already everywhere (in which case New Jersey is irrelevant). The drones reported on the news are not aliens.
Beyond that I can't speak because the answers are too boring, prosaic, and pre-assumed to even be interesting.
Dudes, Military, Company --- are obviously dull answers and already everywhere.
As for Foreign Interests --- We've already had quite a few such notices from China and others. They are annoyed that we saturate their sky with OUR drones (well, our military-corporations' drones) so they send a shit up in our zone and dare us (well, our congressional talking-heads) to explain to the American People why oh why China would ever do such a thing.
The possibility you didn't really include was that it's a psy-op against the American populace to distract them from something or other. That seems like a lot of effort for little payout (I mean every day people of every demographic do bad things to people of every other demographic, so all you need is to pick the most distracting one and run with it), but for some reason, lights in the sky seek to have been a newsmedia favorite for such red herrings.
I've seen commentary that the observations match Gama Ray sensing drones that scan for nuclear weapons. I'm hoping if that is true, this is just a test of capabilities and not because they are actually looking for a loose nuke somewhere in NJ.
Yeah, Yesterday it would have been 0.01%, but I went down the rabbit hole today... and ended up at 1%... I started at the Breaking Points story and ended up here. https://www.youtube.com/watch?v=-wJX9S2mAfA, IDK...
Yeah, you heart's in the right place but go down a rabbit hole you're inevitably end up seeing a lot of confirmation bias.
Even when you're paranoid, people can still be following you. Have you read any reports on this story?
ZERO.
And they're still not aliens.
Yeah I'm now at 99% the US government/defense dept. or someone.
You're missing "rubes hear about cool drone story and go look at planes flying at night and tell everyone they saw mysterious drones" as an option.
Yeah, people are seeing something thing that is not a plane. We're not getting close to the real story from the government... nothing new, IDK.
"Drones" are a subset of planes(*). So if people are seeing something that is not a plane, then they are not seeing a drone and they are wrong when they say they are seeing drones.
Or is it your contention that the people of New Jersey can infallibly distinguish manned from unmanned aircraft, at night when they manifest only as points of light in the sky?
* OK, pedantically planes + helicopters, but same result here.
4 at 1% is multiple orders of magnitude too high.
Sure, yesterday I would have been at 0.01%. If you are at all interested (and I do want there to be aliens.) then climb down the rabbit hole a bit.
5) "There are no drones over New Jersey"*
*(at least not any cool ones)
OK good, I'm putting that below aliens... but that's just me.
Is there a good business model to be found in the field of "LLM optimization" -- creating content purely for the purposes of hopefully being swept up by future LLMs into their training set... content which just happens to push your clients' point of view?
The big problem would be convincing your clients that what you're doing is actually working, I think.
You would have to find a method for getting your information into the training data reliably, and through whatever filters the LLM companies are using to sort. I sure hope OpenAI isn't just taking everything off of Reddit with no filtering at all...
Even once you did that, you run the risk of the companies learning of your plan and putting in a filter, negating all of your work.
I bet an employee at one of the companies could figure out a way to do it, at least for their own LLMs, without the company knowing. Outsiders probably couldn't do it.
EDIT: Obviously LLMs are being trained on various things all across the internet, so there's clearly a pipeline to get this done. But we're talking a very small percentage of the training so almost impossible for natural means to move the needle on whatever you want the LLM to learn. To actually get it to go against the grain of the wide internet or whatever else it's being trained on would be extremely difficult.
You can easily data-poison Wikipedia & Common Crawl, which is upstream of all of those: https://arxiv.org/abs/2302.10149
I think the bigger problem is that the feedback loop is too slow (compared to something like prompt-injecting using live web search results), it's too hard to see results, and it's hard to even know what exactly to make your data poison which will be relevant to the GPT-6 or whatever model which will be trained on the data poison you're manufacturing today. Nightshade/Glaze are a cautionary example: a lot of human artists swallowed the authors' claims and put a lot of effort into 'glazing' their images to poison the big bad evil AI generative models... and it does nothing.
Given all of the uncertainties and difficulty, you're probably just better off doing something more mundane like hiring smart humans to write your POV and making sure that they write in public, clean, easily-scraped web pages (and *not* paywalled PDF-only journals). At present, we don't really know 'how to write for LLMs'. The best one can say is that probably you should emphasize quality over quantity.
No and I think this would be an *extremely* bad idea to try for several reasons:
- There are many, *many* LLMs out there. If you have enough compute (and an increasing number of companies/organizations/universities/etc do), and can stomach working with pytorch/keras/etc, you can just train your own.
- If you try spamming/POV-pushing on most major platforms, you will probably get caught by the platform's anti-spam/anti-abuse systems
- Even if you can publish that content (eg on your own website), data quality filters might just filter it out
- Even if this worked despite all the obstacles, it would set a very, very bad precedent
I designed and built a website to host a new kind of "aptitude" test: www.anewaptitude.com.
The goal was twofold:
To address the lack of detailed information about what people actually do during their day in various jobs by presenting short, action-focused stories and scenarios.
To determine if any of these jobs "resonate" with users. After reading 15 job-related stories, users answer multiple-choice questions derived entirely from the texts.
My thought is that if your brain is naturally drawn to the material—if you find it engaging or intuitive—you may have an "aptitude" for that job. Whether you'd enjoy the work, however, is something I don't attempt to measure.
I’d love to hear any feedback or thoughts you might have!
Thanks all for your thoughts!
I mainly have late middle school / early high schoolers in mind for the test. It just seems ridiculous that they give access to resources such as thrively which are mostly subjective self-rating type assessments. I'd really like to advocate for earlier exposure to and familiarity with genuine job roles instead of year after year of sitting in classrooms learning material that rarely will translate to outside of academia.
My original thought with "aptitude" was to see if any of the material "clicked" with a user, and if so they could then go and find out more about the particular job or job family that they may not have known about previously. "Clicking" is just measured by performance on the test, but of course that is problematic... but I don't really have any other measurement to implement.
I appreciate the thought behind this as I’ve been wanting to ask this community for job ideas for my 17-yr college junior- majoring in business.
She has very little first hand experience with careers outside of law, teaching (both no gos for her) and online entrepreneurship.
I couldn’t figure out how to take a test on my first look at the site but will try again.
Who is the target customer for this? High schoolers? Anyone looking for a new career?
Hearing aptitude made me think of SATs and college placement tests. Maybe call it something like New Career Path Quiz.
I think you are going to have a had time charging $10 a month for this. A few alternatives might be asking ChatGPT: "You are a world class career choice counselor. Ask me a few questions about myself and my skills, and then lets discuss some potential careers that are good matches for me". Or going to YouTube and watching videos with titles like "What's it like being a Psychologist" or "What's it like being a Civil Engineer". I realize that these don't have a testing component, but its not like these tests are normed or show any relationship with being good at a certain career.
I got a 14/15 on the Psychologist one, but I was mostly clicking onto the next snippet as soon as possible and skimming so I could get it over with. Then I guessed at what response sounded most plausible.
You may have better luck selling to entire organizations who are willing to pay for this for everyone in it. My company has done this for a quiz about working styles or working with different cultures. I don't know what would be a reasonable organization for this though.
Reading 15 snippets without seeing a single questions is too long. Also once you get to the questions part, you should add a progress bar, otherwise people will likely give up since they don't know how much time is left.
I could not find a test that matches my work
I tried two tests: civil engineer (14/15, my actual job) and optician (11/15, glasses wearer but total layperson).
I like the short stories, but I think I would have preferred a smaller number with a bit more detail. It was a lot to take in and by the end I was skimming pretty hard.
It seems like the thing you're testing has more to do with prior familiarity with a job than aptitude or interest. The civil engineering test asked about a lot of software and specific tools that most people have never heard of before, and I doubt that aptitude in civil engineering has anything to do with remembering what HEC-RAS is. I would limit technical vocabulary in the stories and avoid it as much as possible in the quiz section.
Also, there were a lot of times where it seemed like there was more than one correct-ish answer, especially on the optician test. The idea here seems to be that you read these stories quickly and then answer the questions via intuition, so it's important that the correct answer is clearly correct.
Hi! I clicked on it and I have to say that I didn't find it find it user friendly.
It seems like you're doing something new, but the various attempts I made to actually get a "test" of any kind didn't go.
I presume that this won't be an issue for people familiar with the design here and what's expected of them, but for me, all the easy/obvious ways to choose a test (any test) didn't get me anywhere.
But thanks for trying something new and don't give up!
It made sense to me. I saw "Try Now", so I clicked that, and then it took me to a page of a bunch of jobs, which kind of confused me, so I re-read Ian's comment, and I clicked on "Start" on one of the three free tests at the top, and then just followed the test.
Why do militaries split CO/NCO ranks? I assume there's a good reason for it, since people have tried mushing them back together again and then gone back. What actually happens that makes the CO/NCO split so valuable/useful?
In a war, you want your front-line leadership to be young.
Young people are less risk adverse, more willing to take a chance, more willing to try something new. A successful military will have a young person leading combat units full of other young people.
Splitting the officer and enlisted ranks accomplishes this. As both officers and enlisted get older, they move higher and also are further removed from where the fighting actually occurs.
The book "War Beneath the Sea" by Peter Padfield describes how terribly the submarine force performed at the beginning of WW2 when the Commanding Officers were all career officers who had slowly progressed up the ranks and when they were all replaced by very young new officers, submarine warfare was reinvented quite successfully.
To follow up on the "class" explanation that a few people brought up, I believe the distinction is useful to keep the "soldier" and "officer" classes separate and prevent the development of a single "military class." Having a single class that almost by definition is better at violence than everyone else seems really bad from a coup perspective.
There's some overlap at the lower levels, but simplistically speaking commissioned officers decide *what* the unit is going to do, and NCOs are then in charge of *how*. These are two different jobs, with two different focuses. The officer has to be looking upward and outward, at how the unit's actions will affect the larger battle. The NCO has to be looking inward and downward, at what his team can do with the resources available. The officer has to command, the NCO has to lead(*).
And since "what are we going to do" sometimes comes down to "a really dangerous thing that half of us are not coming back from", it helps if you can set up commissioned officers as godlike authority figures speaking from on high, and NCOs as Just One of the Guys, we're all in this together.
* It's helpful if the officer can also lead and the NCO can step up to command in a pinch, but the primary focus will be different.
Interesting question and I don't have a good answer, but worth observing that medicine seems to have something similar with the doctor/nurse divide. A big part of the distinction in both cases seems to be practical, on-the-job experience vs more education and theoretical knowledge?
Every profession I can think of has a divide like this: Car mechanics, Automotive engineers. Boiler operators, system engineers. Clerical staff, management.
Ok, qualification--every profession that is focused primarily on the operation of some set of equipment seems to have this division. For example one exception are teachers--teacher's aids are generally graduate students who are in fact expecting to one day be teachers. But they are not primarily focused on operating equipment (of all the ways you could describe kids...)
It seems to me that there is some fundamental difference between becoming highly skilled at simply operating and maintaining a certain type of equipment, from understanding how to design or operate it in context of a larger system.
Organisational complexity. Think call centre employee vs senior customer relations officer.
The call centre drone gets a defined script and has zero decision making capability. You can't give this person any ability to do anything outside the script. Like if you worked in an insurance call centre, the call centre guy cannot be allowed to invent a customised solution that's not in his script - the organisation very possibly doesn't have the capacity to execute millions of bespoke promises made by thousands of call centre employees (and also, how do you make sure thousands of employees deliver consistent outcomes and more importantly, resolve things in ways your company can actually physically deal with?)
If you need to tell thousands of people to do a job, you had better provide extremely clear instructions and have a screening/ongoing regulation process to make sure those instructions are always followed to the letter.
The senior customer relations etc does have the power, because there's like 5 of them total and they're vetted to be competent and they've got a good idea of what the company can actually follow through on. And they are in charge of pointing the thousands of people who are just following scripts/orders in the correct direction, by noticing when environments changed or inputs changed and doing stuff to adjust outputs.
Being a cog in the machine vs operating the machine. (In good labour markets, being a cog isn't necessarily miserable - you know you have some damn good conditions when no one wants the promotion to team lead because everyone knows cogs get cared for and maintained to keep up performance, but the person operating the machine is doing the really hard job of making sure all the cogs work well and machine is pointed in the right direction and so forth)
Well, I'll be pendantic and point out that your example is focused on the distinction between the front line workers and their direct supervisors. I would think that's more akin to the enlisted and the sergeants above them. I'm frankly not sure who the equivalent to the commissioned officers are in your scenario.
But otherwise I think your description is pretty spot-on.
I don’t know if he’s the first to write about this or just one of the more influential, but justification for this split goes back to Clausewitz. From what I remember, his claim is that you want highly experienced people at all levels of an army, instead of eventually ending up with all your experienced people in the officer corps far away from battle. That’s the value of the split
Oh it goes back much, much further than Clausewitz. I am not as familiar with other military structures of antiquity, but the Roman Republic absolutely had an NCO/commissioned (obviously not called that contemporaneously, but functionaly equivalent) officer structure starting from c. 367 BC. Since the Roman military structure was successful at beating the hell out of all comers, frequently while greatly outnumbered and campaigning far from home, for 6 centuries on the trot, the basics of their structure have been copied for the past two millenia, including by essentially all militaries today (except for some Communist inspired ones, mostly to their sorrow).
As a prior Navy officer who went to the Naval Academy, I have also thought about this quite a lot.
I think it comes down to “class”, in a literal sense in the case of the military, where one *must* do what one of a higher class says.
At the enlisted levels, there is much more less of a “I must do what this higher ranked enlisted individual says” than there is a “I must do what this officer says.”
This is codified in law but also by the informal idea of “class” … an enlisted person is not and will never be (except through special means) an officer. Officers—under most circumstances—did not get to their position by being enlisted first. The separation creates an us vs. them mentality (where enlisted often resent officers), but is effective in terms of authority.
I think the class separation also does create a “leader” persona in officers—for better or worse. Often it *is* for worse, but when played out properly, military officers understand their role ultimately is to provide for and protect their men/women. In other civilian leadership roles I have not seen this as often. Rather, civilian leaders more often see themselves as simple administrators or managers of *things*.
That “leader” persona also plays out in higher NCO ranks… but often those leaders will ironically treat their lower ranked enlisted worse than an officer would treat them, for various reasons.
Also, as mentioned by others here, the officer role attracts high IQ, high talent people. If they had to rise through the ranks of regular enlisted, they simply wouldn’t join. The prestige is half the draw for the uber-talented in the military.
After reading about this idly over the years, the shot version I've boiled it all down to in my head is simply,
"The NCO is loyal to the boys, since he was promoted from the boys. The Officer is loyal to the State, since he was promoted by the State. The State keeps the Officer around, because it fears the NCO may not be able to make the tough decisions for the good of the State when the time comes, like ordering your men to die taking that hill."
If I have the time to elaborate, there's some examples I can point to, to justify why I think it's reasonable to oversimplify a no doubt complex phenomenon to this extent:
* Basically every military ever that has abolished the Officer rank, has brought them back; making the men vote on whether to die taking that hill, turns out to be bad for the State, even if it's good for the men. (The PRC is the notable exception, I understand.)
* Older school officers are taught to not get too chummy with the boys, and maintain a healthy emotional separation. Newer school officers are taught to be more personable. This is a vast oversimplification of course, but the older school officers act like that because back in the day, it was simply normal to have to order your men to die. That hasn't been really necessary in the "Global War on Terror" era, since casualty rates aren't that high, but if we go back to fighting big conventional wars, we'll probably have to go back to teaching officers how to order their men to die, without dying inside themselves.
* The historical origin of the Officer/NCO system, at least in Europe, is the Aristocracy vs. Commoner divide; the Officer is a noble, the NCO is a Commoner, and the lowest Officer can overrule the highest NCO because that is simply how it works everywhere else in society. (And the NCO cannot promote into the Officer ranks, because that is also how it works in the rest of society as well). This division applies even at the level of grand planning instead of "boots on the ground" fighting: the famed Prussian "General Staff" system arose from the need to combine loyal but incompetent nobleborn generals, with clever but potentially *too* clever commoners being promoted to General position to lead. The answer turned out to be giving all the *work* of being a general to the commoners, and calling them the General's Staff, while the noble keeps the *title* of General, and thus the authority to overrule them if he feels they're being disloyal to the monarchy. (But in day to day operations, he mostly just rubber stamps whatever they say. Analogies to AI are left as an exercise to the reader.)
* Navies tend to have an even bigger social separation between enlisted and Officers, because the problem of potential disloyalty is even worse. It has something to do with the nature of ships, but I don’t remember why since I don't study naval matters in as much detail as I do land combat. But there's a reason why officers in the Navy eat in separate messes from the men, fraternizing solely with each other instead of with their men, in a way Army officers find surprising. Also related is how the original purpose of the Marines in navies around the world, is to shoot your own sailors / protect the officers against mutinying sailors. The bigger an issue disloyalty is, the more you have to promote for loyalty instead of competence.
Not sure about your army/navy comparison: part just-so-story, part everywhere-is-America.
From a UK perspective, and with some expertise in 18th century history:
1. The original purpose of Marines was definitely as combat troops: intermittently from the 1660s, regularly established from the 1750s. The first references to the idea of the Royal Marines as a protection against mutiny come in the 1790s (post French Revolution and the first period of widespread class consciousness).
2. In the UK it’s the navy that has had the more socially mixed officer class, despite being higher status (the ‘Senior Service’) and more important to the nation. A number of reasons for this, including because the navy was more technical (you need to learn the trade of seafaring to be effective, however aristocratic you are) and because the navy needed a more full-time commitment (army officers could be absent from their regiments most of the time until the twentieth century).
So in the formative periods of both institutions most army officers were directly commissioned, while every Royal Navy officer started in the ranks (albeit some in what we’d now call an NCO role). And unlike in the US, army officers very much eat separately from their men (at least in barracks).
I'm possibly with Melvin on this being a class divide. Officers come from the demographic where everyone has a college degree, and typically enter the armed forces directly as an officer after completing a military academy course. NCOs are typically promoted from the lower ranks after entering with a high school equivalent education.
In practice, I don't think it makes much of a difference who does what. For instance, NATO armies typically place a lot of low-level coordination (below platoon level) on NCOs, typically senior sergeants. Russia has a much lower ratio of NCOs:COs, but has a lot of low ranking COs like junior lieutenants that assume similar duties. I think the US separates commissioned ranks definitionally, because COs are given high level formation command and administrative duties, while NCOs are given low level duties. But it doesn't have to work like this.
I would be interested to see what someone with an actual military background thinks though.
Former USAF officer here.
The duties & responsibilities are just very different, in a way that shows up in civilian contexts as well.
NCOs are your expert ICs with some supervisory & mentorship responsibilities; even at SNCO levels it's still fairly tactical ("how").
Junior officers are management, more operational ("what").
Senior officers are executives, strategic focus ("why").
An NCO pursuing a commission is an IC trying to get onto the management track. Not all can succeed and not all of those even want to. Those that do are typically much better junior officers than those of us who commissioned at 21. But they rarely proceed very high into the field grades if they get there at all (their enlisted time means they're pension-eligible at a lower commissioned grade).
ETA: Junior enlisted can also pursue a commission even if they haven't made NCO; they're younger and less experienced so don't differ from other junior officers nearly as much as former NCOs (primarily just apt to avoid the sorts of stupid mistakes that butter bars are prone to).
Note that there is a 3rd category that comes and goes: Warrant officers.
In the US military Warrant officers slot between Commissioned officers and Non-Commissioned officers so a low ranked Warrant officer (W-1) would outrank a highly ranked NCO and a low ranked Commissioned officer would outrank a highly ranked Warrant officer.
I don't have an answer to your question, but a good answer (from someone else) will account for Warrant officers.
What I've always heard about warrant officers is that they are "highly skilled, single-track specialty officers", technical experts of various specialties which change through time as military tech evolves. That quotation is from the relevant Wikipedia article but it tracks with what people such as my father (a former USN junior officer) always said.
I did not know until reading that Wikipedia writeup that the above tracks from the origins of warrant officers: in the British Royal Navy during the era in which age-old tech (wooden warships) was being updated with new and mission-critical tech (rows of cannon capable to blasting apart enemy wooden warships). So you needed a certain class of gunnery experts who didn't need all the general knowledge of an NCO or career officer, but instead possessed deep knowledge of that particular ship's system.
What the Wikipedia article doesn't address (or I missed it ...) is why Warrant Officers as a rank seem so different between the branches.
The US Air Force, for example, used to have Warrant Officers and still authorizes them but currently doesn't have any.
Navy Warrant Officers seem to come from the Petty Officer ranks (so you get promoted from some sort of Petty Officer to Warrant Officer rather than to a more senior Petty Officer if you go the Warrant Officer route). So in some sense Navy Warrant Officers are similar to the Army Specialist ranks.
The US Army seems to use Warrant Officers a lot.
If this was just a technical specialty I would expect the Air Force to have more than the Army.
And I still don't have a good (non-just-so) explanation for why the CO and NCO split :-)
Re: Warrants
The branches all agreed to sunset their warrant programs in order to be authorized to have SNCO grades. Only the Air Force followed through, but everyone got the SNCOs.
I did not know that!
That makes it seem like the *intent* was to:
(a) Change the Warrant Officers to Senior NCOs, and
(b) Because of this the WOs would be promoted from the NCO ranks rather than be appointed independently.
Which is basically the change to NCO->CO that Chastity is asking about, but with NCO->WO instead.
Possibly? I was USAF so only ever met a WO briefly once; other than Army rotor heads I don't know what they do.
ETA: my equivocation is because I'm not sure what you're indicating with the "->"
It prevents the whole military from becoming a low-class affair. Middle class and higher people don't want to join if they're going to have to mix with Billy Bob and Tyrone.
Romans had centurions (non-commissioned) and tribunes (commissioned).
The camp prefect was the highest non-commissioned rank, and he ranked above the lower tier of tribunes, meaning that the distinction didn't necessarily match degree of authority.
(edit: I understand your question, I wasn't trying to provide an answer, it was just an observation.)
Right, I get that this was really common across all time and space, and the places where they do otherwise tend to be trying some type of utopian experiment that fails and they go back to the old way, I just don't get *why*. What's the actual, nitty-gritty, practical-structural reason that sergeants don't get promoted into being lieutenants and so forth? What happens when you do it?
"What's the actual, nitty-gritty, practical-structural reason that sergeants don't get promoted into being lieutenants and so forth? What happens when you do it?"
I still am not able to answer your question, but ... sometimes sergeants DO get promoted into being lieutenants. The route for this is that the sergeant goes to OCF (Officer Candidate School) and comes out a 2nd Lieutenant.
https://en.wikipedia.org/wiki/Officer_candidate_school#United_States
There is even a term for an officer who has gone this route: Mustang.
https://en.wikipedia.org/wiki/Mustang_(military_officer)
I think having a 4-year degree is a big career edge for commissioned officers in the long run... but a number of enlisted soldiers have 4-year degrees.
A related question might be: "Why do so few army sergeants NOT go to OCS?"
It’s very difficult for enlisted personnel to get the opportunity to become an officer, basically.
Also, by the time one is a Sergeant, they realize being an officer is a lot of responsibility and a lot of paperwork. Much less technical work and being out in the field. This likely is not very enticing. Also, a Sergeant is like the 6th enlisted rank, and they’d have to start at the bottom of the officer ranks, which some may not like.
Sergeant in an E5 and one can make it to E5 in three-ish years. Four years is doable. So four years as an enlisted critter and then OCS to become a 2nd lieutenant.
Clearly, most commissioned officers don't go this route so it can't be terribly appealing to most folks who make sergeant and choose to remain in the army. Probably because, as you mention, the jobs are different and if one COULD be both an officer or an enlisted critter and chose to be an enlisted critter one likely doesn't WANT to be an officer.
I still don't know why the military doesn't start the officers off as enlisted critters and then (maybe after serving a tour) send the ones they want to be officers off to OCS.
Heinlein's military in Starship Troopers worked that way so Heinlein didn't think it was *clearly* preposterous.
But roughly no military does things this way so there is likely a reason.
I mean, I know you *can* go from NCO to CO, but it's not sort of automatic in the way that, say, O2 to O3 is (O3 is competitive so not everybody gets it, but you don't have to go to a special school or anything), which is what I really meant.
> A related question might be: "Why do so few army sergeants NOT go to OCS?"
According Master Sgt. Milt Warden (played by William Devane) in the 1979 television miniseries version of ‘From Here to Eternity’ it’s because someone has to be around
“to swab out officers’ mouths so they don’t choke to death on their own spit.”
The way I interpreted his sentiment was identification with a particular class.
In the final scene he grudgingly puts a sheet of paper into his typewriter to apply for OCS though.
Recent events have made me wonder if Thomas Matthew Crooks would have gotten more approval had he been better looking.
His name doesn't help.
Chris Rock did a joke about good looks and public opinion about vile behavior re Luigi.
“If he looked like Jonah Hill, he’d be in the (electric) chair already.”
Which was precisely as comedic as the rest of his worst-ever monologue. (Seriously, watch it. I've never seen such a trainwreck from such an otherwise exceptional comedian.)
Chris dropped the ball here. The joke should have been "If he looked like me..."
It ain't necessarily just sheer physical attractiveness that matters.
That would have been a different joke. The audience would have taken it to mean "If he were Black..."
Both funnier, and more accurate.
I think it's more because he didn't succeed. You don't get points just for trying.
AT, is playing games. Unless he subsists on MSM or the silliest parts of reddit he knows that looks aren't relevant to the matter.
I published this before there was a single photo of the shooter available and all that was presumed about his motivation came from his bullet casings.
https://ydydy.substack.com/p/brazen-manhattan-murder-of-a-monster
But I won't belabor the point with those who are either knowingly arguing in bad faith or are so very well buffered against the vicissitudes of life that they literally can't see the obvious.
Your point (anomie) is true. Had he succeeded, some silly folk (far from power) who have been devouring nonsense about the importance of "our democracy" since grade school would have immediately jumped on the badwagon.
They would have been "January 6th" very very quickly, and it would have had nothing to do with the shooter's looks.
But, as you noted, he failed, and people hate failures.
Channeling Mencken I simply shared what he would have had he not been subject to the misfortune of being dead.
https://ydydy.substack.com/p/some-thoughts-are-better-left-unsaid
Could you please slow down *a lot* with promoting your blog articles? The norm here is like two links a year, and you have three in this Open Thread already.
I'm not sharing anything that isn't relevant to the matter at hand [and I'm certainly not "promoting" "blog articles"].
If I've already produced a 30 minute video directly addressing murder from a Rabbinic Perspective and how it paradoxically explains and (perhaps) justifies the killing the Brian Thompson, it would be silly not to include it within my response.
If I were just dropping links and going home you'd be right, but what I'm actually doing is speaking at some length to the questions at hand and giving interested readers the relevant footnotes and additional details should they choose to research the view further.
Plenty of people from ACX have found value in these off-page addendums. Some haven't.
But THOSE WHO HAVE tend to be those who are here for more than just swapping fantasy riddles and they tend to come with their own motivational steam and learning, so I'll continue to share what I think is important so long as I continue to receive an appreciative audience for it in my correspondence (if not here quite as much, due to the too-common pile-on experiences that, though less common here than on Reddit, are still far too common here as well).
Perhaps you could give the tl/dr version for those of us who don't like following links all day.
So, what do SSC/ACX folk think of Luigi Mangione?
Turns out Cory Doctorow wrote a story about something similar, published in 2019. He has some things to say here -
https://pluralistic.net/2024/12/09/radicalized/#deny-defend-depose
I can't speak for anyone else, but I think there ought to be consequences to doing Molloch-y things. I don't know if this particular CEO was doing anything particularly bad, as I don't know much about health care, but CEOs generally have a bad track record these days from what I've read.
If you compare worlds where CEOs never have any consequences and worlds where the worst offenders occasionally get assassinated, which type of world turns out better on average?
I don't endorse killing in general because thinking that you know which people make your world worse and which make it better seems very arrogant. In my experience, it's difficult to tell the difference between certainty (because you've thoroughly analyzed evidence that you've seen with your own eyes) and certainty (because you've read some stuff online and you're feeling particularly manic that day).
Luigi is a murderer, the people who are cheering him on are barbarians, and this will not end well. Either we will do the right thing and pointedly ignore Luigi's demands that we Do Something(tm) about the Terrible Evil(tm) of health insurance companies, in which case the cause of fixing real problems in the insurance industry will be set back months or maybe years, or we will take the bait and publicly commit to Doing Something, in which case we'll get more assassinations.
And even if you believe Luigi's was a noble assassination (it wasn't), I guarantee you'll find at least half of the copycats to be wholly abhorrent and antisocial murderous thugs (that part will be true) killing the virtuous and innocent (ish).
Also, if anybody here believes that Luigi's was a noble assassination, did you put any real effort into investigating whether UH's claim rejections were part of the legitimate and necessary baseline for any health care system, and if not whether Brian Thompson was the cause of the problem as opposed to being a reformer trying to quietly fix it from within? Or was it just "CEO in an industry I don't like, so he must have had it coming, they all do"?
Yeah, I thought so. The people who now assassinate the champions of the causes you favor, will be just as careful about that sort of thing. We're all familiar with "Days of Rage" here, right?
>Either we will do the right thing and pointedly ignore Luigi's demands that we Do Something(tm) about the Terrible Evil(tm) of health insurance companies
I'm not aware of Mangione explicitly having demanded that the public do anything. (This could be to his benefit regarding charges of terrorism. Those relate to trying to influence behavior through crime. But) I don't think he demanded any changes; he didn't demand that insurance companies do something different, or try to inspire the public to rise up against insurance companies, for example. Instead, his manifesto just vaguely connects his action to his listed grievances - healthcare is expensive, so insurance is bad, so I killed a guy.
He is a symptom, not the disease. And it ain't just healthcare. I think we are going to see more of this sort of thing. Which is regrettable.
Young male of above average intelligence begins acting strangely -- makes me wonder if he is schizophrenic.
He's the ultimate Karen
I don't think that could be true, or "I want to see the manager!" will never cause a meeting with a manager.
Foolish child. Long on self-righteousness, short on wisdom. He's a murderer and a terrorist and I hope he gets executed.
Killing people is bad, period.
I also think that health insurance might be unnecessary in many ways. I would rather live in a world where hospitals list prices for various treatments up front and people just pay them, as is the case in many other industries. Alternatively, we could treat healthcare as a public good (as it does have very high positive externalities, economically speaking) and just give it out to everyone for free.
He reminds me of the Anarchists of the 1920’s.
Sacco without Vanzetti
We like to blame systemic failings on the "moral" failings of individuals, but that is not a useful lens. The moral failings of individuals, or the extent they are acted on and damage society as a whole *are* systemic failures.
While cliche, it says something about society. The system is broken.
In this case it resulted in the death of a CEO, but it results in the deaths of many more people, in way less flashy ways, every day. Which is bad, independent of any karmic worth we ascribe to anyone in the situation.
The machines we made to fix the machines when they break are broken, so the institutions will slog along creating needless pain and emiseration until they finally collapse under their weight and new institutions can emerge.
If the system was capable of metabolizing signals like "CEO's are getting shot in the street while people cheer" and actually reform our crumbling institutions so that people can go back to ignoring them and living their lives, then we wouldn't have these problems in the first place.
But yeah of course, surely *our* imperial decline will graceful.
By the way, the US economy will collapse by the end of January, so make sure to pack your stockings this Christmas with gold, treasury inflation adjusted bonds, and shotgun shells.
If the system was capable of metabolizing signals like "CEO's are getting shot in the street while people cheer", the signal it would metabolize them into is "Shooting CEOs in the street is a good way of getting what you want."
Or more plausibly "The CEO's compensation package should include armed bodyguards."
If it comes to that, sure, though I suspect the vast majority of people would prefer not shooting people to get what they want, given the choice. But that choice has to exist, and it's the job of the institutions to create that choice. If they fail to do that, shooting it shall be.
Given the widespread differences of opinion surrounding this topic, I don't see how it's possible for the vast majority of people to be given the choice between getting what they want by shooting CEOs and getting what they want without having to shoot CEOs.
If you believe that "differences of opinion" are still the deciding factor whether or not to take action, then the pressure of suffering is apparently not high enough just yet. But it will be. The longer you wait, the more catastrophic the rupture will be.
For solutions, you can always look to just about any other industrialized country in the world, it's a solvable problem for sure. Throwing your hands up and declaring the problem too hard to solve is simply a choice for an unsustainable status quo, with all the eventual consequences that entails and a glimpse of which might be Luigi Mangione.
I believe differences of opinion are the deciding factor because that's what a democrat believes. What's stopping us from having the sort of system you prefer isn't some sort of interference field obstructing the will of a unanimous public-- it's the fact that a lot of us aren't in favor of it. If you do manage to get your own way, there will still be plenty of people mad about that. Maybe one of them will shoot Bernie-- in which case I expect we'll be hearing a lot less about how the mere existence of a terrorist assassin proves that Something Must Be Done.
This. People don't like shooting people.
Though there is very little evidence that shooting people will cause any positive change either.
People want to feel like they have control over that which makes them suffer, because the alternative is that they have no control.
Sometimes that's sacrificing a lamb to appease the storm god.
Sometimes that's shooting a CEO.
It's magical thinking and voodoo either way (also literally voodoo, which developed as a way for enslaved people to curse their enslavers), but people doing those things is a signal that contains information on, which can be acted on rationally.
As a first order derivation, yes.
I suppose I should have phrased it as "properly metabolizing".
I'd bet that message will be the outcome of this situation in particular.
But that is not healthy.
I'm not saying you should blindly react to terrorism, which is what this effectively is (and I intend to mean that in a value neutral way), but you shouldn't blind yourself to it either.
It should be a signal to step back and analyze how this situation got to that point.
If you respond to a healthcare CEO getting extra-judicially executed in the street while the mob cheers with making common sense reforms and regulations, then CEOs would stop getting shot *and* the lives of millions of people would improve.
If you respond to the situation by locking in and not meaningfully changing anything out of principal or incapablility, then it will continue and escalate until more and more things break, while the system continues to hurt and emiserate people.
Is that in itself just? I don't know and I don't care, it is what it is.
And of course the first case will never happen, because again, if the system was capable of making common sense reforms and regulations in response to signals, it would have done so much sooner.
Which leads to the second case, more and more bits of the system breaking, which sucks for us because we live in that system.
You almost stumble on the truth in the last two paragraphs, but manage to walk past it in the end. If there really were common-sense reforms lying around unimplemented, that would be a puzzle-- which should at least lead us to suspect that there aren't, and that the "common sense" ideas are actually controversial ideas.
Addendum, looking at your other comments, I think I understand your position and logic, please correct me if I am wrong.
Democracy is the worst form of government except for all the other ones (I agree here)
America is democratic (I disagree here)
America does bad things
Therefore the demos, the people, are bad
Extrapolating: America will continue to do bad things until Americans are less bad
So you believe we do have a functional political and economic system and everything is working as intended, because nothing better is possible or feasible?
Do you think there are no entrenched private interests in the American economy, especially in *healthcare*, lobbying against legislation and policies which would be beneficial to everyone overall?
Even if there are not simple solutions, assuming the system was functioning, the current US state should be full of the greatest statesmen, economists, and political scientists in human history, presumably they could figure something out and convince the rest of us to support it.
Or is your position "bad things happen because bad people are bad" and any further analysis is somehow just excusing the bad people, so we shouldn't do it?
Well, the particular bad thing we're talking about did happen either because a bad person is bad or because a crazy person is crazy. Though I can certainly understand why someone might be happy to have a news peg for a political argument they're itching to make anyway.
"The best form of government except for all the others" fits my view reasonably well-- though nothing resembling perfectibility follows from it, and I can't even imagine where you came up with some of the other stuff. What I really believe about democratic decision-making is that, while it's generally inferior to market decision-making, it's still capable of a sort of bounded rationality. There's no power on Earth that can make the public believe that the sort of insurance system mentioned by Mark Roulo downthread, where no one ever says no to them but it's still really cheap, is not available: we killjoys are stuck trying to prove a negative. Yet, if you put a specific reform plan before them, they've now got something relatively concrete to grapple with, and with the help enough public debate they can do a reasonable job of comparing it to what they've already got and deciding which axtually-available alternative they like better.
Some people really don't like common sense.
Tip O’Neill once famously said “all politics is local.“
I think in a lot of ways, the same holds true for common sense. The United States was built as an amalgamation of competing interests. Everything is a horse trade; what is common sense in one county might just be crazy in another.
Restricting the phrase to things people actually agree on isn't that big an ask.
I think it's likely he did too many psychedelics and popped a screw loose or three. That or he's just a f$%^ing idiot. Even supposing that killing a health insurance CEO was justified, just for a larf, it accomplished nothing, because there are about 4,000 people lined up to take this guy's job now, only now they'll probably demand their own security entourage, which will be paid for by United's subscribers. So thanks, dude. You just made everybody's health insurance that much more expensive.
https://www.npr.org/2024/12/05/nx-s1-5217617/blue-cross-blue-shield-anesthesia-anthem
Within a few days, BCBS announced the plan to limit anesthesia coverage, CEO gets shot, the plan is scrapped. Whether or not the shooting had anything to do with it directly, it certainly heated up the discussion and increased the pressure on BCBS. And that's just the most obvious, short-term impact the shooting possibly had. It's impossible to say that the shooting had a positive effect for sure, but it's equally impossible to say the opposite apart from the shooting itself, of course.
The axed BCBS proposal has been widely misreported and "limit anesthesia coverage" is definitely not an accurate summary. Here's a better explanation of the proposal (on Reddit of all places): https://www.reddit.com/r/actuary/comments/1h70wic/does_anyone_have_context_or_details_regarding/m0nyrv3/
The shooting having a positive effect on healthcare is the worst possible outcome of this ridiculous event.
Wow, that's a self-own. I didn't think it could be done, but you did it!
You know how sometimes people or companies or countries do really really bad things for a long time until someone loses it from some uncontrollable visceral location?
That's what the American People did last week.
The murder would have made ZERO effect if not for the HUNDREDS OF MILLIONS of Americans and allies abroad saying: "YES. DESERVED. FREE LUIGI."
You can ignore that response, or perhaps make things even harder for them, but it can easily be argued that granting Americans some relief was not only wise BUT NECESSARY, and out of the CEOs' control.
Millions of people work in healthcare and many of them were no longer willing, for $11 and hour to be the bearer of bad news.
You're either too young or too rich to have been unable to intuit this. But hey, nobody forced you to self-own quite as hard as you thought wise.
In your opinion/experience, when you give terrorists some but not all of their demands, would you expect them to compromise, and consider whether the remainder of their demands is worth continued violence, or would they be emboldened by your display of weakness and press for ever more?
What if you give in to ALL of their demands?
>In your opinion/experience, when you give terrorists some but not all of their demands, would you expect them to compromise, and consider whether the remainder of their demands is worth continued violence, or would they be emboldened by your display of weakness and press for ever more?
I'd expect them to compromise. This happened with the IRA in Northern Ireland.
Islamic terrorists won't compromise, but Islamic terrorists are a special case.
Hardly. Luigi Mangione is a symptom of the decline of the US system of democracy, not its cause. There is no End of History, the system of democracy has to continue proving its worth to people, or they will eventually reject it wholesale, with far worse consequences than a single countable death.
Political violence is an ancient, proven act of politics. It will be the fallback if and when peaceful methods fail. The shooting is an opportunity for US democracy to correct course, and if it manages, it will take off the pressure that would produce the next shooting or worse. So I believe the worst possible outcome of this shooting is precisely that nothing good comes out of it.
Dude, come on. One random act of violence is not going to change the economics of the US healthcare system. Did you do too many psychedelics, too?
Do you really think that the only people who must be swayed are the ceos?
A groundswell of hateful anger against this particular aspect of The System makes millions of people who work in the industry uncomfortable.
They aren't afraid of getting shot, they're afraid of being the bad guy at the dinner table or party.
There's A LOT going on here.
Look at it from an economic perspective: do you think it's possible to simply terrorize your way to better, more affordable health insurance? I don't. Shooting the CEO of Honda will not simply scare the company into selling new Accords at the same price as a Kia Sorrento; at least not in the long run. These are two different products with two different cost structures, and the profits on each are a function of the supply and demand for either one, along with how competitive the market for cars is. Insurance is no different. Adding self righteousness and indignation into the mix will not affect anyone's ability to provide a better product or service. Sorry.
I agree with your proposition. I’m not sure the car market and the healthcare market are that analogous though. There is something about the demand side in healthcare that seems an anomaly compared to cars
>One random act of violence is not going to change the economics of the US healthcare system.
You certainly seem to think so. You said that the extra security detail for a couple of CEOs "just made everybody's health insurance that much more expensive." Also, please tone down the ad hominem.
I think this conversation has run its course.
https://www.noahpinion.blog/p/insurance-companies-arent-the-main
I'm an MD and even through my biased self-serving prism I can see that it shows a crazy amount of stupidity to go after/celebrate the killing of insurance company workers when it's hospitals, specialists and Pharma that are jacking up the prices. But yeah - you get the bills from them. Shoot the messenger isn't a new concept.
"Shooting the messenger" is not an appropriate metaphor here. The "messenger" makes a healthy profit himself and is therefore not a neutral party. Also, the messenger is in business cahoots with the road maintenance department, the king's advisors, the scribes who got paid to write the message, and the Pony Express:
https://www.livemint.com/industry/who-profits-most-from-america-s-baffling-health-care-system-11720863433863.html
In the immortal words of Detective Jake Peralta, "Cool motive, still murder".
More specifically and less flippantly, I have two big problems with the killing and with people cheering it on.
On the object level, I think that the case for Brian Thompson's death being justified from either a utilitarian or deontological perspective is much, much weaker than a lot of people seem to be assuming. I haven't examined in detail the premises that UHC is an unusually shitty health insurance company in terms of denying claims, and that Thompson was responsible for this, but on both counts the charges seem plausible and I'm prepared to at least accept them for the sake of argument. But going from that to "Thompson is morally responsible for thousands of deaths and killing him was a praiseworthy act" is a pretty huge leap. For one thing, the relationship between marginal medical care and either life expectancy and QALYs seems to be pretty fuzzy at a population level, c.f. the debate between Robin Hanson and Scott over medical effectiveness earlier this year, where Hanson argued that marginal health insurance coverage was almost pure waste and Scott's conclusion started out thus:
>The insurance literature doesn’t do a great job in establishing one way or the other whether extra health insurance has detectable health effects on a population. Gun to my head, I’d say it leans towards showing positive effects. But if Robin wants to fight me on this, I can’t 100% prove him wrong.
For another, people choose to buy UHC plans on the exchange, and employers choose to offer UHC group plans instead of Anthem or Blue Cross or Aetna or whatever. If UHC is killing people by excessively or unreasonably denying claims, then some of the blame accrues to the people who are choosing their plans, to state and federal regulators for providing ineffective oversight (if UHC is misrepresenting their coverage or otherwise exploiting opaqueness of the exchange system to sell shitty insurance to unwitting subscribers), and to the framers of the ACA for setting up the rules under which UHC operates.
I think the actual object-level complaint charge against Thompson is that he's a symbolic stand-in for the American health care system. I agree in general that the American health care system is a terrible mess, although I disagree with a lot of the conventional wisdom of specifics of causes and cures. But the actual individual who got murdered is a relatively weak stand-in for the target people seem to want shot.
On a meta level, the implied principle isn't that vigilantes should kill high-profile individuals who are objectively guilty of causing (or making substantially worse) major societal problems. Nor even that vigilantes should kill those who you think are sufficiently villainous. The actual principle that is being endorsed by cheering in the killing of Brian Thompson is that vigilantes should kill people *they* think are sufficiently villainous. This is an extremely dangerous principle, and one which I very much do not want endorsed in any way, since the sorts of people who are apt to go out and shoot their chosen villains are likely to be among the very worst people to make that sort of judgement. At best, people are going to get shot (or at least shot at) for bad reasons. At worst, assassinations and other retail domestic terrorism gets normalized and radicalizes more people into supporting similar acts in perceived retaliation, which is how civil wars get started.
The game theory take is you actually do want a few defectors/madmen acting on principle sprinkled about to keep everyone warned. But by definition you have to denounce and punish those actors. It only works if they act self destructively on principle and cannot be expecting some folk hero reward
> In the immortal words of Detective Jake Peralta, "Cool motive, still murder".
+1
Thank you for posting this. 100% agree.
Even if Thompson were a good symbolic stand-in for the failures of the US healthcare system, potentially throwing your life away to destroy his on behalf of a bunch of strangers, many of whom are already dead...well, that seems like it fails any utilitarian calculus.
"But going from that to 'Thompson is morally responsible for thousands of deaths and killing him was a praiseworthy act' is a pretty huge leap."
Maybe the leap for the people cheering Mangione is more: "*NOW* maybe someone in a position to do something will pay attention to our concerns????"
The catch is that the insurer is mostly just in the middle between the folks paying premiums and the folks making claims. If UHC approved 2x as much care (measured in dollars) then premiums would have to double. People want lower premiums and more health care ... that won't happen because of major changes to the *insurance* companies.
And ... if I'm reading the United Health Care financial statements correctly they took in $367 billion last year in premiums and paid out $280 billion in claims. That leaves $87 billion. They then claim $59 billion in operating costs (salaries, rent on buildings, ...) leaving about $28 billion in operating income. $28 billion of our $367 billion is 7.5%.
I expect much of the rage is unquantified ("I want MORE care and LOWER premiums), but I expect that the CxO folks running the company would cut the $59 billion in operating expenses if they could.
If run at cost it looks like everyone's bill would drop by about 7%.
But ... MORE care and LOWER premiums.
But UHC rejected 37% of claims, and BCBS only 8%. BCBS works as a business, so it must be possible to run an insurance co. without being unreasonably stingy about paying for needed health care.
Maybe BCBS plans just cover less stuff from the jump.
How do BCBS's premiums compare to those of UHC?
I haven't checked, but I suspect UHC is substantially cheaper. Not necessarily proportionally cheaper to the claim rejection rate, since:
1. Not all rejected claims are equal in cost. A claim might be a major hospital procedure or a 30 day supply of a cheap generic drug.
2. Not all claim rejections stick. It's pretty common in my experience (as a patient with Aetna insurance) for a claim to be rejected, then the doctor resubmits it with minor modifications and it goes through, or they substitute a slightly different procedure or medication that is covered without fuss.
But to the extent that UHC pays out less money by rejecting more claims, I would expect that to be strongly reflected in premiums, since one of the ACA's major provisions was requiring health insurance companies to spend a minimum of 80 or 85% of premiums on claim payouts, capping profit, admin costs, marketing, etc at 15-20% of premiums. The 80% minimum applies to individual and small group policies, and the 85% minimum applies to large group policies.
Thanks for this sensible and atypical ACX comment.
I think you're probably right about that. It's my observation that people with well-paying jobs tend to have BCBS, people with low-paying ones tend to have UBH. (I'm a psychologist, so hear from patients and potential ones what their insurance is.).
UBH is loathed by therapists. In the past I have had 2 patients with UBH, and for both the company made big mistakes (in their own favor) in payments to me, and straightening it out was a huge hassle, involving multiple phone calls with different people telling me different things, etc etc. After that I refused to take UBH any more. Now I have a patient with UBH who pays me out of pocket and submits my receipts to UBH to get reimbursed, and they are doing the same thing to him. They announce that before paying him they want various kinds of documentation from me about treatment plan and progress. I send it well within the deadline, but their next reimbursement check to him is for $0 with a note saying "required documentation was not received." So then he calls and tells them the date it was faxed to them, along with proof of the fax, and they say well send it again. So we do. But they still haven't paid because they say they are behind in processing things. He asks if they have received the documentation. They say don't know.
I don't object to a company that charges lower premiums covering few things and having bigger deductibles, but I do object to forcing patients or doctors to spend hours getting the money they are owed. Also I object to rejecting a claim that should be paid because it meets the company's own criterion. When I worked at a mental hospital staff had to sometimes fight to get more time covered for people who truly needed it. For instance, the patient might be covered for inpatient treatment if they were suicidal. Obviously there are degrees of suicidality. Still, there are situations where reasonable people would agree that the patient is at substantial risk. For instance they were hospitalized after actually attempting suicide. They have been in the hospital 7 days, and appear and report being deeply depressed. If you ask them if they are feeling suicidal they say they wish they were dead and are sure they would be better off dead, but they don't currently have a well-worked out plan for another attempt. Reasonable clinicians will conclude that somebody who survived a suicide attempt by luck 7 days ago and still wishes they were dead is at very substantial risk. UBH was known for saying, yeah, but where's the beef? It's been 7 days and they haven't made another attempt or even scraped their arms with a fork or something. That's just dishonesty on the part of UBH. They don't really believe the person is safe for discharge, but they pushed, sometimes successfully, for a person to be discharged even though their policy said they were covered if they were a suicide risk.
After thinking about this a bit, I suspect what's going on is that UHC is trying to pare their admin costs to the bone because they have a fixed slice of their revenue for overhead+profit. What you're describing fits perfectly if they're pushing as much admin costs to providers as possible, using poorly-trained (and probably high-turnover) employees to process claims according to rigid scripts, and managing the employees with quotas that incentive them to find any excuse to get cases off their desk.
Maybe they function like the fake-ish auto insurers down here, that cater to people, especially newcomers, who just need a piece of paper saying they have insurance. My father was T-boned by a newcomer to our shores; when he took his vehicle to the dealership (! - it's an old person thing) they looked at this dubious info, and said - we'll just submit it to yours. Perhaps GEICO squeezed some money out of them.
Yes, that sounds incredibly shitty on UBH's part.
I want an explanation for how earbrows grow half an inch in 5 days before I rule out he was "framed" to be a ranting lunitic who will slowly bit by bit seem less and less credible as a reaction to the positivity online/governance failure
Untangling what we know about the shooter-shooter and luigi-shooter picked up at a mc donals with the murder weapon and an open and shut case, will just be hard, its like elon tweeting "put cocaine back in coke" a day before the amber herd trail where she talked about elons drug use or whatever so "elon cocaine" gets the tweet.
He's Italian. Shave him bald and he'll grow new Dukakis-level eyebrows in minutes.
Possibly, and to go even further he may have intentionally experimented on plunking his earbrows and counting the days for which day has the fastest change, while intentionally only showing his eyes
The whole thing about "imma taking a gun to mcdonals after avoiding the cops for 5 days" is fucking weird; did a cia "terror plot prevention by orginizing it" handler fail and then promise him to get rid of the gun? Was it a spy network and they made up the fake tip after unconsitutional spying (btw I heard the tip isnt getting the reward money)
Hero. And murderer.
We live in a day where, by Torah standards, we are ALL murderers.
https://ydydy.substack.com/p/brotherhood-or-the-sword-the-only
ALL THAT WE NEED IS A LIVE PUBLIC DISCUSSION to resokve our issues.
I am convinced that (other than the supposed psycopaths we are all told are irredeemable) nobody who works in the health insurance (or warfare, or gambling,...) industry is a monster.
However they correctly say to themselves that if they don't cause the deaths of certain people for profit than others will. PLUS they have endless justifications - dipped in the sacred blood of millennia - for what they do.
We need to meet as humans.
However, when people like Socrates, Jesus, or Gandhi get too big for their britches they end up dead. Smaller fry, like myself, are left alive to howl into the wind.
https://youtu.be/l4X6J_U0RyA
If this singular murderer allows Luigi, Myself, or some other individual as yet unknown to us, to have the honest Human To Human public conversations with the powerful few who actually own and run the world then ZERO more deaths will be necessary, and retroactively this one will have been cleansed as necessary.
If that fails to happen, then murders and other atrocities wilk continue day in and day out ----- and the killing of Thompson will have been just another one of no redeeming value.
> Murderer. And murderer.
Here, I fixed it for you. You're welcome :)
That what he did is horribly wrong I haven't the slightest doubt. What I think of him for doing it depends on whether he's mentally ill or just evil-- a question on which I'm agnostic.
Google did a massive study on what makes effective teams. It's worth reading if you're still toiling in the mines of Silicon Valley — especially if you're a manager.
>The researchers found that what really mattered was less about who is on the team, and more about how the team worked together. In order of importance:
> Psychological safety: Psychological safety refers to an individual’s perception of the consequences of taking an interpersonal risk or a belief that a team is safe for risk taking in the face of being seen as ignorant, incompetent, negative, or disruptive. In a team with high psychological safety, teammates feel safe to take risks around their team members. They feel confident that no one on the team will embarrass or punish anyone else for admitting a mistake, asking a question, or offering a new idea.
> Dependability: On dependable teams, members reliably complete quality work on time (vs the opposite - shirking responsibilities).
> Structure and clarity: An individual’s understanding of job expectations, the process for fulfilling these expectations, and the consequences of one’s performance are important for team effectiveness. Goals can be set at the individual or group level, and must be specific, challenging, and attainable. Google often uses Objectives and Key Results (OKRs) to help set and communicate short and long term goals.
> Meaning: Finding a sense of purpose in either the work itself or the output is important for team effectiveness. The meaning of work is personal and can vary: financial security, supporting family, helping the team succeed, or self-expression for each individual, for example.
> Impact: The results of one’s work, the subjective judgment that your work is making a difference, is important for teams. Seeing that one’s work is contributing to the organization’s goals can help reveal impact.
The researchers also discovered which variables were not significantly connected with team effectiveness at Google:
Colocation of teammates (sitting together in the same office)
Consensus-driven decision making
Extroversion of team members
Individual performance of team members
Workload size
Seniority
Team size
Tenure
https://rework.withgoogle.com/en/guides/understanding-team-effectiveness#introduction
> found that what really mattered was less about who is on the team
Because they already work at google and have gone through a pretty significant selection process
How do you separate "how the team works together" from "who is on the team"? I'm pretty sure the most significant factor by far in how the team works together, is what sort of people the team is composed of.
Notwithstanding far too many Hollywood movies, you can't hire a bunch of random idiots and assholes and expect them to turn into a well-oiled machine after an extended training montage. Not even if you've read the Bignum Habits of Highly Effective Managers. As you note, safety is important. The assholes will make everyone else feel unsafe. Dependability is important. The idiots won't be dependable (except in providing idiocy).
"The researchers found that what really mattered was less about who is on the team, and more about how the team worked together. "
We've known this, in a broader sense, since the 1960's. Look up research relevant to "leadership style", "Group Roles", and "Group Cohesion" on google scholar.
Re Psychological Safety:
It is not obvious if this generalises. A proverbial google Engineer "asking a question, or offering a new idea" is surely a valuable asset. An idiot doing the same is holding back the entire team. Of course, if true, the morale of the story would be not to hire idiots, and easy to say if you are google. In practice there is a huge middle ground between google engineer and idiot. Advice that works at google might not actually work on that middle ground.
Re Dependability: Successful teams "reliably complete quality work on time" seems to be a massive tautology. The question to be answered is how to get a team to do that.
The whole point of the research is that, in a properly structured team, there are no idiots. Probably there is a level of lack of competence that cannot be integrated into a team, but people at that level are unlikely to be hired in the first place.
"The question to be answered is how to get a team to do that." As I pointed out above, we've know how to do that for fifty years. The problem isn't technical, it's organizational politics (effective front line teams threaten management).
A stupid question is easily dismissed. Someone who is too afraid to ask a stupid question might not realize it's not actually a stupid question. And when idiots are free to ask stupid questions, more people are more likely to realize the person is an idiot and if they're a problem can more easily identify the problem.
If I had to have an idiot on my team I'd rather have one who asks stupid questions, rather than one who repeatedly gets stuck or endangers everything by being too afraid to ask stupid questions.
I believe the concept of some things being "above someone paygrade" was invented to fence which questions are useful and which questions are an attempt to understand what the questioner neither can nor needs to understand.
To repeat, I am aware that this does not apply to high end engineering teams, but anything flirting with "there are no stupid questions" seems to betray ignorance of a relevant part of reality.
One person's stupid questions, may be intelligent questions to another.
I don't know how high-end the high-end engineering team I used to work with was, but we were all getting a training in technology X (I forget what it was), and the instructor kept throwing out acronyms I was unfamiliar with. I asked him to unpack a couple, and he did. But he kept throwing out the damn acronyms. Finally, I asked him if he could put together a glossary of terms with his slides. He said, "Well if you don't have the basic knowledge of this technology, you shouldn't be in this course." He never provided the glossary. But other people in the training asked me to forward the glossary (that he never provided) to them, too. So, I assumed there were a lot of people attending the training who weren't asking questions but were too afraid to look stupid by asking questions.
This is interesting, but I’ve always found specific personnel play a huge role in team effectiveness. If you have a serial exploiter of team systems (someone who actively seeks loopholes, takes advantage of technicalities or weaknesses in protocols while ignoring their intent, etc.) then all the structure and clarity in the world can’t save the team.
And the study doesn’t even mention waffle parties or melon bars.
Thanks for this. Extremely relevant to my job.
I started a blog about the experience of being in-the-closet in the ultra orthodox Jewish community. Would love for you to check it out at https://open.substack.com/pub/exit98
Too much jargon for a gentile like me.
I started reading. Very well written and sincere.
You're not alone.
I'm a musmach of Rav Noach Weinberg (not Aish, I refused theur smicha - Rav Noach) and I can tell you that the God you don't believe in, I do not believe in either.
https://youtu.be/lMproqk7x6g
See also: https://youtu.be/7jesi_6VDOc?feature=shared
Greetings everyone. I think, like many, that the move in governance, but also business, education, administration, even the household, toward more top-down rules and mandates and away from the discretionary decision-making of individuals is not only making institutions less efficient and effective, but also making us all dumber, less moral, and more deceptive. And I'm not even a libertarian, not by a long shot. Norton signed me up to write a short book about this, part journalism, part empirical data from criminal justice, and part philosophical argument from first principles. The books is now available for preorder. https://wwnorton.com/books/9781324051251
Are you familiar with the research on group dynamics and leadership style?
Why are drugs in the USA so expensive? For example, while high costs are often attributed to research and development (R&D), why are older drugs like insulin, which should not require much R&D, still so expensive?
Drugs which are truly old, and in high demand, are generally quite inexpensive in the United States if you don't insist on the fancy brand name on the label. Drugs that seem like they should be old and cheap but aren't, I think fall into three categories:
1. Branded drugs where doctors have gotten in the habit of writing the brand name on the prescription so that the patient is locked in to a monopoly supplier. You can *usually* get the doctor to rewrite the prescription for the generic chemical name if you ask, but sometimes they've bought into the idea that Brand Name is Ineffably Better and also Price is No Object. "Epi-Pens" vs. adrenaline autoinjectors are a classic example here, and Scott had a post on the subject back in the SSC days.
2. Drugs that are variants of classic old drugs, with some tweaks to make them at least slightly better. Your example of insulin fits here. Just plain insulin, like your grandparents' generation used to spend a lifetime Not Dying from diabetes while complaining about all the annoyances, is still I think about $20-30 per month in the United States. But "insulin" now usually points to a tweaked variation on natural insulin that is much easier to properly use in the field. In the case of insulin, this is more than "slightly better". But it does come with substantial research and development and especially testing costs, which someone has to pay for (and governments mostly won't).
3. Old drugs for which there is very little demand. The problem here is that, no matter how old and off-patent the *drug* is, you aren't allowed to sell it until you've spent many tens of millions of dollars convincing the FDA that your specific factory can safely make that specific drug. Yes, even though you're using the old known-safe recipe and your factory has safely made lots of other drugs. The classic example here is Daraprim, which earned Martin Shkreli a well-deserved fifteen minutes of infamy a few years ago. By buying the one factory that made Daraprim, Shkreli was able to gain a monopoly on a low-volume market but one where he could jack up the price to absurd levels. At which point, anyone else on the outside looking in would only see the "opportunity" to spend tens of millions of dollars for the privilege of getting into a price war that would see both sides selling at low prices to a low-volume market.
There are potential solutions to all of these, but it will be a different solution for each case.
I've had a few cases of our insurance refusing to cover a prescription fill because something went wrong between the insurance and either the doctor or the pharmacy (e.g. the doctor wrote a 3-month prescription, and the insurance will only cover 1 or 2 months at a time). From these, I learned that old, very basic maintenance drugs tend to be dirt cheap, in single $ digits per month (I guess we've been lucky, as I haven't seen an exception to this heuristic yet). It's when you need a newish drug that you have a problem.
Not that simple. MAOI's for instance, are very old drugs. They were used in the 1950's as antidepressants. They are currently out of fashion, but the people I respect most in the field believe they are the most effective drugs for depression. The average cost for a month of a typical dose of the generic tranylcypromine is about $240. Could be worse, but that's pretty bad. I believe the reason for the high price is that it is used by many fewer people than, say, prozac.
Aren't they also more dangerous than SSRIs?
The main danger with MAOI’s is that they interfere with a person’s ability to metabolize tyramine, which is a harmless amine for most people. However, if a lot of it builds up in the body it raises blood pressure. While the body is built to do fine with temporary blood pressure raises (bp goes up quite a lot during vigorous exercise, for example, and also when someone is upset), there’s a limit, and a really big excess of tyramine can raise bp so much that it’s truly dangerous. Some foods have high-ish tyramine levels, esp. highly aged cheeses. HOWEVER: In the 50’s and 60’s, studies of the amount of tyramine in various foods found enough tyramine to be a problem for people on an MAOI in quite a few foods. Redoing these measures in the last 20 years found much much less tyramine in foods than the early studies, partly because measurement methods are better now, but mostly because refrigeration is much better. People who take MAOI’s now only need to keep in mind a few restrctions, and even those they can violate in moderation and be pretty safe. A lot of the medical profession hasn’t caught up on MAOI’s, though, and don’t know this stuff. If you are curious about MAOI’s, have a look at Psychotropical, the web site of a world expert named Gillman. And other antidepressants aren’t all that awesomely safe either. There are things SSRI’s should not be mixed with. And tricyclics make doctors nervous because the lethal dose as not all that much higher than the effective dose.
Thank you for providing the counter-example. I figured they existed - I'd just never seen one.
For most goods that are cheaper abroad, market forces will make it so the local price is approximately what it costs to buy it where it is cheapest and ship it to you. This is illegal with drugs.
Because They Can.
I've lived in many countries and the same meds that cost a fortune in the US cost pennies abroad.
Skip all the explanations about, "America subsidizing the rest of the world" and how "there must be a major profit payoff to pay for R&D", etc.
Those are all obfuscatory nonsense responses of no value. The simple answer is BECAUSE THEY CAN.
Oh, and also BECAUSE THEY WANT TO.
If medicine was affordable people might not be obsequious to their bosses and might pursue their passions rather than careers, but when you tie The Ability To Stay Alive to your job, welo you sure as hell better keep your job!
Oh, or join the military -- for "the benefits".
"Because They Can and BECAUSE THEY WANT TO" is also a response of approximately zero value.
Like, yes, obviously, given the choice between "making less money" and "making more money" approximately everyone will choose the first one. But *why* can they make that choice here and not elsewhere is the interesting question that just saying "it's greed, lol" doesn't answer. Have they not invented greed in Europe yet?
Greed was introduced to Europe in the 1500s, but it didn't take well and soon died out, as it was considered "gauche". Maybe it can make a comeback with new marketing, including a rebranding.
Who is "they" here?
The profits from current drugs support the R&D costs of the new ones. The funnel is very leaky: for each marketed drug dozens more failed in an earlier phase.
The FDA's process increases the costs of these phases such that drugs which would only be marginally profitable conditional on success aren't pursued.
And other countries threaten non-enforcement of pharmaceutical patents unless they are charged essentially only variable costs, leaving the US to bear the vast majority of the aforementioned fixed costs (much of which is FDA-inflicted, so it's not entirely unfair).
To get drug prices down without choking off future life-saving innovations, the point of maximum leverage for Americans is reducing the expected cost to get a drug through the process.
People will often flawly justify the insane cost of drugs:
1) R&D costs are off topic for most drugs on the market.
2) R&D costs don't even need to be high, it's mostly a symptom of the insane mediocrity and dysfunctionalism of pharmaceutical companies
3) R&D budgets are significantly smaller than the insane marketing/propagand/corruption budget of pharmaceutical companies
About the actual question: The production costs:
1) There is little to no price negociation with the FDA. Price is artificially set as high as possible to maximize profit. There is often no competition for a long time because of bureaucracy and dysfunctionalism. Even when competition arise, there is probably a high level of cartelization/price collusion between competitors to only marginally reduce prices.
2) Making drugs is generally extremely cheap at scale
3) the prices of drugs, independently of their chemical type, is hyperinflated in diseases where you can milk more the patients which is the incarnation of evil.
https://pmc.ncbi.nlm.nih.gov/articles/PMC10371812/#:~:text=Our%20regression%20estimates%20indicated%20that,drugs%20were%20on%20average%20USD
4) there are broadly two main drug classes where productions costs are orders of magnitude higher than typical drugs: biosimilars/polypeptides and gene therapy.
Of them, biosimilar costs are proven to be fully fraudulent, producing polypeptides is relatively cheap, generally a monthly dose production cost is in the literature estimated at 50 dollars. Which is higher than non peptide drugs but is orders of magnitudes cheaper than market prices.
Moreover this cost can even be reduced via production techniques refinements, as outlined in the literature.
The machines to synthetize peptides are publicly available and cost less than 20000 dollars meaning that it cost less to produce your own insulin and glp1 agonists and to give them for free to people than to buy your yearly individual dose, without any SOTA refinement or economy of scale...
Such level of dysfunctionalism is peak insanity and utilitaristic crime.
The reason there aren't ONGs or biohackers selling such peptides at 100 times less costs is because 1) people don't know they can produce their own biosimilars and 2) it is probably illegal, even for unpatented generics (about endogenous molecules..).
See for example openinsulin, it was funded in 2015 and I don't think (?) they are selling insulin, they are probably standing idle doing nothing because of criminal regulation.
https://en.wikipedia.org/wiki/Open_Insulin_Project
> The average price per vial was estimated to be $7 with each patient needing two vials per month.
This is much more revolting than all wars in the world combined.
The second class of drug: gene therapy
It seems, but I am unsure, to be truly costly however why is unclear and a moral imperative to clarify.
For example, if the cost are driven by viral vectors (why?) then it could be replaced by ultra cheap deliveries like liposomes.
The FDA doesn't buy large quantities of drugs.
FDA regulatory approvals can be altered by economics it is no wonder that gene therapies target rare diseases.
I mostly meant government led pricing caps e.g. by medicare
I'm no expert, but have heard and read several explanations that probably all play a part. Some of the more plausible ones: There's little to no price negotiation, regulation, or price pressure at the national level, like there is in single-payer systems. Medicare is basically prohibited from negotiating prices of medication or preferring inexpensive alternatives to expensive drugs. And the negotiations that do happen between insurers/payers and manufacturers tend to keep list prices high, in favor of steep discounts, post-sale rebates, etc.
Also, there's a lack of transparency in the current system. I think if drug co-pay was fixed at a percentage of list price, regardless of insurance provider & plan, patients would have more of an incentive to ask their doctor for the least expensive effective alternative, that would put more downward pressure on list prices.
Regular insulin has been around for a long time, but the drug companies have come out with insulin analogs. The insulin analogs are either fast acting or long lasting. The drug companies can keep insulin under patent by continuously coming out with improved versions. As DJ says you can get regular insulin for $25 at Walmart, it is the new analogs that are expensive.
Making insulin with long half lives is trivial and cheap, you can e.g. pegylate it. Let's not mystify basic old tech.
"Making insulin with long half lives is trivial and cheap, you can e.g. pegylate it. Let's not mystify basic old tech."
Is getting the insulin with long half lives through clinical trials trivial and cheap?
Some of the costs (not all, but some ...) is because clinical trials can be expensive. The old, regular insulin is cheap to manufacture, off-patent and has gone through clinical trials. If the new stuff is missing any of these three then it can be expensive.
It's worth pointing out that once those bioreactors and purification systems are set up, the cost of *producing* the insulin drops below $5/vial. The current Regular insulin (which you can buy for <$35/vial) is manufactured in bioreactors — just like the expensive stuff that costs >$250/vial. And I can buy the same expensive insulins in Canada for about $45/vial (but I need an Rx from a Canadian MD).
And don't forget that Eli Lilly dropped the price of its insulins to $35 without going broke.
I can't speak to other drugs, but as a diabetic I recently learned I can buy insulin over the counter at Walmart for $25. There are more advanced time release versions that require a prescription but the OTC insulin is fine for my needs.
How long does that dose last you?
About eight weeks. I take ~20 units with every meal.
Ban s on negotiating down prices.
Does anybody here know of any actual *studies* supporting the idea that getting orthodontic work as a child, has positive outcomes for long-term health? We are going through the interview-orthodontists process with our kids, and... there are all these claims that cranking open their upper palates will broaden their airways, reduce headache, etc. but... I've also heard some fairly scary things about American-style orthodontics: that aggressive teeth-moving compromises long-term blood supply to the teeth and then you end up getting root canals and losing teeth in your 50s, that it fails to address the initiating problems of posture, bite, and incorrect forces applied, and so you can end up needing retainers forever, because those forces are still moving teeth out of place, even after all the work.
What's really frustrating is that there do seem to be other options out there, with at least case studies to support... but only if you live outside the US. I keep asking here about things like Planas Direct Tracks, bite blocks, Advanced Lightwire... and they look at me like I just sprouted three heads. And then they go back to their sales patter. I can't get them to explain anything, or go off-script and it's deeply frustrating, when they're asking me to spend $10k+ and 2-3 years doing painful things to my children.
And then they go and tell me "it's genetic" and I die a little inside. Most people did not have crowded teeth three generations ago, but it's rampant now, across the population. THAT'S NOT HOW GENETICS WORKS. We didn't all mutate in that time, and being told that *does not inspire confidence*.
I'm not into cosmetic medical procedures just for the sake of giving my kids a middle-class smile. I know that's un-American. Sue me.
Like, if we're going to go through with this, I NEED to have some kind of evidence that there's a real medical benefit to it. Does that evidence exist? Or is this a whole industry just for cosmetically modifying your kids?
Much of orthodontia is in a grey area. See for example: https://www.romper.com/parenting/palate-expanders-orthodontics-scam
The pop-sci book Breath by James Nestor asserts that many orthodontic ailments are absent in our ancestors because of our modern diets. He suggests that our ancestors chewed significantly more because they ate more hard, tough foods, and that many orthodontic ailments are caused by weaker jaw muscles. The effect, he argues, is exacerbated by mouth-breathing, which can case nasal passages, sinuses, and other air pathways that alleviate e.g. apnea, to shrink.
Apply grains of salt, but I would err on the side of heeding orthodontists' specific claims and rejecting their non-specific claims. Further, Nestor's evidence suggests that there are behavioral alternatives to expensive surgery.
Well, yeah. I've seen the archaeology data. IIRC, medieval people had dental arches ~1cm wider than ours, and neolithic man maybe as much as ~3cm wider. That's not genetic drift. These are genetically modern humans we're talking about. The difference is clearly environmental. But, you know, my kids are 10 and 12-- we can't wind the clock back and raise them on a diet of wild tubers and dried wooly rhinoceros jerky.
That said, has there been ANY research at all, into whether modern orthodontic practices are helpful, harmful or anything in between?
Widening the upper palate is considered beneficial if you haven't taught your children to eat tough food like carrots, whole apples and crusty, chewy bread. It's better to do this before permanent teeth start to erupt.
Yes, but is there any *evidence* for this?
I am a late 20s man who still has his wisdom teeth. The few times I have moved cities I have had a tremendously difficult time finding a dentist. When I go in for regular cleanings, the first thing the actual dentist tells me post cleaning is that I should get my wisdom teeth out. When I ask why, the answer is typically "prevents long term health issues", and if I asked what specific health issues, either the question is either deflected or they say "food can get stuck, and that can lead to infection". The only food that gets stuck in my back teeth on a semi-regularly basis is pretzels, and so as what I deem to be a rational response, I eat fewer pretzels.
I had one particularly bad experience with a dentist on the north side of Chicago. The dental assistant told me they were using a new cleaning technique, which I think involved some type of UV light, but was noticeably painful when used on my bottom teeth. The dentist herself was a cosmetic dentist, and told me on my first visit that I was showing signs on gum loss on my lower teeth. I thought this was strange, as I had never heard it before from any prior dentist. Six months later, I go for a routine cleaning, and I get the same dental assistant, who uses the same UV cleaning process. When the dentist saw me, my "gum loss" was confirmed, and she referred me to an oral surgeon to fix the problem. I was naturally skeptical that I needed this multiple thousands of dollars procedure, even if my insurance would pay for it (it just sounded really inconvenient to go through all that). I checked the instagram page of the dentist, and there were multiple pictures of her and the oral surgeon to whom she referred me at networking events, sharing drinks, and partying. I was somewhat familiar with the healthcare referral process and how some practitioners can earn a sizable chunk of their income from them, and it all felt pretty icky to me. So I scheduled a cleaning with another dentist off insurance for like $100. I didn't mention my purported "gum issue", and the dentist did not make any mention of it. Needless to say, I stopped going to the cosmetic dentist, and stayed with the new one for the rest of my time in Chicago. I do not trust dentists.
I didn't want to have my wisdom teeth pulled out. Issues with overcrowding started showing up in my 30s at which point I had to have them removed. It was a nightmare. Should have done it back at 20. Sometimes waiting for problems to show up is not the best idea.
Had one removed at 35, no issue whatsoever. Was a bit sick though for a day or two, but everything recovered perfectly without issue. Pulling it took like a minute tops.
I am almost 50 and I never had any problem with my wisdom teeth. So, both things can happen, the question is how frequently.
(Trying to ask the experts doesn't help much. In my experience, different experts give contradictory advice, each of them with an expression of absolute certainty.)
Decision making under uncertainty at its finest. The best time to take them out is when the least information about possible issues is available.
The US is very pro pulling wisdom teeth early. Other countries aren't. There's the cynical argument about how it is monetized which I have sympathy for. There is also the argument that people including medical professionals herd. Either way - push back. Wait for an issue to arise - it might never. Downside is having them removed when you are older is it is probably going to take a few more days to recover. Worth it.
My understanding is that a downside is that wisdom teeth, being harder to clean properly (especially if they're partially erupted) get cavities and decay, and then spread that decay to neighboring teeth. More frequent visits to the dentist would have caught it, but at any rate, I got two wisdom teeth yanked in my mid-20s.
But I agree that there seems to be a cultural difference: I have one wisdom tooth that's fine (not perfectly erupted, but no cavities), and my US dentists advise me to get oral surgery to get it removed, and my India dentist tells me to leave it be until there's a problem.
> Ten million third molars (wisdom teeth) are extracted from approximately 5 million people in the United States each year at an annual cost of over $3 billion. In addition, more than 11 million patient days of “standard discomfort or disability”—pain, swelling, bruising, and malaise—result postoperatively, and more than 11000 people suffer permanent paresthesia—numbness of the lip, tongue, and cheek—as a consequence of nerve injury during the surgery. At least two thirds of these extractions, associated costs, and injuries are unnecessary, constituting a silent epidemic of iatrogenic injury that afflicts tens of thousands of people with lifelong discomfort and disability.
— "The Prophylactic Extraction of Third Molars: A Public Health Hazard"
Jay W. Friedman DDS, MPH...
https://ajph.aphapublications.org/doi/full/10.2105/AJPH.2006.100271
And...
> The British National Institute for Clinical Excellence is unequivocal in its recommendation, adopted by the National Health Service: “The practice of prophylactic removal of pathology-free impacted third molars should be discontinued. . . . There is no reliable evidence to support a health benefit to patients from the prophylactic removal of pathology-free impacted teeth.”9(p1–2) The conditions for which extraction is justified include nonrestorable dental caries, pulpal infection, cellulitis, recurrent pericoronitis, abscesses, cysts, and fractures.
And there's a risk of delayed-onset infections after the extractions. The study below suggests the use of clotting agents (hemostatic use), high blood pressure, or certain positionings of the tooth before extraction caused a higher incidence of delayed post-operative infections.
"Risk Factors for Delayed-Onset Infection after Mandibular Wisdom Tooth Extractions" by Ryo Miyazaki et al...
https://www.mdpi.com/2227-9032/11/6/871
Haven't been able to find any mortality numbers, but the data indicates that the majority of extractions are unnecessary and can cause some serious downstream issues.
Can't find the study now, but it found that more people die from complications due to wisdom tooth extraction than from leaving the wisdom teeth in.
As someone who knows nothing about this, I would err on the side of non-intervention until you have a better reason to take action. Just make sure they breathe through their nose and eat food that exercises the jaw.
This has been my approach both personally and as a parent. I share Yarrow's skepticism about this topic. Not a fixed opinion to be clear, I live by the maxim that "skepticism is a process not a position". But if actual longitudinal evidence along the lines asked for exists I've never yet seen it.
Broadly in my immediate family we're on the lucky side as far as dental genetics; my wife not so much. Our son is now 12 and there has been talk of braces but nothing any more serious than that has been suggested. I've now reached my 60s, still have my wisdom teeth and they're perfectly stable and dentists have quit asking me about it; my mother died at 91 with a full set of healthy wisdoms still in place. Whereas my wife had to have hers removed years ago as did my adult son (son of my first wife). My elder brother now has serious dental issues but that is pretty obviously because he's been dumb about it and hasn't followed even basic dentist advice in decades.
So in summary we'll do what our son requires when or if he requires it; I will resist firmly any significant interventions which are not backed up by some real science. (And am irritated by the claim that such interventions represent the "better safe than sorry" approach -- they do not!)
By the way since we're here I will share a recent eye-roll-inspiring development on a tangent but in a way it applies. We have a family dog, a healthy bouncy 50-pound mutt who we all love dearly. My wife and I are both experienced dog owners, in my case since childhood. Our doggo gets solid regular vet care, all shots etc. He's now about 5 years old.
Earlier this year the vet said his teeth needed cleaning. His teeth are structurally fine; his breath at any given moment is not materially better nor worse than family dogs' breath tends to be; he has no actual cavities or other dental issues. Just the discoloration (plaque) that dogs get because, well, they eat like dogs. He gets the "dental stick" type treats which are supposedly helping clean his teeth; they may honestly just be treats but, whatever.
Newly, the vet said that manually cleaning a squirming dogs' teeth is not very effective so they should once a year knock him out to be able to do it properly. And this service costs several hundred dollars.
My wife tends to assume that doctors' opinions are the final word and she was doing the visit so said okay. They knocked him out for half a day, cleaned his teeth which then looked very nice. The dog woke up from the anesthesia moaning in confusion and fear and wasn't himself again until the next day.
I subsequently learned from others at the local dog park that all the vets are declaring this to be a new normal in canine care, once a year. Well it isn't in our house: I put my foot down and said, no. We will not be doing that again to any dog of mine.
Effective generally-safe anesthesia is one of the most-beneficial inventions in medical history, arguably in scientific history generally. It is also not zero-risk. There is a reason that doctors for humans don't do _general_ anesthesia until it's actually needed for the operation.
I treasure our pup, which does not make me think that his teeth need to be pristine or that we can't allow any risk of some normal dental slippage as he ages. Knocking him out entirely for the sake of simple cleaning is in my view on the edge of outrageous as a risk-reward medical decision. I managed not to say FFS out loud but my wife knows I was thinking it.
I agreed that we (which means me, assisted by our son who tries to hold the squirming pooch) would manually do some canine tooth-brushing once a week. Which is fine, kind of amusing actually, and we do it. My expectation is that it's just slowing down the rate of plaque buildup, fine.
But actually knocking him out just to clean his perfectly-healthy teeth...?? I'm still irritated by the whole thing. Including also the several hundred dollars part of course.
I should add, for completeness, that the kids do, in fact, have really crooked/crowded teeth, narrow dental arches, and the older one a pretty marked overbite.
It's just... I thought there were ways (such as PDTs, bite blocks, and even night-guard appliances) to correct bite by adding a little height at the molars, that didn't involve literally three years of metal-mouth and yoinking teeth around. I don't actually care if their teeth aren't perfectly straight, if that means those teeth are still healthy into old age. They've never had any cavities-- I certainly don't want to change that. In my own family, there's a huge correlation between cavities and orthodontic work. My mom and sister both had extensive orthodontics, and also suffered a lot of cavities-- seemed like they were always getting fillings replaced. Me and my other sister both have overbites but escaped orthodontics. Between us, we have zero cavities.
What nobody has been able to explain in any kind of satisfactory way is: with an overbite, if you move your jaw so that your front teeth line up correctly, there's this big space between the molars: they don't touch anymore. So, with all these appliances and crap to "pull the jaw forward" (or the one I hate more: "*grow* the jaw forward" wtf kind of magic car-salesman talk is that?), how are you going to fix the problem where the molars won't touch? Because I'm not letting you file off his front teeth, so you'd better have some other strategy for that.
For what it's worth, I'm someone who has had a huge number of cavities although I eat pretty well and brush/floss obsessively. And on the other hand, my teeth are quite well aligned and I never needed any orthodontics. The same is true of at least one of my other family members. This is no more or less anecdotal than your own reports about your family, but, in lieu of more robust data, I nevertheless suspect that the correlation between orthodontics and cavities you have observed is spurious.
I remember coming upon Scott praising the prose of G. K. Chesterton and filing away that fact.
Just recently I came across a collection of G. K. C. essays and found things like this there:
"Some little time ago I stood among immemorial English trees that seemed to take hold upon the stars like a brood of Ygdrasils."
"In certain endless uplands, uplands like great flats gone dizzy, slopes that seem to contradict the idea that there is even such a thing as a level, and make us all realise that we live on a planet with a sloping roof, you will come from time to time upon whole valleys filled with loose rocks and boulders, so big as to be like mountains broken loose."
Wow. I don't think any imagery hit me like this since first reading E. B. White. And I think I detect a hint of similarity in eg. the recent post, "The Innocent and the Beautiful Have No Enemy But Time."
I'm surprised because the styling of G. K. C. makes the writing feel fresh and strong despite it being over a century old. Maybe because of the contrast with most of what I read on substack?
Why don't more writers use bold, original imagery? Is it just fashion? Is it too difficult? Are there political connotations?
Try it yourself, and I think you will discover the reason why.
Part of it is fashion--Chesterton is using a deliberately archaic style even for his time, although at his time it was a bit more common to use "old-fashioned" devices. Even writers with polished prose styles and a penchant for elaborate images aren't likely to use that kind of language. Although there are exceptions, especially in fiction--Cormac McCarthy and W.G. Sebald are a few relatively contemporary examples.
But in general SSC and rationalist-adjacent blogs don't go for elaborate stylistic stuff. You'd be better off looking to a literary magazine for that kind of thing--the London Review of Books, n+1 (sometimes), The Paris Review etc. Also, writing like this is pretty laborious, so people who do it aren't as likely to do it on a blog, which feels a bit disposable.
I'll go check out McCarthy and Sebald, thanks!
If you don't mind me asking, if Chesterton is using an archaic style for his time, what was the mainstream style of his time?
I love this imagery, because if you've ever seen bars of lead broken open, it hits you "Yes! That's what it's like!" There's something almost surrealist about day breaking like bars of lead, and that's when I remember he went to the Slade School of Art, so he has a very visual imagination:
"The clear moon that had lit up Chiswick had gone down by the time that they passed Battersea, and when they came under the enormous bulk of Westminster day had already begun to break. It broke like the splitting of great bars of lead, showing bars of silver; and these had brightened like white fire when the tug, changing its onward course, turned inward to a large landing stage rather beyond Charing Cross."
I’m an English Catholic (well by birth) and I should like Chesterton but I don’t. His writing is very dated, like an old form A level essay.
"I have long ceased to argue with people who prefer Thursday to Wednesday because it is Thursday." --GKC
I'm sorry you feel that way 😀
Nobody is obliged to like everything, but as an Irish teenager he was the first English writer I read who didn't engage in "and of course those Irish, well what can you do, they brought it all on themselves" type of shoulder-shrugging.
Then I found out he was a Catholic convert and went "Ah, that explains it!"
"Why don't more writers use bold, original imagery? Is it just fashion? Is it too difficult?"
It is difficult to use original imagery. Partially because good imagery often gets copied and then quickly becomes cliche.
George Orwell has an essay on writing that has a bit on this. But I don't remember the essay :-(
Yeah, why use original imagery when cliche is so handy?
But perhaps cliche is also handy because you're 100.09% right in that it will be understood by more people? ...But then, is that the goal?
I think the essay you're referring to is "Politics and the English Language"
https://www.orwellfoundation.com/the-orwell-foundation/orwell/essays-and-other-works/politics-and-the-english-language/
Thank you! Yes, that is/was the essay.
In nominative determinism: today I learned that the president of the Royal Horticultural Society is Keith Weed, CBE.
I wonder; does anybody around these parts maintain a list of nominative determinsms?
When the British Secret Service discovered that human semen could be used as an effective invisible ink, it was headed by Sir Mansfield Cumming.
https://en.wikipedia.org/wiki/Mansfield_Smith-Cumming#World_War_I
(sorry)
No but everyone needs to know of Robin Mahfood, former CEO of Food For The Poor.
There might be a reader here whose real name is List (a fairly common Germanic surname) who could be persuaded to take on the job.
I recall the days when Bob Flowerdew, Pippa Greenwood, and Clay Jones graced Gardeners Question Time
I have no list, but Charles Ragon de Bange was a French artillery officer who designed a bunch of french cannons.
Manfred Freiherr von Killinger was a Nazi (mentioned on Slate Star Codex, I believe).
Does it count as nominative determinism when the names in question are only apt in a different language from that spoken by their holders? "Killinger" probably doesn't imply anything in German, for instance.
For maximum amusement, I think we should take the broadest possible view of nominative determinism.
Oh man, and here I thought that "Dr. Henry Killinger" was uniquely brilliant Venture Brothers joke. I had no idea that reality made the joke first.
Compounding the joke, the real Killinger was the German liason officer tasked with asking the Romanians to adopt a more disciplined implementation of the Holocaust because their more openly enthusiastic version was making them all look bad, As Scott put it in the SSC piece, if a literal Nazi named "Baron von Killinger" is telling you that you need to tone down the war crimes a bit, you may indeed be the baddies.
(Just to be clear, Killinger was not one of the good guys in this story, he was just a more pragmatic and prudent sort of baddie)
Managed to not even hear about brain zappers until recently. Anyone here who tried it? How was the experience?
Related question: has the potential of ECT been under-researched due to stigma, e.g., can it actually induce wellness despite no clinical depressive/manic diagnosis?
I’m not an expert, but have observed ECT and received TMS. There is indeed a lot of stigma around ECT in particular, which may partially explain a general sense from psychiatrists that it is woefully underutilized. However, it also requires anesthetic sedation and can cause memory loss, so you’d have to have good reasoning for why it would work on mild cases in order to ethically subject research subjects to it.
I think TMS is underutilized because the machine is very expensive and it requires very frequent visits, so it’s hard to get it covered by insurance. I don’t think it helped me but could be wrong
Thank you. Much appreciated.
Would it be efficient to build multiples of expensive telescopes in sequence?
I recently attended a fascinating talk on the possibility of life on hydrogen ocean worlds (https://www.alumni.cam.ac.uk/festival/events/the-hycean-paradigm-in-the-search-for-life-elsewhere) and it quickly became clear that a major barrier to his and other astronomer's work was getting time on the James Webb space telescope, which is consistently over ten times oversubscribed.
The overall cost of the JWST was about $10bn, but a huge chunk of this (a majority?) must be the R&D, testing, building the tools, etc. Now that we know it works, couldn't we double or triple the scientific output we get from the JWST by building a few more at a fraction of the initial cost?
Others have pointed out that the cost savings aren't as great as you might expect; as a card-carrying rocket scientist, I concur.
But there's also the question of what you are really going to accomplish. There's no near-term practical value to life on hydrogen ocean worlds orbiting other stars; we aren't going to be able to make use of it to make our own lives better. Except that we want to know, and we would enjoy the discovery, and those are good enough reasons to devote some level of resources to the endeavour.
One JWST, well maintained(*), will eventually image everything in the universe that can be imaged by JWST. There's only 4*pi steradians in the sky; divide by the JWST field of view and multiply by exposure time, and that's it. The life on hydrogen ocean worlds, or whatever, we'll find it. Well, maybe not us, but our descendants will.
What have you got against our descendants, that you want to hog all the glory of discovery for our generation and leave none for theirs? Hedonic adaptation suggests that we'll get less out of each new JWST-scale discovery, but each new generation will get the full first-time experience. Maximum utils at minimum cost = build one JWST and build it to last. Same for any similar instrument, including the hypothetical even better successor to JWST.
Where the spam-the-sky-with-telescopes strategy would be of value, is in observation of unpredictable ephemeral events. It would be nice if we could have a dedicated space telescope looking at each potential supernova candidate within say a thousand years. And we could use a few more telescopes looking for asteroids and (especially) comets on Earth-impacting trajectories.
* JWST is not actually well designed to *be* maintained, but that's a different issue and we're working on it.
> we're working on it
Please elaborate. Are you expecting Starship to reach L2 soon?
I'm not expecting JWST to need repairs soon. When it does reach that point, the best solution would probably be a robotic servicing mission, but if we need an astronaut to handle it (and we might), then a Starship or an SLS/Orion should be able to reach L2. And really, I expect we'll have something better than either by then.
The problem is, JWST wasn't designed to be serviced on orbit, so even if you've got astronauts to do the clever fiddly stuff, it may be that the parts which need to be replaced (probably the gyros) will be buried too deep in the system to be reached without disassembling the whole thing.
We've got a good many people looking into how to design a satellite or spacecraft so it can easily be serviced on orbit, and we've got some other people looking into how to do what we can for satellites that were not so designed. So, best case, when JWST needs repairs we'll be able to send a customized robot to do the job and it will just be more tediously annoying than it would have been if we'd designed JWST to play nicely with standard off-the-rack servicing robots.
Worst case, JWST itself will be a lost cause. In which case, don't replace it until you've updated the design to include all the fiducial markers and grappling points and quick-connect fittings, etc, and then replace it with one such vehicle.
OK, your comment implies we can reliably keep track of JWST after it drifts out of its halo orbit (and it won't accidentally fry itself looking at the sun), which was the part I was missing.
I had simply assumed it wouldn't last long enough to "cover the sky" with a reasonable amount of time, but if it's as you lay out I'm much less concerned then (footnote noted).
I am the human that has studied the most the ways to revolutionize astronomy metrology including the scalability of space mirrors, given it is a philosophical imperative.
The answer to why we have no large space telescopes and even less in large numbers is the extreme prevalence of mediocrity, self sabotage and institutional fraud.
I could go on about:
the prospects of alternative mirror materials (aluminium and other metals (there is no oxidation in space)
liquid mirrors (recently tested in the ISS) (microgravity naturally gives perfect liquid mirror smoothness)
alternatives to mirrors (lenses, etc)
foldable or stretchable mirrors
expandable mirrors
rectangular spinning mirrors (kinda crazy almost nobody on earth though of that) (no spin energy loss in space)
segmented mirrors
etc
Science progress is always about Hanlon's razor, and even If I am the human that most exhaustively know how to disrupt mirror scalability and economics through innovation, it is, sadly, as often, a distraction, a digression versus the actual bottlenecks that do not stems from technology but from the realization that the world is led by people that have borderline, scoped dementia.
Not realizing humans are universally, locally insane is part of the mental disease, a blind spot bias.
Not only is producing JWST clones cheap, but JWST is also peak overengineering about weak optimization for IR, that are neither potent, nor needed for UV and optical.
The use of materials (platine, etc) with no industry in this domain is pathological, doing a segmented mirror is actually trivial.
But it also comes from a false premise, not only are people wrong and distracted but they are chaining consecutive errors and false premises.
It is a criminal false premise that we even need segmented mirrors to reach JWST aperture size.
A monolithic space mirror is simply limited by the rocket max payload diameter.
JWST aperture is 6.5meters.
Starship payload diameter is 9meters which means it can make a monolithic space mirror larger than JWST.
Saturn V was 10 meters.
Even if we limit ourselves to Ariane V, it has a diameter of 5.4 meters, which is sufficently close to JWST to completely revolutionnize science (you have no idea of all the things this enable)
So if we already had rockets that enable monolithic mirrors to be JWST like, twice the size of hubble (and that would have higher resolution that JWST since being optical increase resolution ten times (the crime of JWST as not being UVOIR is another insanity), why
didn't we do it and why isn't it even planned in the decades to come?
Production costs of large monolithic mirrors are negligible.
The reason for the ESA is dementia level self sabotage:
I am among the extremely low number of people to actually know why Hershel and euclid have precisely their specific and low apertures.
It is because of contingent limitations in the crystallization of silicon carbide mirrors, herschel was the max size and it wasn't diffraction limited which means this low quality mirror was relegated to far infrared.
Euclid size was the limit diffraction limited max size for an optical telescope.
The limit was recently broke to 4 meters by chinese manufacturers, again a harmful distraction. The concept of silicon carbide mirrors and the cartelization of this industry ESA created itself, is one of the largest scientific crimes of the 21st century.
As to why NASA which enriched glass mirrors have no scalability limits (up to 11 meters) did not launch large monolithic mirrors cannot be explained by self sabotage of ad-hoc technological reasons, the answer must stems therefore from inept managerial reasons
1) NASA is regularly stopped by the defense industry and coerced to use repurposed military built old mirrors.
2) conflict of interest with the segmented mirrors lobby
3) probable inept induction from N=1 that since JWST blew up costs and delays that any large space telescope must be too budget and delay "risky" and therefore have in a self fulfiling prophecy way, delayed funding to post 2030s missions.
I can singularly and multiple times answer how to disrupt space mirrors and therefore science both on the mediocrity and the fraudulent aspects, but the dismissal of the existence of dementia level self sabotage and fraud by the "rationalists" is what allows the world to keep beeing so broken.
Whilst I definitely can't assess the science you're discussing, I am somewhat familiar with comparable *types* of projects, and find this entirely plausible.
"it quickly became clear that a major barrier to his and other astronomer's work was getting time on the James Webb space telescope"
That's not really true. Professor Madhusudhan was awarded 3 observations of K2-18b, but a competitor was awarded 7 observations: https://www.stsci.edu/~nnikolov/TrExoLiSTS/JWST/trexolists.html
JWST is still young. If the competitor confirms Madhusudhan's findings and the planet still seems interesting, astronomers can propose for more JWST time to look at the planet again.
The R&D is a one-time cost, but the economies of scale between producing 1 and producing a small N aren't great. The fabrication and validation still have to be performed by experts - imagine if each iPhone had to be hand-assembled by a team of PhDs. While building more copies is more cost-efficient to some extent, and linearly increases the sheer quantity of science that can be performed, this has to be traded off against performing an entirely different mission. If 5 JWSTs can be built for the cost of 1 JWST and 1 novel mission, is the incremental benefit of 4x JWST hours going to lower-priority astronomy projects (that weren't judged worthy enough to be allocated on the first/only telescope) more useful than what the 1 novel mission would provide? These are the sorts of things that the project administrators are considering.
We *have* made twins before (Pioneer 10/11, Voyager 1/2, Spirit/Opportunity), but I haven't seen cost breakdowns. It would be interesting to see historical accounts of why those projects were designed as twins versus alternatives. (Curiosity and Perseverance aren't twins and were separated by a fair amount of time and significant upgrades, but they do have a lot of commonality.) All of the examples I can think of (other than crewed missions like Apollo that ran increasingly powerful versions before being cancelled, which is a different kind of project structure) are exploration missions, where the benefit of being able to send the probe/rover to a different location means that you get to do different science there, and you can't change your mind after the trajectory is set. In contrast, telescope time can be allocated anywhere in the sky, after the telescope is launched.
Realistically though, JWST was so massively over-budget and over-schedule that it's a wonder it wasn't cancelled entirely. There's no way we could have gotten another copy, with how science is funded in this world. (And it's a wonder that it worked, speaking to the incredible skill of its team. Stuff unfolding in space is scary hard - look at Galileo's messed-up antenna.)
The considerations here, which I don't know how to estimate, are:
1) How good are the people in charge of allocating JWST hours at rank-ordering the most useful science to do with it?
2) How quickly does the value of the JWST projects drop, is it a sharp decrease (so basically all the value comes JWST #1), or does it drop off gradually, such that JWST #2 contributes 80% as much?
Contrary to popular belief, it’s not actually that hard to build a deep space telescope. The necessary precision of the reflectors for the same resolution decreases with the area.
The JWST is such an amazing device because of its constraints, and the ingenuity that it took to work around those constraints. If you had a bunch of guys with some steel sheets, welding equipment, some precision measuring, and a bit of polish, you could imaginable build a better telescope by hand.
That’s of course an oversimplification, but the idea holds, if you have a larger payload bay, and more mass to orbit, you can make the telescope bigger and therefore higher resolution (or bigger and less precise for the same resolution). If the cost per telescope is lowered, the pressure to get everything absolutely 1000% right goes way down, and thus the price of the whole thing goes down dramatically.
"The necessary precision of the reflectors for the same resolution decreases with the area."
That's not true. The precision of the reflectors has to be a fraction of a wavelength, and the wavelength of light doesn't change as the telescope gets bigger.
I don't know what your experience is with building telescopes in space, but I have actually helped build a few, and it is not quite as simple as you are making out. Telescopes like the James Webb needed new technologies to help shield them and keep them cold. You need precision to keep things aligned and together under really tough thermal environments, you have to make sure your materials and welds work as expected in a vacuum and in zero-g, and to work out ways to keep the telescope orientated properly in space to avoid overheating things, exposing them to micrometeorite damage or just to point at whatever you are looking at. It's really not something you could weld together in an afternoon.
That said, I think rather than mass production, some kind of standardisation could help with building cheaper telescopes in the future. People are still very much in the mindset that every observatory is a very expensive one-off, whereas we might be able to get more data by producing more but cheaper probes and telescopes. Even so, we'll always have the flagship observatories like the James Webb which are oversubscribed, simply because they can push the limits and that's where a lot of the most interesting stuff to be found is.
Thanks; I work in infrastructure finance, and safe to say I have seen the impacts of having a learning curve Vs not in the past decade or so on various technologies, so was wondering how applicable it would be for something like this.
Also, I love this comment section, this is the place I was most confident I would get a reply from someone who was actually involved in a space telescope!
You’re right, it was mostly an imagined experiment to illustrate the point that bigger = less precision necessary for the same resolution. If we can launch bigger, less precision is necessary, if less precision is necessary, there’s lower cost. If there’s lower cost, we can launch more or otherwise not spend the huge amounts of money to be as near to 100% sure the mission will succeed.
The scale where welding things together would actually work as an effective telescope (in combination with all the other things like temperature control and proper orientation) is probably humongous (it would take a lot of math to figure it out). If we could just snap our fingers and put whatever we wanted in deep space though, we could probably build a higher resolution JWST on the order of tens of millions, or hundreds of millions of dollars.
A lot of the complexity of building JWST had to do with folding it up so it could fit into an Ariane 5 payload bay. With the much larger and potentially much cheaper Starship it might make sense to design the next big space telescope for that.
I just wrote a reply to another comment rubbishing AI safety/Pause AI. Despite standing by my core point (it's not clear it's tractable and also problems that impact rich people are unlikely to be good EA candidates A Priori), I want to go against one argument I've seen, which essentially argues that international coordination is in principle impossible.
It isn't!
CFCs were globally banned and the ozone hole is retreating, chemical weapons have been held down pretty successfully (yes, they're less useful for maneuver warfare anyway, but international pressure has kept it down, and plenty of countries don't have a maneuver doctrine), we've collectively decided to not go there with human cloning, bioweapons have been suppressed very effectively...
I also think one can make the argument that at least fusion weapons were a (bad) choice and that we *could* have backed off from that cliff, but chose not to, with potentially world-ending consequences. We did successfully back from from having 70,000 nuclear warheads to c.12,000 now, which is still enough to wipe out every major city but maybe not enough to *definitely* end human life.
Advanced AI is, I will freely admit, particularly challenging here given the non-weapon applications and a more multipolar world, but if energy and compute requirements continue increasing then the ability to train a super intelligence looks more like a Manhattan project than a terrorist cell (if it becomes much cheaper in materiel and energy then that obviously becomes much more difficult to control).
Yeah, you should always be skeptical when someone personally interested in a niche issue (especially if it's in their personal financial interests) tries to explain to you that regulation of their niche issue is impossible. That's not to say that regulation is *never* impossible, but when "guy shitting in the deli meat because that's his fetish" gives you his 50-page treatise on how there's no point in a law that says he can't shit in the deli meat, because some Chinese guy in a Beijing processing plant can still do it so its meaningless to even try to start regulating, you should consider that you're dealing with a motivated source rather than an honest one, and "impossibility" may be less parts disinterested reflection on reality and more parts thin excuse because he'd personally rather not stop.
It is easy for pessimism to become a self-fulfilling prophecy: "X is impossible, therefore we shouldn't even try it." No one tries X. Unsurprisingly, X does not happen. "See, I told you X was impossible!"
Our ancestors could fly to the Moon, and eradicate smallpox. If we didn't know that it happened, it would be too easy to provide clever arguments why something like that is impossible and anyone who claims otherwise is just an uneducated fool who doesn't understand that the world is complicated.
This reminds me of the covid vaccine rollout. Without Israel as an example, who managed to more than double the rate of other western countries, nobody would have known it was possible to do that.
Covid made a lot of insanity visible. For me, a fascinating thing was during the first months (before people got tired) how everyone who remained in Slovakia posted on facebook that we need to be careful and protect the people around us, especially the older ones; and everyone who worked in UK was like "well, the virus is unstoppable, obviously the old and weak must die so that the young and strong can live free". And everyone was 100% sure that their way is the obvious one.
And, looking at the numbers, during the initial spring virtually no one died in Slovakia, I think we had by far the lowest numbers in EU. Then in autumn we fucked up completely, but that's another story. In a parallel universe where the vaccine was developer a few months earlier, it would be an incredible success.
When the vaccines came, again the cultural differences were great. Many countries made the vaccination mandatory, but with a lot of waiting, and often inconvenient. In contrast, in Austria it was voluntary, but the vaccines were easily available. Like, in some big parks in Vienna they built tents where they vaccinated anyone who walked around and wanted to give it a try; no need to register in some system and then wait until they call you. They didn't even care if you were a foreigner outside their healthcare system, so we went there to get our kids vaccinated a few months earlier.
So yeah, often things that are "impossible, and all reasonable people know it" in one country, are perfectly possible in another, sometimes the next door.
I'm not particularly familiar with Slovakia's example, but I'm skeptical of "but that's another story" w/re the autumn fuck-up. I think, broadly speaking, there are quite a few countries that achieve anomalously good results in the first 3-6 months of the pandemic, but by using methods that were not plausibly sustainable in the long run. And which thus mostly just postponed a bunch of deaths by 3-6 months.
If "and then we fucked up" refers to not sustaining the unsustainable, then no, that's not a different story, That's just the second half of the same story.
And yes, if we were going to get a highly effective vaccine available in quantity in 3-6 months, then aggressive temporary containment measures at the outset would have been a much better decision. But that didn't happen, and I don't think anyone ever seriously believed that it was going to happen.
Good point about sustainability. I still don't understand what people find so unbearable about wearing the masks, but... apparently they do.
I don't remember the exact dates, but RaDVaC and Winfried Stöcker's vaccine came relatively fast, they just seemed mostly ignored.
Something I'd be curious to know is what the odds of developing a working vaccine at all were. Because early it was commonly heard that it might take years to develop a vaccine if ever. But what did the actual experts think at the time? And what would have been a good estimate in hindsight?
If we need to develop a vaccine for the new bird flu, what are the odds we can do it and in what time frame?
Seems to me that *developing* a vaccine is relatively quick with the science we have today. The "source code" of covid virus was available practically overnight. Synthesizing parts of RNA is business as usual. See: https://radvac.org/
Now getting the vaccine *approved*, that is the real challenge that takes several years and costs millions (i.e. you can't do it on your own). A naive person might imagine that we could streamline the process during a pandemic, but that's not how these things work.
The only things that have been successfully banned are things that aren't very useful or which have easy substitutes.
I'm not sure that's true; human genetic engineering / cloning seems both useful and to not have obvious substitutes, but that has been pretty effectively banned.
Have you not heard of polygenic embryo screening? Human genetic engineering proper doesn't exist because it's not currently technologically feasible. If it was possible to guarantee an extra 15 IQ points for children, there is no force on earth that could prevent it from happening.
Similarly the promise of AI is simply too valuable for it to be held back by decree. Banning CFCs was free by comparison. The proper analogy is drug trafficking. The war on drugs did basically nothing to curb drug use. The lure of outsized profits is simply too strong for any centrally-planned policy to disrupt.
Harvesting eggcells is slow and difficult for a woman, instead of harvesting hundreds of eggcells for +15 points gain you could have cloned 150 iq person and skip most of inconvenience. It already exists.
What already exists, human cloning? Are you kidding?
Cloning of pet animals is long commercialized and humans are barely different.
Human cloning already is used, conditional on destroying embryo after harvesting stem cells ("therapeutic cloning").
I agree with much of what you say here, and agree that if it was easily and safely available it would be widely used.
We need to ask, however, *why* it isn't technologically feasible at the moment, and my view is that restrictions in place (such as not being able to cultivate embryos beyond two weeks) have significantly slowed down research in this field, such that it has remained technologically infeasible to do. Do you disagree?
I'm actually not sure. I think the only thing that was inhibited was direct human cloning. The tools that could potentially enable real genetic engineering (Cas9, CRISPR) have developed rapidly.
The tools have indeed progressed, but the particular application of those tools is very difficult and requires substantial empirical research!
<If it was possible to guarantee an extra 15 IQ points for children, there is no force on earth that could prevent it from happening
I actually think more people would want to select for beauty, health and longevity than would for IQ.
> I actually think more people would want to select for beauty, health and longevity than would for IQ.
100% - you forgot "athletic potential," sure to be popular among fathers aiming for the next quarterback, and height, too, which will be a big one for sons that will likely set off a fun arms race that will enable ordering people generationally on sight (much like we do with names such as Ethel and Gladys, or Jaden and Aiden).
Arguably more important than any of them would be "robust mental health" or "high happiness set point," but to my knowledge we don't have GWAS's on those things yet.
I peeked down the rabbit hole of jihadist propaganda (hezbollah children shows, how Iranian public intellectuals discuss their strategic aims when not speaking to a western audience, etc.) and it feels unhinged and scary. If I wasn't observing an objectively low/decreasing amount of terrorism in the world, or say Saudi Arabia de emphasizing antisemitic Islamic verses from their curriculum, I would be quite concerned. What is a sane way to understand/track the radicalizing and moderating forces in the Muslim world, and to project into the future likely outcomes. I am also aware that if I listened to Maga world for a similar chunk of time I would be genuinely worried about civil war, even though I actually think everything is totally fine and we are in the middle of a wonderfully productive, moderating, cultural self reflection and reckoning.
Look at the new masters of Syria: they’ve adroitly transitioned from jihadis to nationalists to maybe modern trade-oriented technocrats over a couple decades. People change in unexpected ways.
Bill Roggio's Long War Journal once upon a time was an excellent tracker of various Islamic issues and it's valence to the West (read: the US), with some ability to interpret various ideologies. I know it was pretty influential in US Mil circles; not sure if that's still true after the fall of Afghanistan, though.
Consider a change in perspective: how much do you think people should change their views after listening to the views showcased by LibsOfTikTok, for example?
My vague understanding is 'teachers brag about indoctrinating their students with far left cultural dogma in a way many parents would strongly object to', and stuff like that. If that's wrong please correct me.
I think that's less a change of perspective, than another example of the type of thing I'm asking about, which is broadly 'how do we judge the strengths of seemingly crazy movements that don't publicly self-reflect or have a internal diversity of views, when those movements actively threaten anyone outside of the movement who disagrees'.
Sure, you can add 'libsoftiktok content providers' to 'magaverse' and 'jihadists' on that list.
That's easy: just look at whether they're winning. In the case of the new right, well... They just won control of every branch of government of the most powerful country in the world, with popular support to boot. And they did that despite their party spewing blatantly fascist rhetoric in a supposedly democratic country! So that makes them pretty strong!
Ah, blatant fascism. Also known as "Centre-leftism circa 1980"
The deep state may not be officially a branch of government, but surely they count as one for this purpose.
...They'll capitulate just like the rest once they realize what they're dealing with.
Would you care to put your statements in a testable way so we can check later?
I disagree with your implication that "fascist" and "democratic" are incompatible: it's not LIBERAL, certainly, but a lot of decidedly illiberal policies are wildly popular.
Hezbollah and Iran, as Shia, are not part of the global Islamic mainstream. For all their rhetoric, they don't have much influence outside of their local struggles, and with rare exceptions they aren't responsible for global terrorism.
You'd get a better read on the pulse if global Islam by checking out Salafist channels on YouTube.
Metaculus
Metaculus is useless for prediction of "Black Swan" events, or even "light-gray swans" like the recent events in Syria, or October 7 (the evidence was there, but it wasn't receiving the right kind of attention).
I agree with NoRandomWalk that some Islamism rhetoric is alarming and under-exposed to western audience.
It could also have an effect on European cities with a large amount Muslim immigrants / refugees.
I'm struggling with an ontological account of causation and laws of physics.
First, I'm sure that there is a certain last level of depth that a correct account can reach. In the end, the world just is how it is. So if one finds a correct account then one will still be able to ask questions of why things behave like that, but there will be no further answer.
Second, physics has revealed some fundamental facts. For one, changes of things have, in a certain sense, no cause, that is, a thing that's moving will keep moving like that, it will reach other places, then other, and then other again, just because it *stays* in its state of motion. It's not that something would be pushing it continuously and causing it's ongoing motion therewith. Also, for a thing to change it's state of motion it needs to interact with another thing and this other thing will change then simultaneously and proportionally in its own state of motion. There is only interaction, no action on one thing by another that is not simultaneously reciprocated. All this can be subsumed under the principle of momentum conservation.
Third, this means that in a certain sense everything is just happening in accordance to certain "rules". Especially the feature of causation that a cause happens or exits before its effect is nowhere to be found at the fundamental or most general level of what is happening in the world.
Now, I'm pretty sure that that means that notions like "Susan got ill, because she drank that dirty water" make only sense because of epistemic limitations.
I mean, she did not get ill *only* because of that. She also got ill, because of her immune system being weaker than that of Li who never gets ill from dirty water. And also because she was born, because if she were not, then she for sure would not have gotten ill.
However, the dirty water is of most interest when one is looking for a cause, because for the near future and most people it's what one can do most about.
But if one considers absolutely everything in Susan's past light cone, and everything that once had future light cones that contribute to what is in her past light cone, then it's hard to speak off causation anymore. Then it again looks like events just happening according to the rules. But that's no problem for me.
The difficult thing for me is to give an ontological account of "the rules" then. At first I tried to say that things have properties and that these properties are interconnected in various way such that if a thing changes in one then it also changes in another. That's easy to see in a piece of metal that one heats up. It then also gains volume and at some point will even melt. That its properties are interconnected like that would then be a property of it too.
But the conservation of momentum and the fact that things "interact" would require it to say that even the properties of multiple things are connected. But if interconnections of properties are properties themselves, of which thing is such a property then a property?
The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them.
The only satisfying idea I have how to fix this lies in considering the possibility that there are not really multiple things. All electrons are excitations of one thing, the electron field, and there is already a working theory of this, likewise with all other fundamental "particles", and then all fields are somehow ... just one single field. There's no working theory of that.Yet.
If that were true, then "rules" like the conservation of momentum could be a description of a property of that one thing, call it the world, so that it's parts behave like they do.
The laws of physics just floating around as anything like commands is deeply unsatisfying to me -- after all, how do rocks know about them -- and so to anyone who feels the same:
Do you have other ideas?
"But the conservation of momentum and the fact that things "interact" would require it to say that even the properties of multiple things are connected. But if interconnections of properties are properties themselves, of which thing is such a property then a property?
The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them."
What do you mean by "taken as part of reality and not just the descriptions of reality"?
Continue with the example of conservation of (let's say linear) momentum. In a Newtonian world, this isn't a separate "law" - it follows directly from Newton's third law, that forces are equal and opposite. So which of these two "rules" is the real one: conservation of linear momentum or forces being in equal and opposite pairs? Are they both "part of reality"? What would that mean? Or are they really the "same law" - and if so, why not instead just say that reality exists and these are two ways of describing it?
Of course we don't live in a Newtonian world, but I don't think that matters at the level we're discussing (in any case we don't yet have a complete theory of everything). Even so, modern physics is similar with respect to symmetries and conservation laws. Again sticking with linear momentum: why isn't translation symmetry just a fact about reality? What would it even mean for it to be a separate "part of reality"?
>"The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them."
>What do you mean by "taken as part of reality and not just the descriptions of reality"?
>Continue with the example of conservation of (let's say linear) momentum. In a Newtonian world, this isn't a separate "law" - it follows directly from Newton's third law, that [...]
I distinguish between laws as part of reality and laws as descriptions of reality in this way: if the law of conservation of momentum *follows* from something then it is not a part of reality, at most it is about it. Because conclusions or propositions are what follows from something, not reality.
Often people talk like this about "laws", just like you, and I like to put quotes around it then, because they aren't *pre*scriptions but *de*scriptions.
But I believe that those "laws", because they are only descriptions, do not explain the relevant behavior (and even less cause it), and the behaviour of the relevant things must come from those things themselves. The way they behave is a characteristic of them. It's their way of being. It's how they are. And when one knows how they are then one can make really good guesses (calculations) about what they do next.
And so I conceive of laws, as part of reality, as properties of real things. Which leaves me with the difficulty of to which thing a law-like property belongs that governs the interaction of two. Hence the idea that those two things really are just parts of another thing and the law-like property a property of it.
I'm afraid that distinction might be too abstract for me. I chose conservation of momentum because it was mentioned up thread, but if that is only descriptive, can you give an example that is prescriptive?
And there are no prescriptive “laws” of nature, because the only things that follow prescriptions are things that understand them, not things in general.
There is a difficulty for me, but it's not with finding an example, but conveying my apparently hair-splitting distinction.
But I suspect you know that distinction very well, just not in the context of laws of nature
Here:
The volume of some amount of gas is proportional to its temperature. What I have just asserted is a fact, it is something real, namely a property that any gas has, with a gas being an aggregation of molecules that <insert further qualifications here>. It's a law as part of reality.
"The volume of some amount of gas is proportional to its temperature." is, however, not a fact, but an assertion that the volume of some amount of gas is proportional to its temperature. Its a law as a description of reality.
(On the next analytical level it's a sentence, composed of words, to express that assertion.)
It's the difference between, love being an attitude towards something and "love" being the name for that attitude towards something.
(On the next level "love" it's only a word. That's why it's stupid when someone says with a belief of saying something deep and profound that love is just a word. That's only true in a trivial way and just plain wrong if he means the actual attitude. I have this example from Daniel Dennett.)
It seems that this is describing the "map–territory relation" but then blending them again. I agree with your example's second paragraph. I disagree with the first.
Consider the following substitution: Some configurations of molecules are apples. What I have just asserted is a fact, it is something real, namely a property that such a collection of molecules has, with an apple being an aggregation of molecules that <insert further qualifications here>.
Now, is the form of an apple a "law as part of reality"? I don't believe so. When I say that a certain collection of molecules has the "property of being an apple", that's just a description in my mind. There is no platonic form of an apple which the molecules take on as a property. The molecules don't care what I call them (and the bonded atoms don't care that I call them molecules, and so on).
Similarly, if asked for an ontological account of Charles' law, my answer is that it's one model of reality: one map of the territory. An ideal gas could exist in reality, but Charles' law itself is only "something real" in so far as we can think about it and that our brains are real.
Or from another angle: if Charles' law is indeed "law as part of reality", yet also "conclusions or propositions are what follows from something, not reality"... consider that one could just as easily assert the axioms of kinetic theory to be the true property of an ideal gas. Then Charles' law merely "follows" from kinetic theory in the same way that conservation of linear momentum "follows" from Newton's laws or translation symmetry. So is Charles' law no longer a "law as part of reality"?
The laws of physics don't "float around like commands" and this is what I think is bothering you: You're taking "laws" to be something fundamental on the level that "things" seem to be.
Laws are just descriptions of how a model behaves. Or, in a different sense, you could say that the laws are rules to update a model which, if correct, should have properties matching measurable properties of reality.
However, the laws are related to *the model*, not directly to reality. There's no need to postulate that the laws are "somewhere" or that they "exist" in any meaningful sense.
"Laws are just descriptions of how a model behaves"
No, of how a thing behaves. The "laws" of which you speak *are* the model, or part of it.
As I have just answered plmokn, the "laws" you mean are only descriptions and as such do not explain the relevant behavior (and even less cause it), and the behaviour of the relevant things must come from those things themselves.
The way they behave is a characteristic of them. It's their way of being. It's how they are.
And so I conceive of laws (as part of reality) as interconnections of properties of real things, and with that as properties themselves.
Which leaves me with the difficulty of to which thing such a property belongs that "governs" the interaction of two. Hence the idea that those two things really are just parts of another thing and the (with a "law" referenced) property a property of it.
The laws we can describe are descriptions of models, not of things. There is no law which we know of which we can show applies to a thing in itself and not to our model of it. We can't even prove that the world is lawful, rather than just appearing to be so. Hence, I see no reason to postulate that laws are a part of the world or to explain how they could be so.
Now, supposing the world is somehow lawful, the idea that everything is only a single thing seems to be a decent way to interpret it, but definitely not the only possible one.
This is the (to paraphrase) "nothing but a bunch of math arguement" arguement stated famously at the end of Brief History of Time", not that I disagree, in fact it seems more and more likely. As you say, the math relates to the model, not the actual interrelationship between processes. The idea that there are actual "things" out there or Indeed "rules", now seems somewhat naive.
Years ago, I cofounded a company working on problems like this.
I'm not joking. We were working on a technology that enabled formal models of information, such that one could map any existing database onto it, generate databases from it, integrate multiple data sources using it, and so on. We had a working prototype of the third thing. And we weren't idiots about it - or at least, we were smart enough to avoid the sorts of problems one might see illustrated in an XKCD cartoon. We knew to scope the problem to real-world data systems making real decisions, and to not pretend everything could be expressible in relational algebra (it can't), and to not pretend one could just assume any data schema was authoritative just because it drove a billion-dollar business. Our unspoken motto was that if it drove real business logic, it was important, no matter how hard it was to implement in the RDB or stored procedure framework of the month, and if it couldn't be modeled in first order logic, then it wasn't worth modeling.
One of the things we had to model was causation. We had to, because we needed a model of events, and some events were considered to cause others, whether it be "the next opcode entered the CPU because instruction pointer incremented and the clock ticked" or "Fred killed John because John slept with Fred's wife". For this, we researched various causation models in metaphysics (I don't remember which ones our chief philosopher looked at, but we were accustomed to metaphysics research in this domain), discussing candidates such as "X caused Y iff Y occurred in all possible worlds after any possible world in which X occurred, and after X".
The good news for you here is that one virtually never has to break causation down this far. Nor to stuff like particle/waves propagating or anything like that. For most everyday problems, it's sufficient to define causation arbitrarily, at the level of everyday actions. You brought an umbrella because it was cloudy. A wingnut costs five cents because that's what the store said. If you're having to break the process down to finer detail, it'll be because there's an underlying criterion that naturally informs what will be the best way to lay out the subcauses, and they in turn will also be arbitrary (or empirically available by experiment).
"But if one considers absolutely everything in Susan's past light cone, and everything that once had future light cones that contribute to what is in her past light cone, then it's hard to speak off causation anymore"
Why? !maybe it's a departure from the film idea having a single cause, but maybe that just wrong.
"But if interconnections of properties are properties"
The interconnections are relations.
>"But if one considers absolutely everything in Susan's past light cone, and everything that once had future light cones that contribute to what is in her past light cone, then it's hard to speak off causation anymore"
>Why? !maybe it's a departure from the film idea having a single cause, but maybe that just wrong.
I mean not because of that angle, but because none of the little changes of each little thing is thought of being a cause of anything in the relevant physics, but just as going on according to laws.
When I said causation talk about Susan's illness makes only sense because of epistemic limitations, I meant that it *does* make sense, I don't belittle it, but it only makes sense at a point at which one is not looking at all and each of the fundamental interactions of particles.
At that level, to use this picture, the x's in a state-space representing all things with all their properties just slide around in a fancy fashion and with no position of them causing any further positions and with no possibility for them to slide any differently (except for quantum complication).
>"But if interconnections of properties are properties"
>The interconnections are relations.
If they are only relations, like this earring's temperature is now double than before while its volume is now 1.03 times greater than before, then we have only "laws" and that's what I find unsatisfying.
How does the "law" make things behave like that?
In the case of temperature and volume we found the mechanism and it is one that belongs to every piece of metal. It's a property of metal things to behave like that.
Hmm there is a hierarchical nature to our understanding of the world. We're not smart enough to grok the whole thing from just physics. So our path to understanding is to find new 'laws' at new levels... it works OK, but... Have you read "More is Different" by P.W. Anderson? https://www.tkm.kit.edu/downloads/TKM1_2011_more_is_different_PWA.pdf
I will. I just skimmed a recent article about it, and coming to the point of conservation of momentum being a symmetry thing, I'm strengthened in the each thing being a part of the same thing idea.
When all things (except the world) are parts of the same thing, the world, then every behavior of them is really a behavior of it.
Things don't need to manage to mirror (quite literally) their behavior in an interaction as described in a textbook or demanded by a "law", because it's really just one thing acting.
To formulate it provisionally: things are more parts of the world than the world consists of those things.
And that would also open a way to explain a system having properties which are hard to explain by the properties of its constituents. It would be a property of the world that it only shows under that circumstance or only shows along with that circumstance.
Sounds like a Platonic theory of forms could be a route to a satisfactory explanation for you.
The idea seems to be, every thing is like it is and does what it does because there's an immaterial blueprint of it that says it should. Or, a form the things want to take on.
That's not much different than thinking of laws of nature as commands to be followed.
You still wonder, how do things know?
It’s something that’s very hard to explain well without just reading a bunch (or all) of Plato along with commentary.
In essence it’s a theory that posits the reality of things, as we describe them. When we say “that’s a tsunami” we have identified something real with the word “tsunami” even if the atoms composing it at its start are all different from those at its end. It doesn’t claim the existence of these things in the material world as we experience it, as everything is imperfect in some way. We should really always be saying, “that resembles the form of a tsunami.”
The justification for it isn’t exactly empirical, but essentially our use of language (and perhaps thinking in general) requires the existence of forms (or concepts) that are permanent, singular, indivisible, and in a fundamental sense, more real than reality. If these forms are necessary to talk about anything as Plato generally claims, then the fact we can successfully talk about anything at all (which we obviously do) is evidence of their existence.
This also serves to explain what “causes” things to be, if the world is purely a deterministic dance where cause and effect are mutual.
Whether this is a tool for thinking, or truth, or something in between is for the individual to decide. Plato also emphasizes the important of finding the truth through one’s own thinking rather than receiving the knowledge from others.
First: I am not an ENS so the subtleties of your use of ontolgial definitions of 'things' and 'properties' might be lost on me.
I think you could make some progress by thinking about properties and things as entirely arbitrary classifications we came up with through shared language. It might seem obvious, that parts of a tree are its stem, leaves and roots, but the dirt surounding its roots and the birds in its crown are not part of the tree. Hower, we have made these distinctions through assigning names to specific subgroups of atoms because such descriptions are usefull to us in a practical sense, not because they reflect a ground truth description of reality.
Les intuitively, the properties of such subclassifications or 'things' work in the same way.
Our classification system through naming things forms layers, and properties of certain things exist only on a given layer of description. The higher level word for 'The forest' includes all the things described above and a multitude of other things in the lower subcategory layer. We can assign properties to the forest such as 'it's dense', but this property seizes to exist on the sub-layer of individual trees. All you can say at this level is: "well this one tree is closer to the other tree than this tree is to the third tree", but 'denseness' is a higer level description (and also requires for other forests to exist as its a relative decription which is true for most everyday properties).
Similarly saying: 'look here is a piece of metal and it is hard, but if it's temperature rises above a certain threshold it becomes soft' is a useful description of reality if you are talking to a blacksmith or an engineer. 'piece of metal' differentiates a set of atoms from the surrounding environment and describing properties like hardness and temperature describe something happening a the scale-layer of everyday human objects. But if you are a physicist and start looking for these properties on a deeper layer, they loose all meaning. You now need to use words like 'atoms' and 'bonds' and 'kinetic energy' to describe your observations. While you can establish an explanatory relationship from the 'kinetic energy' of individual atoms to the higher layer property 'temperature', and observe that the bond between individual atoms breaks if one of the adjacent atoms wiggles too fast, the property of 'temperature' holds no meaning on this layer.
And if you are so inclined and become a particle physicist you can start defining 'things' on an even deeper layer. You start talking about 'protons' and 'electrons' and suddenly the properties of 'wiggling speed' and 'bond strength' loose their meaning.
And of course for the very daring there is even one further layer if you participate in quantum phsics, but on this layer our agreement on definitions for 'things' is still vague and thus explanatory power for higher layer properties limited.
So in a sense you are right, the properties of electrons do not exist in the electrons themselves, but in their interaction (see atom bonds), and if you are a physicist and it seems usefull to you, you could define an 'electron field' as an entity that simply holds the properties of electron interaction. This is a usefull definition for Physicists because it simplifies talking about math ("this complicated formula describes electron interaction so we say it describes an electron field so we can categorize our various formula better"), but the main takeaway here should be that subdividing reality is up to us and not hard coded in anything and at deeper layer descriptions higher level properties are simply undefined.
In this sense the difference between describing electrons as particles moving through an electron field or describing electrons as exitations of a singular thing is the same as the difference between describing a bird flying through a forest as seperate from the forest versus talkig about 'the forest' as a singular thing; trees, birds, dirt and all.
All the level or paradigm switching you are talking about you are talking about inside the same ontological system.
The subdividing of reality (which I think cannot be made successful if done just arbitrarily) is only a subdividing of things into other things.
What each way of looking at reality keeps in common is that there are things and "laws" they obey.
And what bothers me for a reason I'm not sure about are the "laws".
If I can conceptualize things to change according to their own properties then I'm fine. But if I cannot, as in case of two things affecting each other -- you know, changing in an absolutely synchronized way, somehow "aware" of each other -- than I have this stray property, the "law" governing what they do, which seems supernatural to me.
> "The laws of physics, taken as part of reality and not just the descriptions of reality written in a textbook, seem to resist the characterisation as properties of the things "governed" by them."
Im not sure, that "the laws" exist as anything other than part of a textbook (in the widest sense as something we came up with to subivide and describe our shared experience of reality)
Just as the temperature of a piece of metal or the denseness of a forest is a a property which originates in a different layer of our descriptions, so are the fundamental forces that ultimately make up the textbooks Laws of Nature emergent from things interacting at a deeper layer.
Physics describes the 4 "fundamental" forces: gravity, electromagnetism, weak and strong force. They are fundamental because their nomenclators realy thought they hit ground truth in their description of reality. Newton described gravity as (in combination called classical theory) a central force without a communicating medium. But Maxwell, in defining electromagnetism, already talked about an abstract "field" communicating the force, or in our terms: the field contains the property of charges attracting and repelling each other.
Together these two form the Classical Theory and for a time Physicists believed they where almost done with this process of dividing up reality as the atom was the indivisible smallest thing, and only some confusion remained on where the forces came from, or what this "field" exactly was.
Well and then the Atom was split, first in theory, then in laboratory and we had to go a layer deeper in our descriptions. Since then we have come up with some usefull classifications and a lot of confusion, dug even one layer deeper, and gave names to things which produce these forces through some kind of interaction.
So if your question is: "if the laws of nature dictate, that two objects are pulled together, then where does that law reside if not in either of the objects?" Then the Physicist answer must be: "well we call it The Law of Gravity and its seems to be a phenomenon somehow related to something on a deeper layer; atoms seems to interact with a "field" called spacetime somehow and that interaction is mediated through something we aren't too sure about, but we called it Higgs boson."
So in short: these fundamental laws aren't fundamental after all and they originate at a deeper layer of reality we can't observe well enough to make satisfying classifications and explanations.
Sure, you can consider fields as separate things that interact with each other, or as one complicated thing that evolves according to internal rules. But the problem (if you want to call it that) is that what is or is not considered a "thing" is just a human judgment without any physical relevance.
At the end of the day the whole universe could also be seen as a single "state" that evolves from one instant to the next, but even then, you could say that the totality of states at every point in time, taken together, is actually the fundamental single thing, and slicing it up over time is just a human judgment once more.
"But the problem (if you want to call it that) is that what is or is not considered a "thing" is just a human judgment without any physical relevance."
Only kinda. I cannot *honestly* say that I believe there are no mind independent things. It might be worth treating a tsunami wave as a thing coming your way, even though it is *not*. For most of its existence it's just a row of changes in height of the water it consist in from moment to moment. The water it consists of at its creation is not the same as the one at its annihilation.
And the same goes for the supposed particles of the supposed fields.
But, at the end of such reasoning there always remains something, and that is a thing. The fields are things.
But I agree that judgements about that can be wrong. Maybe the fields are not really things either and even less the fundamental ones then.
"At the end of the day the whole universe could also be seen as a single "state" that evolves [...]"
It can be said, but not seen/thought-of to be *just* a state coherently. Because states are states of things, a state is the whole of the properties of a thing.
I cannot do without the concept of thing, and I always find on deep looking that no one else can.
Real thing: something that just exists, with properties, in a certain state, and is changing, into another state.
"but even then, you could say that the totality of states at every point in time, taken together, is actually the fundamental single thing, and slicing it up over time is just a human judgment once more."
Slicing it up, when? Judging, when? How long does that take?
If humans miss-perceive or miss-construe reality, then there is a succession of states of at least themselves. Their miss-perceiving or miss-construing proves that, or at least one is speaking self-refuting by asserting they miss-construe it.
"I cannot *honestly* say that I believe there are no mind independent things."
The question of whether things exist independent of minds is a different question than the one about how whatever exists should be divvied up into discrete "things", and whether that specific division has some fundamental kind of existence in nature, beyond the existence of whatever the things are made of.
"But, at the end of such reasoning there always remains something, and that is a thing. The fields are things."
To me it sounds like you are saying that once one has described what a field is, how it behaves, and how it interacts with other fields, one has not given an exhaustive description of the field. Instead, you think we should add an additional fact to our description of the field, namely that it is a "thing". The problem for me is that I cannot imagine what role this additional fact would play in nature or physics. The fact that a field is a thing does not play any causal role, it doesn't change anything, so why is it a necessary part of the complete description of the field, according to you?
"I cannot do without the concept of thing, and I always find on deep looking that no one else can."
There are many things I cannot do without, but that doesn't mean they are all a fundamental part of the universe. Even if I grant that I cannot do without the concept of thing, still the way I draw the boundaries between them can be different every time. Therefore the boundaries between things are (as far as nature is concerned) kind of arbitrary, and not fundamental.
"Slicing it up, when? Judging, when? How long does that take?"
Now, always. How long does "now" take? How long does "always" take?
"It can be said, but not seen/thought-of to be *just* a state coherently. Because states are states of things, a state is the whole of the properties of a thing."
Yeah, I meant a state of a thing.
In fact, we cannot prove that the fields themselves are fundamental. They are a model that happens to work well enough for everything we've tested them.
We have no reason to believe the fields are "things" other than we have no better theory.
I agree with your worries, sometimes I just wonder if it’s Turtles all the way down. It seems true that human reason is insufficient to answer base ontological questions, but is that insufficiency due to the former or latter - we don’t have enough of reason, or is reason not enough?
On how do rocks know - this relates to free will and I heard this cool answer that this point is actually a misapplication of language than a real problem. By natural law or rule we do not mean the same as legal rule - something compelled - by natural law we simply are saying what happens. At the same time that sounds like trickery to me and not satisfying.
The whole oneness thing u reach makes a lot of logical sense, I never really got the oneness thing until u said that
I’ll say why I’m not thinking about these things more - I think what we have is what we got, such that our objective laws and reason is objectively what we have; they’re Truth. They are true - (for us) permanent and unchanging, all that we can have, what we ought to have. (faith for the ought)
So I’m not too interested in explanations of oneness or the illusion of time, higher dimensional truth, etc. Ants debating the weather.
Super excited if superintelligence will surpass that truth and if it can show any of it to us.
You know, I don't want to know every fact about the world, but I would like to grasp the basics. And right now in a formulation of the basics I have "things behave as if they were following rules". "As if". Methapher. Aaargh!
But maybe I should settle with this. No one else seems to to better, so why of all should I :-D
Please Shinji, you've got to pilot the EVA and settle this once and for all... for your own sake. Find out why you came here; why you exist at all! Answer your own questions. And when you've found your answers, come back to me. I'll be waiting for you.
No, I just can't! I will never go in that thing anymore!
Isn't it more along the lines of "Things behave in certain ways. Those ways are repeated. That makes them predictable. From observation of these behaviours over time, we derive a set of predictions about how those things will behave in future. If we do this right, we are able to create a set of rules that we then call 'laws' of nature about how things behave, are likely to behave, and have behaved, and we check new observations of new behaviours/new things against those laws. If it all holds together, we've got science".
"things behave as if they were following [our] rules" sounds pretty good to me.
Our understanding can be only EITHER
1. "direct" experience (which is merely direct experience of the near-real-time simulation that our senses present to the brain (little more than a bare-bones summary of reality),
OR
2. metaphor where we compare our perception to mathematics or billiard balls or planetary systems or whatever. Nested metaphors all the way down to direct experience.
Science is pragmatic. If you can predict how a system behaves and you can affect its action then you can control it. If you can control it, it becomes technology and you end up with the iPhone.
If you want absolute truth, that is the domain of religion, but they universally fail by scientific standards.
So what would a deeper explanation look like? Are you asking for a theory that ends with "How could it be otherwise?"
I'm asking for a (ontological) theory that works and isn't dualistic.
Right now we have matter, that is, things, whether particles, radiation, fields, meteors or supermarket chains, on the one hand, and "laws", on the other.
Similar situation on the consciousness front. There are things and some of it have the "content" in their brain, that they holding a red apple in their hands.
Maybe I'm just neurotic. But it bothers me so much.
Laws are just descriptions. They are not supposed to be fundamental characteristics of the world, just statements about how we think the world behaves.
Perhaps you should think of it as "things" and "properties of things". Properties of things are what arises when you put things in time.
So your complaint is that the dimension of time behaves differently from three-space. :) Not much either of us can do about that. Fortunately!
Two thoughts:
1. it's a feature of language that we have nouns (objects), verbs (actions), and adjectives/adverbs (properties) so three things. This forces us to describe reality in those terms and, pragmatically, this has worked pretty well.
2. Our mind is at a distance from perceived reality so our perception is always us and it.
This is very well framed. As I was reading, I was thinking to myself (everything is the same thing, duh) and then I got the part where you say that.
But all fields aren’t necessarily one field
:)
Does anyone have a link to a 'as much as is knowable' about how to reduce chance of having difficult pregnancies, or identifying risk factors?
Possibly relevant details: age=30, occasionally quite difficult periods, very small hips, prone to mood swings. Open to alternative medicines whether or not they work in theory if they work in practice. Guessing that 60% of risk factor is psychological 'pill that genuinely makes babies super cute and desirable would be nice'
It is possible to considerably improve the healthspan, lifespan, future well being and IQ of neonates through optimizing pregnancy nutrition and cytoprotection. Actually there is a critical period in pregnancy that in in vivo models alter as much lifespan as state of the art lifelong supplementation of geroprotectors.
Not only is it likely that the future lifespan record will be via an optimized pregnancy, but even more likely the future IQ and RQ records will be driven by the optimization of such critical periods.
IIRC neural tube defects can be prevented close to 100% by concomittant administration of folate with inositol.
DHA and choline dietary levels (though max beneficial and safe dose to determine) have probably extreme consequences in brain organogenesis.
Many nutraceuticals can significantly improve nephron number, etc
This: https://www.ncbi.nlm.nih.gov/books/NBK573942/ is a 2021 meta analysis on interventions to deal with nausea and vomiting in pregnancy.
It finds strong evidence for the efficacy of a dopamine D2-receptor antagonist (metoclopramide hydrochloride - brand name 'Reglan') at improving overall feelings of wellbeing, and reducing nausea and vomiting.
It finds that ginger (tablets or food) shows some effectiveness in improving overall feelings of wellbeing and reducing nausea. In the study's conclusions, they recommend ginger.
Recommended doses are about 1.5 grams a day.
It discusses a number of additional interventions which I can summarize, if you like.
You can also see here: https://www.uptodate.com/contents/nausea-and-vomiting-of-pregnancy-treatment-and-outcome# which links to studies regarding various drugs.
a D2 antagonist seems insanely unsafe to the baby's brain, heuristically.
Thank you!
https://www.astralcodexten.com/p/obscure-pregnancy-interventions-much
Sorry let me clarify. I'm not concerned with the health of the baby, but the experience of the woman of the pregnancy itself, minimizing negative experiences during and physical consequences thereafter.
I asked GPT4 for predictors of unpleasant pregnancy symptoms, specifying that I was not asking for predictors of a damaged baby or of dangerous conditions a pregnant woman might develop -- I only wanted to know what made it likelier that the woman felt awful during pregnancy. I did not check its responses, but have found that GPT4 is usually reliable for questions of this sort. Here is what it said:
----------
- Individual Susceptibility
History of Motion Sickness or Migraines: People prone to nausea in other contexts (e.g., car rides, hormonal migraines) are more likely to experience severe pregnancy nausea.
Sensitivity to Smells (Hyperosmia): Heightened sense of smell during pregnancy, which varies between individuals, can worsen nausea.
Pre-Pregnancy Mental Health: A history of depression or anxiety increases the risk of prenatal depression and fatigue.
Pre-Pregnancy Fatigue or Sleep Issues: Sleep disturbances before pregnancy can predict higher exhaustion levels during pregnancy.
--------------
-Previous Pregnancies
Severe Symptoms in a Prior Pregnancy: If nausea, vomiting, or fatigue were prominent in a previous pregnancy, they are more likely to occur again.
Multiparity: Carrying multiple pregnancies previously can increase feelings of physical discomfort due to weaker abdominal muscles or ligament laxity.
-------------------
-Multiple Pregnancies (Twins/Triplets)
Carrying multiples can increase symptoms like nausea, fatigue, and discomfort due to higher hormone levels and a greater physical toll on the body.
---------------
-Lower BMI: Underweight individuals are more prone to severe nausea and vomiting (hyperemesis gravidarum).
Higher BMI: Being overweight before pregnancy is associated with greater fatigue, joint pain, and risk of developing sleep apnea, which worsens exhaustion.
--------
-Lifestyle and Daily Stress
High Stress Levels: Chronic stress can exacerbate symptoms like fatigue, nausea, headaches, and even prenatal depression.
Lack of Social or Emotional Support: Feeling isolated or unsupported can predict worsened mood and overall well-being.
Poor Diet or Nutrient Deficiencies: Low iron levels can worsen fatigue, and inadequate intake of vitamins like B6 may worsen nausea.
----------------
-Women who were physically active before pregnancy tend to report fewer symptoms of fatigue, back pain, and mood swings.
Lack of movement can contribute to stiffness, muscle fatigue, and sluggish digestion.
---------------------
-Sleep Quality
Sleep disturbances during pregnancy are common (e.g., frequent urination, discomfort, vivid dreams). Poor sleep quality in early pregnancy is often linked to worsening fatigue, mood swings, and irritability.
--------------
-Age and Health Factors
Maternal Age: Older pregnant individuals (35+) may experience more fatigue, aches, and joint pain.
Underlying Health Conditions: Conditions like anemia, hypothyroidism, or diabetes can increase symptoms like exhaustion and nausea.
---------
-Pregnancy-Related Discomforts and Conditions
Constipation and Bloating: Often caused by progesterone slowing digestion. Affected by diet, hydration, and physical activity.
Round Ligament Pain and Pelvic Pressure: More likely in later pregnancies and in women with poor abdominal muscle tone.
Heartburn: Worsens with certain foods and becomes more prominent in the third trimester.
----------------
- Genetics
Nausea and vomiting severity can run in families. If close relatives experienced significant morning sickness, the likelihood increases.
Summary of Key Predictors:
High HCG and estrogen levels (e.g., early pregnancy or twins).
Personal sensitivity (to smells, motion sickness, migraines).
Mental health history (depression, anxiety).
Previous severe pregnancy symptoms.
Stress and lack of support.
Low fitness or poor sleep.
If symptoms are significantly affecting quality of life, there are often safe, doctor-approved strategies to mitigate discomfort (e.g., dietary changes, supplements like B6, physical activity, or mental health support).
I've never come across any information on how to prevent bad pregnancy symptoms, but for the most common one, morning sickness, Diclegis and zofran are considered to be relatively safe and are also quite (although not universally) effective at treating the nausea.
https://parentdata.org/qa-zofran-morning-sickness/
Thank you!
In a failing attempt to understand why Luigi Mangione can't be tidily labeled on the political binary, we have mainstream news articles quoting Scott, like this one:
https://www.latimes.com/world-nation/story/2024-12-14/luigi-mangione-politics-united-healthcare-shooting
In it, the writers reference online speculation that Magione is Grey Tribe, attribute the phrase to Scott, and then quote him, with a link to the relevant post. (And, I'm pleased to report, refer to him by his pen name!)
Weirdly, I feel pretty okay about this? The think pieces trying to explain Grey Tribe / independent ideology to the binary masses don't feel as antagonistic as one would expect.
You've got it. It's that whole friend/enemy, ingroup/outgroup distinction. Now that he's done something they like, they can't go all-in on attacking Grey Tribe ideology. They'll try again later; a lot of it's that whole Silicon Valley-MSM rivalry that doesn't go away.
I'd point out Grey Tribe is a very specific 'liberaltarian-ish' group of ideas ranging roughly from old-school liberal to libertarian: pro-progress, pro-technology, pro-free speech, usually pro-immigration, 'socially liberal' in the 1990 sense of 'gay is OK' and 2010 sense of 'trans is OK' but not in the 2020 sense of 'no member of a marginalized group's feelings must ever be hurt'. It's adjacent to the more old-school-liberal types, but not really the same thing as, say, the remaining IDW people--Grey Tribe isn't nearly as anti-trans, for instance. They believe in biological sex, but if you want to change your body so it doesn't match that, go right ahead--it's the T in TESCREAL!
Non-binary ideologies that don't clearly map on a left-right divide are more various--I can think of Catholic social thought/Chestertonian distributism or generic populism on the other side of the political compass, but you can imagine others. A lot of the neocons are politically homeless now, for instance--look at the Cheneys stumping for the Democrats. Richard Hanania has more or less turned himself into one and you can see how much he hates both sides.
Theoretically political views are a space in R^n, where n is the number of issues, and we map that 'spectrum' by analogy with ROYGBIV onto a left-right R^1. There's a lot of information loss; even R^2 doesn't really make sense since it's not clear whether speech codes are 'socially liberal' or not. (Even the terminology gets difficult as 'liberal' means 'free-market' in Europe.)
Arguably the two political parties basically try to do a support vector machine (modeling and machine learning people PLEASE tell me if I've got this right and if you have a better description please say), drawing borders to include the largest clumps until they get to 51%. This doesn't really produce coherent ideologies or good governance, though Europe and East Asia have their own problems.
"look at the Cheneys stumping for the Democrats"
Who would have thought we would live to see *that* happen, eh?
Cheney wants America to rule the world. Trump doesn't care about the rest of the world.
It's not the least surprising to me. Cheney is from Wyoming, where the GOP tends to be extremist in their anti-environmentalism. To put it as plainly as possible: someone like Cheney doesn't believe there is anything worth conserving. The world just looks like money to them. As such - and the principle way such people can express their hostility to other values, to values of any kind - is to be for open borders. Again, because there is nothing to conserve, whether culturally or in terms of flora, fauna, and the landscape. Everything and everyone - is fungible with money - and they have made a great deal of money while disregarding the fate of the country, to which they are ultimately indifferent.
Of course their cause would lie with that of humanist progressives. I am not even suggesting they are cynical about it. Someone like GWB doesn't seem like a man with the capacity for cynicism. That's probably one thing that made him seem likeable.
Even Bernie Sanders understood this.
Such people of course tend not to choose to live amid ugliness they celebrate, but we're to imagine that's coincidental.
There's a principle I have that goes something like "If you're trying to explain someone else's behavior by claiming, 'Despite their stated reasoning their true motivation is actually $secret_motivation_x,' then $secret_motivation_x better be something that you yourself can actually see the appeal in and imagine being motivated by. Otherwise you're probably just making shit up to justify thinking your opponents are evil."
I think this comment fails that test.
This is a principle that by construction precludes your enemies being too different from yourself.
That's true. I don't view it so much as a universal rule, more as a useful warning that you should be extra suspicious of this sort of line of thinking. Sometimes after applying that extra suspicion you'll conclude that, no it really is right in this case, just as there are many other situations where you need to be on guard against motivated or self-serving reasoning, but sometimes the self-serving conclusion really is correct.
Money is something to value, that’s true enough. But evil - I don’t know if I can agree with you that humanism is evil, or a perversion. But you said it, and you do have that principle guiding you :-).
Oh, c'mon, "The Cheneys support Democrats because they hate the environment" is hell of a take. I'm going to tell everyone I know about this.
I like this test, and also noticed something was off reading the above comment, though I couldn't put my finger on it.
Of the reasons for Cheneys to defect from the GOP, "Wyoming conservatives want open borders to destroy the world" doesn't strike me as within the top 10.
I see that folks here aren’t real familiar with western GOPers and the history of conservation in this country lol. That’s perhaps not surprising.
I’m not sure why what people accept about the most famous GOPers of Kansas - is strange to them about Wyoming. Or indeed of Reagan, whom I’ve not heard described as “evil”.
And of course, Mike Lee didn’t achieve the power he holds, to halt conservation, by accident.
But admittedly, this is a big subject - other bigger than, uh, talking about talking, the preferred mode on much of Substack.
Everybody says antidepressants take weeks or months to have an effect. This does not match my experience at all. I've found antidepressants of various kinds to have a strong positive effect on my mood almost immediately. That effect fades soon, and after a month I'm not sure they're doing anything at all. What's going on?
This has made me pretty reluctant to increase dosages multiple times, since I've expected that to only lead to a temporary boost. Should I instead assume that while some of the initial effect fades, some of it remains, and eventually after a few rounds the dose might be appropriate and the drug would provide a longer-term benefit?
I had the same experience as you a few years ago. For both antidepressants I tried, I had near instant effect, and then the effect was gone in a few weeks.
Was wary of increasing dosage indefinitely and so quit them after a period of having side effects with no mood benefit.
I have no good answer to this, but I've heard this from several people.
You're probably responding to the actual boost in serotonin, which is fast but quickly adjusted for. Most people seem to respond to some downstream effect of the adjustment to higher levels of serotonin.
More claims at https://www.psychiatrymargins.com/p/how-antidepressants-work , which I haven't personally confirmed:
>> "It is clinically very obvious that a subgroup of depressed and anxious patients demonstrates a substantial and sustained response to antidepressants. This group appears to be the one in which positive effects on emotional processing, cognitive flexibility, and neuroticism are all experienced, and they are experienced in a mutually-enhancing manner. Because emotional processing changes happen quickly within days, these patients often start noticing signs of improvement in their emotional outlook within days. Patients whose depression is characterized by greater psychomotor and cognitive symptoms and low baseline neuroticism rarely demonstrate a quick response to antidepressants in my clinical experience. Response to antidepressants is often painfully slow, perhaps because neuroplastic changes have to slowly produce therapeutic effects without a synergistic boost from emotional processing shifts or improvements in neuroticism."
...Sometimes I wonder if the people saying that about anti-depressants actually have the kind of depression that's treatable by those medications. If you have low mood that's due to normal causes, it's going to get better over time because mood naturally has it's ups and downs. That would continue to be the case even after you took the medication (unless the medication has obvious side effects); you don't even need placebo.
For me, anti-depressants definitely work. Last time I stopped taking them, after a few months I completely lost the will to live. Not as in I wanted to kill myself, but as in my body literally stopped trying to sustain itself. I completely lost my appetite and my will to do anything, and my body was barely generating any heat. I was legitimately going to die if I didn't get back on my meds. Of course, once I did, my condition immediately started improving.
So, uh... my prediction is that if you're no longer feeling like that, or never felt like that in the first place, I doubt you're going to see much improvement from increasing your dose. Of course, I'm not a doctor, so please just discuss this with your psychologist. Also, if you're living alone, don't taper off of your medications without telling anyone, because you might die.
Did anybody teste the newest batch of ergonomic pillows. There is an aggressive marketing on my feed of this new version that looks more like exotic geometric figure than like a pillow. I am intrigued by the design and there is a claim that it helps side sleepers so I am thinking of getting one. However I have loads of different unsuccessful products in this area and I am not interested in another regular pillow, so I would like to hear if anybody had positive experience with it and should I risk. Otherwise I might try a quite more expensive solution that is a combination of snake pillow for pregnant women and some kind of platform to have your bottom arm and shoulder less impacted by the weight
I doubt there are studies on their effects on sleep apnea, brain bloodflow and glymphatic clearance
Do you have a link to what you mean?
Here is the link to one of the options of the intriguing one:
https://soya-paris.com/products/oreiller-soya-2-0
And here is the link to the expensive but somewhat more blunt solution platform:
https://fr.medcline.eu/products/dispositif-anti-douleurs-aux-epaules
Just like Scott I bought something like the first pillow once
I always ended up waking up in a position which defied their shape. You have all these images of people sleeping in a perfectly "ergonomic" position and it might work if you sleep like you're dead and don't move at all, because you can certainly lie down on one of them in a way that feels "correct". But if you move at all while asleep, you will probably not stay in that position. With ordinary feather pillows or synthetic equivalents you just mold the pillow as you go.
I guess it's worth mentioning that I also prefer harder mattresses and I really don't get the "memory foam". It feels too floaty and unstable as a support for the head or the body and I usually wake up with sore muscles when sleeping on something like that (I sleep better even when hiking and sleeping basically just on a patch of grass).
I sleep on my side, mostly the right side, and have a sore right shoulder a lot of the time because of it. I've tried lots of piling-up-pillow things, and am actively fiddling around with making a home-made version of what you're looking for. Here is what I am sure of so far: What you use has to be a negative version of what you will be supporting. So imagine lying down on your side in some soft molding material, with your arm in whatever position you like it in when you sleep: stretched straight forward in front of you; or upper arm straight forward and forearm pointed up at a 90 degree angle; or arm bent so that you hand is under head pillow, and upper arm is parallel to your body and slightly in front of it. You sink down into the mold until the highest part of lateral you, the side of your cheek, touches the molding material. Ok, when you stand up you will leave a hollow, and that hollow is what you want to duplicate with support you use. You need a support that involves a large hollow on the surtace of the bed — a hollow into which your lower shoulder and arm fit. In brief: You want innies that match the outies.
The most promising thing I have seen is this bed: https://sonusleep.com/?gad_source=1&gbraid=0AAAAAoe18FQscvigbb9P5D_TumjjMU-i2&gclid=Cj0KCQiAvP-6BhDyARIsAJ3uv7aIKnD5IG1GCG-UJe3DODOgpt4aBPMT9j-Z4TAYSouItEuH9G7uMlQaAhKlEALw_wcB
I have not tried it, but I’m pretty sure it actually would work.
I’m trying is to improvise something similar for the bed I have. I’ll order a wedge-shaped piece of very firm foam for the upper half of the mattress, and carve a hollow in it that is innie version of my downward-jutting shoulder, and my arm pointing forward from my body.
So looking at the things you have in mind: I’m sure the first one would not work. It’s really just a firm version of an ordinary pillow. It lifts the head a lot, but not enough to keep the woman’s shoulder from twisting forward of her body and taking all the weight, and that’s the thing that gives you sore shoulders. It is not possible to take weight off the shoulder just by lifting the head. In theory you can, but in practice it is very uncomfortable. You need to lift the head *and* make a hole for the shoulder. I think the second one you link would probably work, and now that I think of it maybe I should just buy one.
Later edit: Amazon is selling the Medcline one for about $250. It gets lost of pos reviews, but some neg ones that sound pretty valid. The most serious complaint is that the thing is made of soft memory foam, and squishes down so much when you use it that it isn't supporting your head and sparing your shoulder as it's supposed to. People also say it permanently loses some of its shape and height pretty fast. So actually I think I'll stick with my first plan and make something similar out of very firm foam, possibly with something even firmer inside to help it hold its shape.
I waited to respond because I already had a chiro appointment scheduled and wanted to see what he found. I too wake up with my right shoulder sore, and it's gotten to the point where it became a real problem. This is confounded by heavy use in grappling.
Long story short, the shoulder was out of its socket, and developed a nasty case of bursitis. Out of the gym for me for awhile. But - he reset the shoulder and last night it already made a big difference.
Where I'm going with this is it may be good for you to have that shoulder checked - if it's out of socket no clever pillows or support structures will really help until it's back in its place.
Wouldn't having it out of its socket be very painful? How bad was your shoulder hurting? The pain in my shoulder is pretty mild. I only notice it if I stretch my in certain ways -- straight overhead is the worst -- and even that is bearable, & I can do it if I really need to reach something. I know someone who used to have a shoulder that dislocated spontaneously, and he said that was the worst pain he'd ever felt.
The shoulder wasn't too far out. All it took to get it back was a mild click. Followed by an immediate sense of relief, like there was this tension inside that got released.
My morning pain was pretty severe, a burning sensation that I'd rate 6/10. This, I believe, came from bursitis. That inflammation is still there of course, will take weeks to clear out, but this morning the pain was less bad already.
The thing is, I suspect my shoulder was out for awhile, and it just got aggravated in the last 2...3 weeks. Before that I would wake up with a mild but annoying pain, maybe 3/10, and it would kind of gradually ease during the day - unless I stretched, especially overhead (yes, exactly as you describe)! So I pretty much stopped doing overhead presses, for example.
I am thin and wide, a bad build for shoulder health. It's not the first time this shoulder has gone out of place.
I tried something like the first one once. I didn't get anything out of it and couldn't really figure out how it was supposed to work.
I think https://xkcd.com/335/ has you covered
By the way, following up Scott's couple of posts on architectural tastes, the acclaimed new movie "The Brutalist" debuts on Friday. Those who read Scott's review of Tom Wolfe's "From Bauhaus to Our House" might have some doubts about the historical accuracy of this fictional movie about how an impoverished immigrant Bauhaus architect (Adrien Brody) is pushed around by a self-assured American capitalist (Guy Pearce) in the 1940s and 1950s.
I think it's a funny coincidence that after Francis Ford Coppola spent decades spent trying to make "Megalopolis", it would be released the same year that someone else released a cheaper & better received movie about an architect.
I haven't seen The Brutalist yet, but I've come across some reviews that say the one black character seems shoehorned in to establish that the protagonist is a good guy for being friends with him.
I've got some weird conspiracy theory about how the Democratic Party leadership is going back to making movies playing up Jewish-WASP antagonism to corral Jews back into the Dems after a revolt cost them some Ivy League presidencies...but I doubt these things are that well organized.
I think the British film industry might be more affected by PC views of diversity. Period settings are their bread & butter, and Britain really was overwhelmingly white until relatively recent immigration. I noticed this recently in the film "Joy", where there were two notable non-white characters, in which the black woman was a white woman in real life that they changed, and the South Asian man was an entirely fictional invention. They also changed the protagonist to fit better with socially liberal attitudes than she was in real life, and apparently never bothered to contact the one surviving member of the team to see if they were portraying her accurately.
https://slate.com/culture/2024/11/joy-birth-of-ivf-jean-purdy-movie-netflix.html
Holy crap. Is it really in the aims of making characters 'relatable to a modern audience' that the female main character has to be sleeping around?
I'm biting my tongue here because there's a lot I could say but it would be more heat than light, but sweet sassy molassey. Even for a 'period piece' (and now the 60s/70s are period pieces?), the characters have to behave like 21st century people. No, you can't have a woman who was Christian and didn't believe in sex before marriage be the heroine! In order to show that she is bold and brave and challenging the stuffy old orthodoxy, she has to be liberated! It cannot possibly be that someone with socially conservative attitudes also works on ground-breaking new research!
The review sounds like one of those online bingo card quizzes: check off all the squares! We got "Christians bad", we got "POC characters", we got "sexual liberation", we got "repressive societal attitudes", we got "making stuff up for dramatic purposes", we got - of course! - "Strong Independent Woman". Ah, I can't go on, it's all so tiresome.
EDIT: Looking around to find out what this congregation could have been, this is the contrast between what the review says the film does and real life:
"Before joining the research team, Purdy, who lives with her devout mother, is a frequent churchgoer and committed member of her congregation. But as word spreads about the nature of the team’s experiments, Purdy is told she’s not welcome at the church unless she quits her job, and ultimately her mother says she can’t live at home anymore as long as she continues the IVF research. Purdy nevertheless refuses to quit."
Well, first she was living with her father and brother at home as well as her mother, so far as I can make out:
https://www.sciencedirect.com/science/article/pii/S2405661815000064#s0045
"Purdy lived at the family home, a semi-detached house at 61 Langham Road, off Perne Road, Cambridge with her father (George), her mother (Gladys May) and her elder brother (John; for family tree see Figure 6)."
Second, again so far as I can find out, the Grantchester church she belonged to is Church of England. Maybe this is a very conservative church, or it was back in the late 60s, but whatever the attitude I can't see an Anglican clergyman telling a congregant "You cannot attend here anymore!" in that way. It's just not *Anglican* to be that direct:
"Jean is buried beside her mother and grandmother in the Church of St. Andrew & St. Mary, Grantchester, just outside Cambridge; where Louise Brown, the world’s first test tube baby, unveiled a memorial to Jean on 20 July 2018."
As for official Anglican attitudes, I can't find anything in the resolutions of Lambeth Conferences in 1958, 1968 and 1978 about IVF. The CoE has broadly accepted IVF (unlike us Papists) and issued a report about it in 1983
She was very young when she died, only 39, of malignant melanoma.
It's interesting you mention the stigmatization of non-promiscuous women. I never thought of that angle, but it's a really good angle, and a lot of women are very much not into sleeping around. It's odd (or perhaps it isn't) how removing old prejudices tends to create new ones.
I get the funny feeling Chesterton wrote something covering this exactly 100 years ago.
To be fair to the film, after someone at church rebukes her for showing up the actual clergyman disagrees... saying the church always wants people to repent and come back, implying that her work is still unacceptable to them.
If we're going for movie conspiracy theories, possibly more about "See? Immigration is great! Without immigrants, we wouldn't have had this beautiful concrete bunker building!"
Which, uh, may not have the appeal factor they think it does.
Going purely by the description in Steve's comment, I am up for seeing a Brutalist architect get pushed about by a self-assured client who don't want none of that ugly stuff 😁
This is another update to my long-running attempt at predicting the outcome of the Russo-Ukrainian war. Previous update is here: https://www.astralcodexten.com/p/open-thread-355/comment/76655933.
9 % on Ukrainian victory (unchanged from November 11, 2024).
I define Ukrainian victory as either a) Ukrainian government gaining control of the territory it had not controlled before February 24 without losing any similarly important territory and without conceding that it will stop its attempts to join EU or NATO, b) Ukrainian government getting official ok from Russia to join EU or NATO without conceding any territory and without losing de facto control of any territory it had controlled before February 24 of 2022, or c) return to exact prewar status quo ante.
36 % on compromise solution that both sides might plausibly claim as a victory (down from 38 % on November 11, 2024).
55 % on Ukrainian defeat (up from 53 % on November 11, 2024).
I define Ukrainian defeat as Russia getting what it wants from Ukraine without giving any substantial concessions. Russia wants either a) Ukraine to stop claiming at least some of the territories that were before war claimed by Ukraine but de facto controlled by Russia or its proxies, or b) Russia or its proxies (old or new) to get more Ukrainian territory, de facto recognized by Ukraine in something resembling Minsk ceasefire(s)* or c) some form of guarantee that Ukraine will became neutral, which includes but is not limited to Ukraine not joining NATO. E.g. if Ukraine agrees to stay out of NATO without any other concessions to Russia, but gets mutual defense treaty with Poland and Turkey, that does NOT count as Ukrainian defeat.
Discussion:
Since previous update, there has been continuing moderately alarming Russian advance, especially in the southeastern corner of the front, even though US military aid has not been curtailed in any way.
Also Ukrainian government so far totally failed to publicly articulate any coherent strategy, diplomatic and/or military, in response to US elections.
Their failure to take the initiative increases the likelihood that Ukraine will have to accept conditions dictated over their head by some bargain between “the West” and Russia, which will be probably not good. And I don’t think they have some secret genius-level plan; so far Ukrainian civilian leadership, as opposed to high military command (which has been consistently extremely competent), demonstrated mainly their haplessness, except for one brave decision at the very beginning.
* Minsk ceasefire or ceasefires (first agreement did not work, it was amended by second and since then it worked somewhat better) constituted, among other things, de facto recognition by Ukraine that Russia and its proxies will control some territory claimed by Ukraine for some time. In exchange Russia stopped trying to conquer more Ukrainian territory. Until February 24 of 2022, that is.
I wrote a second substack on Ukraine which might be of interest. It is a direct response to comments in Astral Codex Ten from people who disagreed with me for 3 good reasons:
"Russia and Ukraine Should Keep on Fighting, Not Make Peace": Making the best case for U.S. opposition to a peace treaty (Part I) https://ericrasmusen.substack.com/p/russia-and-ukraine-should-keep-on
Based on your definitions, I think "Ukrainian defeat" is the likeliest outcome, and it will be like how the Korean War ended. Both sides will sign a cease-fire, with Crimea and parts of eastern Ukraine under Russian control, but Ukraine not recognizing the legality of it.
I wouldn't bet much on this scenario. Russia would find it very unsatisfactory, because the example of South Korean clearly shows how unconquered part of Ukraine could be provided with forces and weapons in the future, even if it's not a part of NATO (any kind of NATO or NATO-aligned forces presence in Ukraine is simply unacceptable to Russia).
Further, consider that Korean ceasefire was signed after front lines stalled for a long time, with neither side achieving anything significant. For now, Russia keeps steadily if slowly advancing: as long as this advance continues, Russia has little incentive to accept any kind of peace talks aside from complete Ukrainian surrender. Importantly, this advance does not look like some kind of large concentrated push, like last year's Ukrainian counter-offensive: it's just Ukrainian army stretched thin after Kursk (mis)adventure, and Russian forces using this to gain previously well-defended positions.
But what is the alternative? I see three:
1) Likely: after Trump's plan that proposes Korean-like scenario, or anything at all that includes term "peacekeepers" in it, is soundly rejected by both Russia and Ukraine, Trump is outraged at Russia and orders to continue aid to Ukraine. The war drags on for another year or two, with Ukraine still mostly losing ground. It ends when either Ukraine or Russia has a sudden change in leadership (Ukraine is more likely, but all things are possible). The new government quickly signs some deal, based on realities on the ground.
2) Less likely: same as above, but Trump is outraged at Ukraine instead of Russia, and manages to actually cut US aid to Ukraine completely (he will have a hard time convincing the whole Republican party to support him in Congress and Senate in this, but maybe he can do it). The war drags on for maybe a year, with EU trying (and probably failing) to replace US. Then Ukraine gives up and agrees to whatever deal Russia is ready to accept at the time.
3) Very unlikely: Trump's plan is actually somehow acceptable to Russia (I would guess this includes some palpable guarantees of no NATO membership for Ukraine, conquered territories recognized by USA, Ukrainian military is reduced to self-defence forces, Russian language is re-instantiated as the second official language). Ukraine is probably furious, but terrified of losing USA aid, and so accepts, maybe haggling a little (mostly about size of reduced military and defence guarantees - it already has some treaties with EU countries, but will want another with USA).
You might note that all 3 versions end up with Russia getting what it want, and mostly differ in amount of suffering inflicted on both sides. This is because I don't see a better end for Ukraine, unless NATO intervenes directly, which is unlikely. At the same time, USA and EU should be pretty happy to prolong this war for as long as possible, weakening Russia without any loss of life their voters would care about, which makes alternative number 1 most likely, as it guarantees the longest war. The only thing that might change that is if Trump is crazy enough to think he can turn Russia against China, or at least to some kind of neutrality, by throwing Ukraine under the bus (I can see no other reason he might want to end this war, otherwise, unless maybe to win a Nobel Peace Prize, which, admittedly, might be a good enough motivation where Trump is concerned).
I wrote a second substack on Ukraine which might be of interest. It is a direct response to comments in Astral Codex Ten from people who disagreed with me for 3 good reasons:
"Russia and Ukraine Should Keep on Fighting, Not Make Peace": Making the best case for U.S. opposition to a peace treaty (Part I) https://ericrasmusen.substack.com/p/russia-and-ukraine-should-keep-on
You are underestimating how succesful Ukraine is at building their own weapons industry. Also overestimating length at which Russia can keep up its current pressure.
Obviously cannot replace the US, but in EU + Ukraine + serious funds (they get like 15bn/year from interest of Russian funds alone) + depleted Russian armor stocks + depleted Russian forex funds + depleted Russian manpower (losing like 4-500k/year at current rate), they will not need the full might of the US supporting them in 2026.
If all Russia can do is throw meat waves at Ukraine, with a total lack in armor support and air support (their bombing run frequency has dropped like 80% in past month) then simply delivering enough artillery shells + FPV drones will be sufficient to wear Russia out at not too great a cost to Ukraine.
15bn/year sounds like a lot, but Ukraine has a 40bn budget deficit. It also has manpower problems, or US won't be recommending it to start drafting 18 years old men (which Ukraine refused to do so far, because this will likely cause a lot of internal discontent). According to a recent article, it also has a problem with a (relatively) high desertion rate. EU also failed to fully reach its goal to supply Ukraine with shells, although admittedly it did a lot better than seemed possible at the beginning of the year.
I would be very wary about losses figures for both sides, as there are no good sources for them. Ukrainian officials crudely under-report them (80K killed since the beginning of the conflict is so transparent a lie it doesn't stand up to any scrutiny), and Russia just doesn't report any figures at all. Based on openly available sources, it seems to me that losses are about equal on both sides - and Ukraine has a smaller population. The myth of meat waves is told about both sides from the beginning of the conflict, but the truth is probably both sides sometimes have to send in troops without armour, though both prefer not to. Ukrainian officers like to talk about overwhelming number of Russian troops to explain the loss of this or that village, but combat footage, I think, doesn't support their assertions. By the 3rd year of war, both sides converged on pretty similar tactics.
I'd like to see your source for 80% drop in frequency of bombing runs, though if it really is there, the most likely explanation is weather, which is pretty shitty in these parts in November and December. Certainly there has been no shortage of missile and long-range drone strikes in past weeks.
Additionally, I think that if US stops aid, EU is soon to follow. The war is growing unpopular there. While AfD is unlikely to win Chancellorship in Germany, its popularity grows, and that usually means that its opponents will begin to accept some of its ideas, to stop their voters from defecting. This is already reflected in tougher immigration rhetoric from other parties, and this might happen to Ukraine, too, because it's a "cheap" topic: it doesn't need new taxes, new limits on freedoms, or any other things that has to be "sold" to a voter.
France is entering its own political crisis, and it's not obvious how it will resolve it, considering that the president and parliament are completely at odds. Hungary and Slovakia will continue their campaign to bring back cheap Russian gas.
Russian meat waves a myth? Come on man. There are thousands of videos out there that show it is very much not a myth. Hard to have a discussion with you if you deny this basic and obvious fact. OSINT estimates of equipment loss ratio sits between 2-6x in Ukraine's favor. So it is laughable to say that loss ratio is 1x for manpower.
Bombing runs adjusted for weather are down ~80%. And Russian Forex reserves are rapidly dwindling (about 1 year left) with inflation already sitting at 15-20%.
Russia will not be able to keep up this intensity for more than 6 months - year. They are gambling that Trump will pressure Ukraine and they can squeeze out a favorable peace deal. If Ukraine holds on, the conflict will continue in low intensity mode after that with Russian leverage rapidly dwindling.
> OSINT estimates of equipment loss ratio sits between 2-6x in Ukraine's favor.
Comparing https://datalens.yandex/t8or7ozcm6jeh?clckid=2d2578ba&tab=J8&state=349f4ee0102 (for Ukrainian losses, based on LostArmour OSINT site) and https://www.oryxspioenkop.com/2022/02/attack-on-europe-documenting-equipment.html (for Russian losses), the ratio is "at best" 1:2 per category, and some are 1:1 (e.g. artillery). With Russian capacity to strike beyond front lines, and advantage in number of artillery, I wouldn't be surprised if personnel losses are, indeed, quite comparable, and certainly are far from 1:6.
As for meat waves, I believe that BOTH sides use them SOMETIMES when they have no other choice, or when a miscommunication happens. But the tactic is not unique to Russia, neither is it used constantly. That you don't see videos of Ukrainian meat waves only means they are not widely posted on western platforms (quite unsurprising, considering dominance of pro-Ukrainian posters).
> Bombing runs adjusted for weather are down ~80%
I'm really interested to see the source for that now. I did a quick Google search, but couldn't find anything.
> inflation already sitting at 15-20%
That part IS true, but Turkey shows that a country can sustain that rate of inflation for a longer time. Inflation causes some grumbling, but by itself, it does not prevent a country from making war.
> And Russian Forex reserves are rapidly dwindling (about 1 year left)
This part I'm unsure about. News shows Russian gold-and-currency reserves actually growing a little during 2024. National Wealth Fund also grew a little. The economy is growing better than IMF prognosis. All in all, I'm fairly sure that, unless something drastic happens to the oil prices, Russia will probably be able to sustain war effort for more than a year.
All in all, I feel this discussion is going nowhere, as we both have different bubbles of information which are hard to break out from, and no special knowledge available to either of us. Aside from my request for the source of information about bombing runs, which is really interesting to me, I propose we conclude this thread and see what really happens in 2025.
> I can see no other reason he might want to end this war
...Is it crazy to think that Trump wants the US to be allied to Russia?
Yes, because it's too late for that. First, there is a matter of what Putin sees as broken promises from the West - he won't trust anything not written in stone, and probably not even that. Second, this is Trump's last term, and his successor is more than likely to try to renege on any agreements made by Trump.
The only way Russia and USA could be allies these days is in a case of direct war threat from China. But this is not likely: China doesn't need to conquer Russia by military force, because currents of policy already push Russia closer to China. All China needs from Russia is cheap resources, and it already has that, because western sanctions drive down the price.
However, I think that Russian top officials are not very satisfied with the way China supported Russia in Ukraine, and so there is a small opportunity for USA to use that dissatisfaction, not to drive any real wedge between the two countries, but to weaken the ties between them a bit. Kind of like Franco's Spain didn't directly join Hitler's war, only supporting Germany a little. But I don't know if it makes any sense from a strategic perspective, because it's not obvious what a "full weight" help from Russia would look in case of US/China armed conflict: while Russian army right now is one of the most skilled forces in the world in modern conflict, Taiwan and South Sea is a very, very different theatre where lessons learned in Ukraine would apply only partially, and Russia just don't have the right forces to participate there, anyway.
> Also Ukrainian government so far totally failed to publicly articulate any coherent strategy, diplomatic and/or military, in response to US elections.
Why would you expect Zelensky to *publicly* articulate their strategy? Besides, you are assuming a *single* strategy - the smart thing for a leader to do is to keep several doors open.
Because he kind of did. I mean, when you present something called "Victory Plan" people assume it to be a plan for victory. It's in the name. Yet, a plan is, usually, a sequence of steps, that, if followed correctly, should bring about the desired result. Instead, "Victory Plan" turned out to be a list of demands to Ukraine's allies. It's not "here's what we'll do", but "here's what we want someone else to do". Worse than that, it's "here's what we want someone else to do, even though we know they won't do that", since the linchpin of this "plan" is NATO membership. But Ukraine don't need NATO membership as a keepsake, you know. It's not like NATO can admit Ukraine, and then Ukraine will go "oh, great, that part of plan is complete, now we will go all Hulk on Russians, because our enthusiasm from being a NATO member infuses us with supernatural energy". It's "C'mon, NATO, please fight Russia for us, possibility of nuclear apocalypse be damned".
Well, OK, this "Victory Plan" was just to reassure the domestic population, so it's kind of unfair to make too much fun of it (no more than of Trump's speeches about "draining the swamp"). But the problem is, no one else came up with any kind of workable plan for Ukraine that I know of that does not involve direct NATO intervention. And since NATO is unwilling to cooperate, it raises the possibility that any kind of victory for Ukraine is impossible.
Just to be clear, Victory plan predates US elections (though I do agree it was not a good idea).
I am not expecting it, exactly, but I do think it would be a smart thing to do, in order to steer public conversation in both Ukraine and globally in a favorable direction.
I like going through links roundups, like Scott's monthly ones, or MR's and FTAV's daily ones. Does anyone have a best of 2024 round up anywhere?
I would be interested in that as well
Extraordinary. https://x.com/KelseyTuoc/status/1867686568787607746
2014 Rationality: Beware Isolated Demands For Rigor; touchy about any comparison to religion; exceptionally scrupulous norms around truth-seeking, centrality, consistency, etc.
2024 Rationality: explicitly making a case for selective rigor; "see, we're no worse than your friendly neighborhood religion"; downplaying differences of epistemology for PR / community building reasons.
One can quibble about the details of the positions or the degree to which the tweet is representative, but that's almost irrelevant as a matter of culture. The vibes have shifted starkly.
The core groups that later become Rationalists seem to have put a lot more of their identity in the early 2000s as being explicitly Atheist. Not lower-case, but specifically part of an acknowledged movement. Scott has talked about that a lot over the years, how the Atheists became cringey and lost favor, etc.
When a movement that used to be attached to specifically anti-religion loses the anti part, it can talk about the rest of the movement (EA, Rationality, etc.) using religious or quasi-religious phrasing and thoughts, without losing any part of the core message. Atheism, it turns out, was an add-on that wasn't necessary for the ideology. Though it makes sense given that a lot of Rationality came from a certain background that overlapped heavily with both atheists and Atheists. Atheism, perhaps, was a rally point for the movement to talk to each other.
There's a separate vibe shift that I've been seeing in a lot of Rat-adjacent writing that recognizes missing factors in a stable and enjoyable life. People that don't believe in God but see something of value in a religion or religious community. Surveys showing religious people as generally happier and better adjusted (fewer mental health problems, etc.). Religious people having more children. Whether this will result in Rationalists become religious or quasi-religious remains to be seen, but there's a lot of reasons to be skeptical about it. Perhaps they'll find a way to add ceremony and such to their activities, but it'll be difficult for a lot of their core reasoning to survive very much of that (and uncertain to be effective if that's not what makes religion "work" for their purposes or if the people involved get frustrated by the dissonance). SF Tech "Move fast and break stuff" mentality giving way to community and culture can either redefine the movement or destroy it.
I'll quote it in full, it's not that long:
"I don't think there's any ideology that, taken seriously, requires you to be a neurotic wreck. You can always take it seriously and simply not be neurotic about it.
But there are definitely communities and cultures that do a more or less successful job of integrating ideological claims that make a lot of people neurotic (whether that's 'Christ will return' or 'most people go to Hell' or 'everything trades off against saving children's lives' or 'the world is ending soon') with a healthy culture and a vision of the good that makes people stronger and more functional.
You can reject out of hand all claims with terrifying implications or you can try to build communities and intellectual cultures that are functional and non-neurotic and non-self-destructive while taking those claims seriously where they deserve it."
I'm not sure it's really representative of all that, I think; I'm more seeing it as reacting against some of the perceptions of EA as people being obsessed with the welfare of shrimp or hypothetical people a million years in the future.
The alternative is to just refuse to live by all the more challenging claims of your ideology du jour, like Catholics who use contraception or Muslims who drink alcohol. This annoys intellectuals who really believe in the ideology immensely but is what most people actually do.
Not that you're wrong about the vibes; you'd expect something to shift in ten years.
I agree with NoRandomWalk that I can't see what you're talking about.
I...don't see it/am confused?
This tweet is something I would not have been surprised by in a less wrong post in 2014, with a thoughtful/helpful discussion about possible tradeoffs but not reflexive pushback.
Everyone seems to believe that Luigi became mentally ill, perhaps became psychotic or schizophrenic before the murder. It bothers me that it's almost tautological that someone living such a normal, successful life in our society who commits such a violent act *must* have become mentally ill merely because we have no model for someone behaving that way who *isn't* mentally ill.
But isn't it at least possible that, like Raskolnikov or Carlos the Jackel, Luigi simply developed a violent philosophy and decided to live by it? His Goodreads review of the Unabomber manifesto in which he notes that violence is a valid method of communication was from 2021, well before the period when he possibly had a mental breakdown.
It isn't just the murder. It is the behavioral changes that took place before the murder.
All we know is that he disconnected from his family and most of his friends. Someone planning a murder might do such a thing for reasons including trying to put themselves in the right frame of mind and making sure they didn't in some way tip off what they were planning.
Yes, but the point is that there is evidence of mental illness apart from the murder itself. It is of course possible that he is not mentally ill, and that said behavior was caused by something else, as you speculate. Nevertheless, it is incorrect to state, as you did, that the claim of mental illness is tautological.
I believe that, for the majority of observers, the claim of mental illness is tautological, meaning that most people in society don't believe that such an act committed by a person with the background of Luigi Mangione was possible without mental illness present. The murder itself is considered sufficient evidence for illness with no room for daylight between.
I think people are just playing the odds. Your explanation is certainly possible but judging from history it's not particularly likely. Wealthy people don't usually martyr themselves. His ideology certainly played a big part but I'd be surprised if he wasn't suffering from acute mania or other serious psychiatric condition.
Psychiatry/psychology has a real blind spot when it comes to wickedness.
Disappointing that you don't see the murder of someone you don't know as an irrational act. We as a society need to agree that that is well outside of the norms of acceptable - crazy.
Yeah, but there could be a third category: Sane but evil. That label puts him outside the norms of society without sticking on a label, "psychosis," that we do not have enough info to use. And don't think it's self-evident at all that he had to be psychotic. In case you haven't noticed, people, including well-educated thoughtful sane ones, murder each other all the time. Consider the middle east. And it is really not that hard to inculcate young males into a culture where killing is OK. "This is my rifle this is my gun/ this is for work and this is for fun."
Rationality depends on what you want to achieve with it.
All I see is a bunch of pigs wrestling in the mud. All thinking they have the moral high ground. Murderer who used his pen, got murdered by a murderer who used his gun. Spectators who may have done some questionable things with their pens are outraged that a gun was used.
Pushing through corporate policies that will deny or delay treatment to make more profit and will kill people but is perfectly legal (partially because your corporate lobbyists have helped write the regulation that determines legality), is that well outside of the norms of acceptable? WIll you express an adequate amount of outrage at them too? Or is everything that is legal perfectly moral?
I want to clarify, murder is 100% wrong, but I am amused by the outrage against this particular murder that people are expressing now.
Legal versus not legal is a good proxy for moral and not moral. That's how society defines acceptable.
Outrage over bad policies should be directed at policy makers.
Me not paying for your healthcare is not the same as me shooting you in the head. You know that.
I'm getting sick of the prissy outrage too. I felt a flash of fierce joy when I heard the guy had been shot. Do I really approve of murdering CEO's if their company is unusually greedy and inhumane? No absolutely not. But I doubt my initial private reaction was rare. And now all these people are going around saying "I'm shocked, *shocked*" like the police chief in Casablanca when confronted about the gambling that has been going on in his bar.
How Soviet of you.
See also: Every Single Theocracy Ever
"If it ain't legal it's bad! John Brown and Hannah Senesh got what they deserved."
Someone not paying for your health care is not the equivalent of shooting them in the head.
If you can't figure that out for yourself, you can use the law as a guide. Or, you can lobby to change the laws.
Also, it's a little silly to equate our medical insurance system, as bad as it is, with slavery or genocide.
Crazy, sure. Evil, definitely. Irrational? No, not if his goal was to be a hero who is loved and supported by millions of people. If that was his goal, he wasn't delusional or irrational.
As I wrote below: It seems relevant that he's been unemployed since at least 2023. Why? He was a very ambitious student. He went to work at a company that developed software to help car shopping, then got laid off, or maybe he quit. It's not clear. Perhaps the corporate world broke his spirit, and he redirected his ambition in a completely different direction.
You are sane washing. His behavior is unacceptable by our social contract.
We need a new social contact.
Sure, like tax evasion or speeding. The more concise way to say "unacceptable by our social contract" is "illegal."
You want to judge. I want to understand.
Continuing from our previous conversation: I think you massively underestimate how much it takes to get psychologically "normal" people to actually make the decision to kill people. Grossman's "On Killing"* describes at length the process the US military has to take train its members-somewhat by definition mostly psychological normal-to kill other human beings, and it's not clear that it works all the time. Thus, it's reasonable to assume someone able to succeed in our society (which necessarily implies pretty good levels of sociability) that then decides to break one of the biggest social taboos we have is not acting rationally.
To get people** willing to kill without that training, you have to either get people who grew up with norms where killing is part of the culture (basically an honor culture) or throw extreme rage/intoxicants into the mix (if one watches enough The First 48, one notices how often "two guys got drunk/high and into a fight and somebody ends up dead" is a basic narrative of a murder, albeit it's usually not the modal one), be too dumb to work through the consequences, or yes, be psychologically defective (tends to go with the explanation before that, actually).
*I'd be curious to know what Scott thinks about Grossman's work; my understanding is he's viewed as something of a crank in the psychology world, but my impression is also he's deliberately avoided being part of that world anyway so there may be some element of him just not fitting in.
**as always, offer only valid on WEIRD people and not extendable to non-WEIRD psychologies, although most of the time we're talking about people living close to honor culture anyway
There seem to be a lot of psychopaths killing their spouses in cold blood in our society, if all those popular true crime shows are correct.
You write "normal" but I'm not claiming Luigi is normal in any sense, in quotes or not. I agree that the overwhelming odds are that he became psychotic or manic or schizophrenic or something pathological. There has to still be a tiny chance that he's simply a sane, however unusual, man who killed another man in cold blood for somewhat calculated reasons.
“There seem to be a lot of psychopaths killing their spouses in cold blood in our society, if all those popular true crime shows are correct.”
[insert airplane with chickenpox meme]
But they aren’t correct, is the thing. There are in fact so few of those people that those shows can go back decades and spend weeks (AI says an estimate for documentary shows is shooting about 1.5-4 minutes a day, and generally you can estimate about 45 minutes of show for an hour-long block with commercials, podcasts seem to spend about 2-3 hours per hour of research of actual air time) to produce these shows. Whereas even a Quibi (10-minute run time) focusing solely on homicides in 2024 the City of Chicago alone (595 at current count) would run for more than four straight days non-stop without commercials. An hour-long episode discussing Chicago’s murder would be a supercut of 6 second segments.
https://heyjackass.com/2024-year-to-date-totals/
If they went back to 2021 (856), they could almost run their very own Shark Week! There’s simply no time to produce that much content, though.
Indeed, IIRC Peter Moskos has noted that one reason for the declining murder clearance rate in America may be that prior to no-fault divorce, a high modality murder case was the “husband standing over the body of his rapidly cooling wife, sobbing and sobering up (note my point about “extreme rage or intoxication”-probably a factor in guys murdering their gals!) with a bloody knife nearby”. These were very easy murders to convict people on (the husband almost always confessed, usually to the arresting cop). You can actually see this trend in this old BJS report from 2005 (should be page 103 of the PDF).
bjs.ojp.gov/content/pub/pdf/htius.pdf
From the 2021 BJS data, more men are murdered by non-intimate family members (1797) than women are murdered by intimate partners (1689). The prevalence of those stories is not actually related to their frequency. Also “man killed by intimate partner” (something that happened to former coworker) is not that rare an occasion-approximately 1070, which is way more than the number of women killed by non-intimate family members or strangers.
https://bjs.ojp.gov/female-murder-victims-and-victim-offender-relationship-2021
Indeed, the primary reason for those stories is that the audience for true crime shows/podcasts is overwhelmingly female, and mostly middle/upper-middle class women who really strangely like hearing about other women (especially white women!) get murdered. Strangely, they don’t seem to enjoy hearing about black guys murdering other black guys over $250 of marijuana, or two Hispanic guys getting high and trading bullets, or even the white guys who end up in a drunken brawl that ends with one dead over the baby-mama they share, all of which are far more realistic depictions of what murder would look like.
https://psyche.co/ideas/listening-in-on-horror-why-many-women-love-true-crime-podcasts
I watched a lot of The First 48 at one point (which follows departments in a few high body-count cities) investigating murders. “Things that you would expect to see on true crime shows” was exactly one of the high dozens of murder cases that I saw (middle class white guy murdered his mom for insurance money; in the interview, they gave him a very quick and pro forma opportunity to confess-which he refused-because he’d left so much evidence in the process that they were looking forward to taking him to trial).
As I understand it Grossman is viewed as a bit of a crank in the history world as well. "On Killing" is largely based on the work of S. L. A. Marshall, which is now regarded as somewhere between dubious and completely fabricated. Although most reviews I've seen have admitted that "On Killing" has some directionally correct points, especially with regard to the rise of PTSD, I wouldn't put too much stock in its factual and historical claims.
Also, as a bit of an ironic side note for an author who wrote about the difficulty of getting people to kill their fellow man, Grossman and his institute have been running 'warrior mentality' police training courses, which seem to have increased police aggression and lethality.
I honestly think that's cope. Look at all the praise he's getting. Those people are one step from following him, with no obvious Schelling Point between them and him that stops them from going out and committing murder. But they don't want to commit murder, and very much don't want to admit it if they did. They know on some level that Luigi was *wrong* to do it. That his actions will not actually fix our healthcare system, and that murder is bad. So they have to rationalize between the very positive visceral reaction they are having and the reality of the futility and evil of Luigi's actions.
So he's crazy. Good-intentioned perhaps. He did it for the right reasons! But sane people know that's not how you go about social change.
Admitting otherwise says that you are a bad person who condones murder. Worse, someone who thinks they personally should go out and kill a healthcare CEO but is too scared to actually do what they think is right.
This seems to be recapitulating the arguments around bombing abortion centers or shooting abortionists a while ago: the so-called "pro-life" movement doesn't have a coherent response to "If you believe abortion is murder, act like it's murder," so they said the same kinds of things, "murder is always wrong," "violence is never the answer," "follow the law," etc.
The difference is whether clinic bombing and doctor assassination was cheered on by the respective community. It's hard to argue that a *lot* of people haven't been cheering Luigi on. I don't recall anywhere close to this response from the pro-life movement to clinic bombings.
Was Raskolnikov not mentally ill? He arrived at the culmination of his grand philosophy after steeping in his room for a month ruminating.
I doubt Dostoevsky would agree Raskolnikov was mentally ill. Was "mental illness" even a mental model in 19th century Russia*? The main theme of Crime and Punishment is guilt. Through a guilty conscience, the murderer seeks his own punishment. You could say that Raskolnikov's philosophy at the beginning of the book, which is that one can weigh the costs and benefits of a crime to determine if it is really a crime, is shown to be wrong after the psychological costs of the crime destroy him, because he has a conscience. One could say that proves he was healthy, just very misguided.
Our Luigi doesn't appear to have a conscience, but having a conscience seems like a quaint notion in the 21st century. I think Dostoevsky was warning about such a future.
*It looks like the modern concept of mental illness appeared (in Germany) in the 1880s, which is post-Dostoyevsky.
To be clear, I didn't mean Raskolnikov had a diagnosable disorder which may render a person morally unaccountable for their actions.
I meant that his mind was not healthy. His thinking was disordered and delusional, and Dostoevsky is making a point there. Raskolnikov's philosophy is taken seriously by many in intellectual circles, who toy with insane and destructive philosophies as social statement and hobby.
To take such a philosophy seriously, to act upon it to its logical conclusion, would require a madman - like Raskolnikov, after he shuts himself up in his broom closet of a room for a month, driving himself insane with constant rumination. And because Raskolnikov does have a conscience, the act drives him even more mad, until he acknowledges the evil of what he's done and repents.
The main theme of Crime and Punishment isn't so much guilt, as the inhumanity of ideological thinking as opposed to Christian morality. A sane, healthy human being regards murder as evil, full stop. To deviate from the Christian ideal of treating every human being as a sacred icon of God is to derange and dehumanize oneself.
I agree with all of that. Well put.
He dropped out of contact with his family and friends for about a year just before he committed the murder, which is odd behavior that wants an explanation.
One explanation could be that he didn't want any of them charged as accessories.
There is (probably) a difference between (a) actual psychosis; and (b) using violence for instrumental ends, even if the person is mistaken -- in the ordinary sense of being mistaken -- about whether their plan is likely to work.
So we want to see some evidence of actual psychosis.
Old school terrorist groups like the Provisional IRA are usually considered to not be diagnosably mentally ill. But the more recent "lone wolf" type killers ... seem more like psychotic breaks.
[There may be a very important diagnostic difference between "belongs to a large group of people who are all using violence for a shared political end" vs "it's just him, and he didn;t have any co-conspirators"]
It’s hard to square a level-headed, intelligent man planning a murder in his right mind with Luigi’s actual actions.
He left DNA where it could easily be found. He gave an ID to be duplicated with his real face on it where it could easily be found. He left his backpack where it could easily be found. He kept the murder weapon on him. He departed the city via public transportation. He didn’t alter his appearance before the murder. He didn’t change his appearance after the murder.
Like, unless he had a good plan to disappear off the face of the earth forever (he didn’t), there was no way he would have gotten away with it. The picture on the fake ID would be more than enough for his friends and family to recognize him, unless his murder got little to no publicity, which would invalidate the whole point of it in the first place.
Raskolnikov was definitely mentally ill by the way. The philosophy that justifies the murder definitely was not what actually motivated it.
I’d be more willing to attribute such a thing to Ted Kaczyski, who through careful planning was able to get away with murder for decades (although he wasn’t incredibly effective at killing people, he was very effective at spreading terror). If it wasn’t for something that a person in their right mind planning a terrorist campaign would reasonably not account for (a brother he hadn’t seen in years recognizing his prose style of all things), he probably would have kept getting away with it for a while longer, or perhaps never been caught. Kaczynski’s Manifesto is quite coherent and has a mostly self-consistent view for justifying his actions if you accept his premises as true (not justified in a general sense, but at least self-consistent with his philosophy).
Luigi’s Manifesto literally says nothing of substance and basically just gives his reason “Man, this whole healthcare thing sucks! I think. I don’t actually know that much about it but I’m sure smart people could tell you why it sucks. Anyways, I’m off to commit murder.” This is not the manifesto of a man with a philosophy that accepts murder as a necessary evil for some grander goal that makes it justifiable.
I don;t think he was mentally ill, in the sense of psychotic. I do think that living with constant pain can do very bad things to a person's outlook in life and make even "crazy" things seem rational.
Right, there's no room for fanaticism.
I suspect a lot of people are holding back from making the utilitarian 'if it saves more lives through helping create national healthcare it's justified' argument they really want to. After all, you could be accused of fomenting violence (a First Amendment exception) and a lot of people here might have cause to fear both the left and right.
I agree it's possible, it just seems less likely.
I bet >5% of people here have read the Unabomber Manifesto and would say something edgier than "completely wrong and 100% against this kind of thing" if asked to review it.
You can apply this reasoning in the opposite direction though: if I think of the many schizophrenics and people who have had mental breakdown that I have met, none of them have had the persistence, competence and wherewithal to build their own gun, get fake ID, and organise and commit an assassination.
Schizophrenia has different impacts / presentation in people with high IQs. Per this study: https://www.tesble.com/10.1016/j.eurpsy.2015.02.007:
High IQ schizophrenics had more mania (not always to a statistically significant degree) and lower negative symptoms. Those of them who showed a decline of less than 10 IQ points from their estimated pre-morbid IQ also showed significantly fewer "negative-disorganized" symptoms.
They note that this lends support to the idea of a psychosis spectrum or continuum.
The relationship between IQ and negative symptoms that they discuss had previously been found at lower IQ levels, and they found them at higher IQ levels. But even the high IQ individuals in the study, only averaged pre-morbid IQs of around 120.
Mangione probably had a significantly higher pre-morbid IQ than 120.
Accordingly, the symptom pattern that they describe for high IQ individuals, most notably, the lack of the negative-disorganized symptoms, would likely have been even more pronounced in his case
Psychosis is weird. I have seen lots of psychotic people whose faculties seems pretty preserved. I think the guys who shot Lennon and Reagan did some okay planning, and Mangione made lots of strange mistakes, like continuing to carry all the evidence on him.
Yeah, that's the other weird thing. He seemed to go to some considerable trouble to avoid getting caught and get away in the initial shooting, but then kept evidence on him that was 100% damning after he'd gotten away. It seems like a continuing desire not to die in prison would motivate you to dispose of the gun and not carry around a written confession with you.
People don't act on their moral convictions like that unless they're mentally ill. Stable individuals have the good sense to value their own survival and social acceptance over all else.
I think the discussion is a bit tautological because "not acting on mental convictions like that and valuing own survival and social acceptance above all else" is almost baked into our definition of mental health. So anyone who behaves differently is assumed to be mentally ill, it's just a matter of slapping on a diagnosis.
Sometimes those point in opposite directions (joining the army or dodging the draft).
I'm not sure; I suspect there's a bit of a U-curve here, where the extremely unstable and extremely stable both have the ability to act on their convictions, for different reasons. The 9/11 terrorists were not, as far as I could tell, unstable in any way. Quite the opposite, they had an unusually stable internal belief system that could survive being embedded in an environment which was very different to their internal beliefs.
Moderate stability leads to cowardice, like me. I'm definitely the sort of person who wouldn't say anything about people using slave-grown sugar even though I privately believed it to be abhorrent ; this is my current position regarding factory farmed meat, after all.
Mental illness is defined relative to society, and always will be, "Sluggish schizophrenia" "Drapetomania"
Motte: There are instances where a mental illness is basically socially defined.
Bailey: See, that's why this dude who has stripped off all his clothes and is out in the parking lot waving a knife around screaming that he is Jesus is really just socially defined as crazy.
In some places the rates of adhd dionisois in male children reach staggering heights; while there maybe a real thing of true adhd in 1%, I but call bullshit, genetic factors airnt this well correlated by state
https://www.ncbi.nlm.nih.gov/core/lw/2.0/html/tileshop_pmc/tileshop_pmc_inline.html?title=Click%20on%20image%20to%20zoom&p=PMC3&id=8580963_41598_2021_1233_Fig1_HTML.jpg
whats going on with that 16%?
so no, Im not mott and bailey this; it could easily be 10% of the population, where society assumes its ok to do something, and blames the victims.
It's the environment.
By which I mean, modern life is much, much more chaotic and demanding than it was, e.g., a century ago, and therefore the threshold for ADHD symptoms becoming debilitating has lowered significantly. As a bonus, social media and the like are actively making them worse. (And as a bonus, as you seem to expect, ADHD medication is highly regulated and if people really want to drug children to stop them from acting like children, or drug themselves because the drugs can also genuinely improve lives of otherwise fully functional individuals, it's all or nothing. You really need that diagnosis, so that's what you get.)
If it was self medicating or modern *work*, there be more adults then children with adhd. Its children tho, so its the schools, a constructed unnatural environment imposed on everyone a mere century ago but treated as if its foundational and unchangable.
I don't think "drapetomania" was taken that seriously in its own time.
This sounds a lot like the argument that ancient people didn't believe their religions. I don't know if there's good reason to believe this: it sounds absurd to us, but a lot of views currently taken seriously are no less absurd.
But "drapetomania" isn't at all analogous to ancient religion. It was just one guy who came up with it, and little evidence that it was widely accepted.
It is. And it isn’t.
A mafia boss isn’t considered mentally ill, or schizophrenic when he kills or orders a kill
That's because he effectively functions in a different society from ordinary non-criminal folks. In his society, killing and ordering to kill is required for "own survival and social acceptance".
It's not the killing that makes us conclude someone is unstable, it's brazenly going against social norms despite drastic negative consequences. Mafia boss's behaviour is in line with the norms of his in-group, Luigi's is not.
Given the popular support for Luigi's alleged actions, I question whether Luigi's behavior might be in line with the norms of his in group after all.
Beautifully said.
Luigi acted well within most Americans' understanding of our social contract.
To clarify: Most Americans over 25 whose net worth is in the bottom 65% percentile.
He acted within our social contract and we want him freed and given a parade.
No joke.
If you want a bloodless revolution THAT would be the way.
Right. Perhaps Luigi's goal was to become a popular hero. He wasn't delusional in thinking that he would be one. Maybe he doesn't mind being in prison as long as millions of people love and support him. It seems relevant that he's been unemployed since at least 2023. Why? He was a very ambitious student. He went to work at a company that developed software to help shopping for cars, then got laid off, or maybe he quit. It's not clear. Perhaps the corporate world broke his spirit, and he redirected his ambition in a completely different direction.
Often, I wished for a listing of young Scott`s aka Squid 134 writing on livejournal. Turned out, an outstanding person did it, six years ago! Here you are: https://archive.fo/fCFQx
Should I read Principia Discordia? When I look at that book's cover it gives me a lot of anxiety. I've never encountered a book that caused such a strong reaction in me, which makes me a bit curious.
Many of the ideas are now in the mainstream and not too earth shattering. I remember when I first read it, being quite shocked.
If you can find it, the Book of the Sub -Genius has something very similar, in a purified form.
Should be easy to find a PDF with no cover at all.
Whether it's worth it? For me, 60s counterculture is interesting for the ways in which it both sets the ground for modern wokeness and fails to meet its standards (a lot of them were young straight guys trying to get laid). However, I find that sort of stream-of-consciousness disorganized text to be extremely irritating. That's a very individual review and may not apply to you at all.
Read it but don't take it too seriously (I think the book itself would agree with this).
I think there are many versions with different covers; I'm curious which one had such a strong effect on you.
I just looked it up on Wikipedia, so the cover that's shown on the article.
Yes, my fellow pope, you should indeed read it.
I love bookcovers. The scary ones make me rethink my worldview
Hi all. I created an AI companionship platform focused on realistic interactions.
The platform is growing steadily and we have a nice comunity of users on Discord. My users recognize my platform as the most realistic out there. However, I have trouble keep driving traffic as this requires a lot of time that I'd rather spend developing the platform. Also I am not an expert in SEO, back linking etc. Is anyone willing to partner with me about this and help me organically grow the platform? Not looking for developers (I am happy to keep doing all the development myself), but for partners in the marketing area.
The site is https://theAIpeeps.com
If interested send me a DM :)
It’s generally a mistake to focus on SEO and backlinking in a crowded market, like AI companionship. You’ll be competing with larger players who can justify dropping $100,000 on a Bloomberg article talking about their product, which does about a hundred times more than any SEO expert with a moderate budget can accomplish. There’s just no way (without a meaningfully superior product, lots of free publicity or large amounts of money) to compete with large funded competitors when it comes to SEO.
Unless you’re selling something almost no one else is, or are ranking for a local search terms (Something like: “Computer Repair Shop St. Louis”) there’s no trick to getting it right.
There are other guerrilla methods of success besides SEO though. Reddit is a good one if you can grow your subreddit. Making a really good product that serves a niche, that gets picked up by a niche-celebrity is another.
Scott - I remember your position being that minimum wages don't cause harm. I wonder if you've revised this view since? How does a rationalist even arrive at this position in the first place?
The value of money is non-linear; while we may have games for making it more linear, such things need high volume efficient markets, job searching is not one.
If your starving, actually starving and you need 10$ to eat enough to fill full, it could be rational to enter into a 3$ pot Russian roulette game.
If your a multi billionaire and you were forced to pick between 1/10th of your wealth or a 50/50 odds of keeping everything.
If you were forced to gamble at a casino(or live somewhere where taxes are slightly lower if you play the stock market cough cough) it maybe rational to make a large spread of small slightly losing bets if your not confident in your capability to take large bets.
Given a non-linear curve of money, people with very little money, moving the floor up a little may make sense even if an economist would say "what happens if minimum wage is 1000$, would you be employed?"
This assumes (I think?) that minimum wage laws increase the real income of the poorest quintile in the long term (2 years out or more.) I'm moderately skeptical that this is true. To the extent that minimum wages are helpful, they seem to be an extra cudgel for things like wage theft from the poorest quintile of workers. Which still justifies them, but not for the reasons everyone else thinks. There are also some union contracts which tie wages to a multiple of the minimum wage. I'm not clear how influential such contracts are.
> This assumes (I think?) that minimum wage laws increase the real income of the poorest quintile in the long term (2 years out or more.)
How? I never referenced a single number?
Theres a minimum wage point that the poorest want, if only to enforce that all of them play Russian roulette if it causes unemployment
It's kind of broken to talk about MW 's causing harm without consideration of the amount. A lot of people do, though.
Indeed, or the distribution of harm.
Policy makers may be particularly keen that low skilled, perhaps low ability workers who none the less are willing to get up and do their best, get a decent amount of money for it. MW may help with that.
Whether it does in practice is, like all questions in economics, a matter for empirical study, and it may be hard to untangle effects. Broad suspicion about interventions by the state is warranted, but then again the arena in which those interventions play out is a product of political, cultural, legal etc factors going back years. It's seldom fair to see MW as the only intervention and everything else as being in some sense the free and fair market.
It follows from this that an MW policy may play out differently in different places at different times.
I'm not particularly concerned to advocate MW. I just really don't like the implication that for someone broadly "rationalist" there can only be one answer.
Interestingly, as the Neumark and Shirley paper points out, the distribution of harm from minimum wages is particularly concentrated among those with the lowest skills and ability, almost exactly what economic theory would predict
Any minimum wage, no matter what the amount, causes dead weight losses for society.
We've had minimum wage in the UK since July 1998, but it only applies to the employed, not the self-employed.
There have always been a lot of one-man businesses or small groups of partners subsisting below the minimum wage (or, before that, what those in employment would regard as a minimum wage). How do you think the big names in silicon valley got started (the socially acceptable phrase is "started in their bedroom/garage")?
More recently the minimum wage has forced a lot of people to become self-employed and join the gig economy.
My point is, whatever a government does, the people and/or the market will try to circumvent it so beware of unintended consequences.
I very much agree with the proposition that in balance minimum wage causes harm
I currently think it's really complicated, and it depends on the exact wage increase, the industry, and the prevailing economic situation.
(or rather, benefits > harms under some way of valuing things, since I think it's near-impossible for any intervention to cause NO harms)
I think this is what the literature says. Certainly there are many papers showing net benefits in some situations and others showing net harms in others. I understand that maybe one set of papers is wrong, but when I look at analyses by people I trust (sorry, I've forgotten specifics), they suggest it's more likely that they're observing real variation based on the particular area under study.
How is it possible for a minimum wage not to cause harm? I thought I'd previously seen an argument that made sense based on sticky wages, but I can't find it again or reconstruct it. All the arguments that make sense to me either assume companies were making dumb mistakes before the minimum wage increase, or that it increases jobs at the company's expense by shifting to an equilibrium where they hire more employees but get less profit.
Hopefully a commenter here can give a good explanation; otherwise I guess I'll retract my previous position and go slightly more in the direction of thinking the studies were flawed (though of course it could still be true that some minimum wages don't decrease employment *very much*).
Would you mind enumerating the buckets of 'harm' and 'benefit' that you're employing? I think that there may be an argument to be made that, in some situations, min wage laws reduce average hours worked. This would be a benefit to consider, independent of income. And I'm trying to figure if you're accounting for that, and how much weight you give to people having more free time, on average.
>How is it possible for a minimum wage not to cause harm?
Worker's productivity rises with the rise of his quality of life, and quality of life raises as the amount of available resources rise, and available resources for a worker are determined by his wage. (Meanwhile, society-wide, the amount of available resources rises as productivity rises.)
There are suboptimal equilibria where society as a whole would benefit from increased quality of life of its members, but no individual company can expect the resulting rise of productivity to offset the cost of raising wages, so it won't happen on its own. (As for why it might be so - note that companies are in many ways shielded from the consequences of their workers' low QoL. E.g., they're not affected by workers' health deteriorating faster than it should, they can simply replace them with new hires.) This is where the government steps in and solves the coordination problem.
>Worker's productivity rises with the rise of his quality of life
How much and what's your reference? I'm very skeptical. In any case I'm also skeptical that there's a firm connection between MW and quality of life as the former can cause inflation which harms low-earners disproportionately - the 2020 covid stimulus is a good example of that, though of course that wasn't a wage. Putting money in the hands of the poor tends to be inflationary because they consume almost all of their income.
A few years ago, Jennifer Doleac (an economist who mostly focuses on criminal justice issues) had some comment on Twitter implying she doesn't trust the minimum wage literature because it is so political.
https://x.com/jenniferdoleac/status/1350255741274349570
You shouldn't but that does leave you with no good evidence either way not one side winning decisively
If the bias of each source is known, we might be able to find something useful by looking at multiple sources with opposite biases (also if any specific conclusion is against the bias of its source, it's a lot more trustworthy than the other way around). I believe Scott used this approach recently in his prison article, let's see if we can do something similar with the minimum wage.
In April, California raised the minimum wage for fast food workers by 25%, which should be a high enough amount to see some effect. Limiting the increase to fast food workers should also give us some natural control groups like non-fast-food restaurant workers or fast-food workers in other states. The whole thing has been heavily discussed, so we should be able to find many sources with various conclusions.
As expected, every article I found was heavily biased, but let's see if we can learn anything useful:
- UC Berkeley study (https://irle.berkeley.edu/publications/press-release/new-study-analyzes-impact-of-californias-20-minimum-wage-for-fast-food-workers/) found a 3.7% increase in prices. Given Berkeley's bias we can probably treat this as a lower bound and conclude that at least some of the cost has been passed to customers. A mere 3.7% cost increase for an 18% wage increase (same source) could still be considered a win for minimum wage, but given the source I'm not comfortable saying that 3.7% is anything more than a lower bound -- so, inconclusive.
- The governor (https://www.gov.ca.gov/2024/08/20/icymi-after-raising-minimum-wage-california-has-more-fast-food-jobs-than-ever-before/) cites Bureau of Labor Statistics data and claims an increase in fast food jobs from 747k to 751k (interestingly, the UC Berkeley study which should have a similar bias didn't claim any job increase).
- The conservative magazine National Review (https://www.nationalreview.com/news/california-fast-food-restaurants-shed-thousands-of-jobs-after-20-minimum-wage-hike/) cites the same data but claims a decrease in fast food jobs from 571k to 565k (slightly different time period, but more importantly the numbers differ by almost 200k so clearly they used some different slice of the BLS data).
Since the governor and NR both relied on BLS data, we should probably dig in some more. Neither source links to the exact data they used, but the governor at least links to the relevant search form (https://data.bls.gov/PDQWeb/sm). It's not clear to me how to find fast food jobs specifically, but here are some relevant categories I could find:
- "Restaurants and Other Eating Places" slightly down based on September (most recent available month, 1352k to 1351k) but also slightly up if you pick a different month
- "Limited-Service Restaurants and Other Eating Places" (sounds like fast food but I don't know if it matches California's definition) slightly up (737k to 738k) but was slightly down in June (conveniently the exact month picked by the NR article above) and also the growth did seem to slow down significantly around the time the new law took effect (year over year growth was 10k in January but hovers around 1k since May)
- "Full-Service Restaurants" (the control group?) went down (615k to 613k), not sure what to think about that
At this point I lost interest, but I'm convinced we can learn something useful about minimum wage from California's fast food experiment and I'll probably take another look at the most recent data in a few months.
Might I suggest also considering total income and total hours worked since both are likely impacted. There are arguments in favor of a minimum wage, but the whole notion of a 'living wage,' specifically, which is perpetually trotted out becomes moot if total wages of the poorest quintile don't increase significantly.
Here’s two pieces of economic theory that support a minimum wage. The first is the idea of a “reserve price”, which can be used in an auction to maximize revenue. In this case, it says that even though there may be less jobs, a properly set minimum price causes more total revenue to flow to the lower class. The second, which is less theory and more analogy, are unions. Minimum wage is the government collective bargaining on behalf of low-wage workers. As is standard, a union can push too far and tank a business, but can also simply change the distribution of profits within a system.
An auction is a terrible model to use in most settings that are not auctions. But even a reserve price in an auction is set by the person selling the good, not by government!
And government privileged unions are terrible news for most industries. Government is the most govt. privileged of government unions!
>causes more total revenue to flow to the lower class
And why is that a net-positive for society? Boosting low-end income with the covid stimulus in 2020 just resulted in a lot of inflation.
Not only did giving the poor more money cause inflation, it led to them infiltrating higher income areas and bringing their dysfunction. E.g. the massive increase in unruly airline passengers following the stimulus.
I don't pretend that any of these are knock down arguments. And I also don't argue that min wages necessarily increase the real earnings of the poorest quintile. But...
1. Productivity also decreased significantly during Covid, which also contributed to inflation. 2019-2021 is a problematic time to base studies on, because there's a lot of noise.
2. In normal times, some level of consumption among the poorest individuals could also be considered investment. i.e. If you can't afford a reliable car and gas, you cannot optimize the value of your labor. The cost of borrowing for low-income individuals tends to be particularly high.
3. Property crime is reduced by government programs to provide a minimum level of consumption. Presumably, property crime is a negative sum game. Though how much of one is beyond my pay grade.
4. We currently have social programs in place. Increased wages may reduce some of these government programs. It may also better navigate benefit cliffs, which are problematic. (But which could be easily addressed without a minimum wage.)
5. If consumption could be reallocated to some extent, it should increase average utility. (At the expense of maintaining property rights, and the bulwark against authoritarianism that tends to entail.)
Again, none of these are knock-down arguments. But they do address the (possibly rhetorical) question of 'why?'
Sure it's complicated, and I agree that supply problems were partially responsible for recent inflation. But certainly not entirely.
Agree that some level of wealth transfer to the poor is necessary to keep society stable but I think that minimum wage is probably the worst way to do that. It just distorts price signals and carries inflationary risk. A negative income tax is much better.
>"How is it possible for a minimum wage not to cause harm?"
The trivial case is when it's not a binding constraint, id est, if every worker can command a higher-than-minimum wage through normal market mechanics then the law is inert.
Re: How is it possible for a minimum wage not to cause harm?
See: mathematical step function. If the increase in the minimum wage is less than a certain amount it won't cause any effect in employers (who, unless their business is right on the edge of failure, have some "cushion" to absorb price change shocks.) Price increases do happen in other business inputs all the time-- why should wages be any different from the electricity rates going up? Compare this to personal behavior if the temperature in a room goes up by one degree, Most of us won't even notice and will continue to go about our business. But if the temp goes up by ten degrees, yes, we will notice and react.
Human behavior is rarely a smooth linear function, and assumptions that rely on linearity in economics are most generally wrong.
Does this not cause harm to the employers? You may look only at the businesses right on the edge of failure, but then you should also look at the ones adjacent to that, who will now be on the edge of failure. Then look further at the ones who now must be concerned their "cushion" they carefully engineered to withstand the unexpected must now be reevaluated, etc.
Fundamentally, why should the employer have to pay more? As far as I can see, it is only if the workers are worth more than what they are getting paid, and do not have the freedom to seek an alternative job where they can be fairly compensated. And getting "fairly paid" is often a judgement call without any objective way to calculate it.
Having to pay tax hurts employers but that's not a flat argument against tax.
If you're running a business you certainly cannot count on your expenses being set in stone. If labor goes up why is this any different from increases in raw materials, utilities, commercial rents etc. etc.
I should also mention here that the US federal minimum wage is a joke these days. It is far below the"natural" minim um, wager that the market itself has set. You can't hire people for $7.25 these days. A few years back I recall people outraged at the suggestion the minimum wage should be $15. Nowadays in a lot of places that's the true minim,um wage set by supply and demand.
There was a short period in which it appeared that state minimum wages (which apply to well more than half the national population) were getting close to true market minimums in various regions. That moment passed though, the state minimums are now mostly well below market and aren't catching up anytime soon if ever.
If the Trump plan for mass deportations of immigrants actually happens then that will further reduce the national unskilled-labor supply and presumably make minimum wage laws even less relevant. Particularly if Trump's announced desire for the deportations to include some non-citizens who are here legally as well as citizens who are the children of illegal immigrants, is realized.
Trump will not be able to deport citizens (other nationalized citizens who have been convicted of serious crimes, as currently).
As a business, yes, your expenses can go up, and if they do, you must decide whether to eat those costs or pass them on to your customers. This is straight-forward inflation. I don't see how artificially setting any kind of minimum wage is not harming the employers.
Except, as you mention, when the mandated minimum wage is irrelevant. And in that case, why have one?
Thanks. Someone else has shared the Neumark and Shirley review which I personally think is very much right. But also, the Bayesian process on this (imo) should go as follows -
1) We have good reason to believe that prices encapsulate useful information generally and that demand reduces as prices increase
2) Artificially set price floors or other price controls will reduce demand where it does not need to be reduced and otherwise introduce a host of inefficiencies that result in poor outcomes for everyone
3) We have loads of good evidence from many many different sources for the above propositions - rental markets, product markets, labour markets, you name it
4) This one weird trick in labour markets lets us believe a politically convenient free lunch exists where a price floor will not cause outcomes to be worse.
5) We should not update on 4 very much, if at all!
What do you think?
Neumark and Shirley (2022) - "Myth or measurement: What does the new minimum wage research say about minimum wages and job loss in the United States?" - Is a recent review of the topic that came to the conclusion that the expected effect of minimum wage is probably negative, with a nice approach of asking the researchers directly which of their paper's estimates is the most relevant. From the abstract:
"Our key conclusions are as follows: (i) there is a clear preponderance of negative estimates in the literature; (ii) this evidence is stronger for teens and young adults and the less educated; (iii) the evidence from studies of directly affected workers points even more strongly to negative employment effects; and (iv) the evidence from studies of low-wage industries is less one-sided."
Though I don't think their conclusion is the new consensus, They changed my view on the topic (update from "probably no effect" to "probably negative") though the literature is very American, and if you have some local estimates you might prefer it. In Israel for example the last estimate I've seen is for no adverse employment effect and some redistribution from small business to low wage workers (either the Treasury or the Central Bank published it, I don't have it on hand).
Oh I know the literature very well, but thank you for posting it. I was just curious about Scott's views
There’s more to the world of economics than Econ 101. You havent rigorously defined harm there but the position against to the minimum wage is either that it causes high inflation, or higher unemployment.
Sometimes when the claim to higher unemployment doesn’t pan out there’s a claim that the increases in employment aren’t as high as they would have been which is hard to verify.
One might venture that for any economic configuration there is a natural value of "minimum wage". One that identifies the lowest "value" that configuration places on labor. If the official minimum wage is less than the natural value, then there is no economic impact to increasing it. The natural value is determined by the aggregation of the feelings (waves hands) of the various cultures that are part of that economy. Huh, I guess this really doesn't cash out to anything. Oh well, it's typing practice.
No, you're very correct in your thinking here. There's a wage at which employees are willing to work and employers are willing to hire them. That's the natural minimum wage. Raising the legal minimum above that rate will cause disruptions, while leaving it below that rate will have minimal effects.
There's another aspect, which is that some employees are worth more than others, even filling the same exact role. Two people working an automated assembly line, where you would expect employees to have very similar levels, may have OOM differences in productivity.
There's some obvious reasons for this, even beyond how much that machine produces with minimal human interaction. Someone with really bad attendance is obviously less productive than someone who shows up on time every day. Someone who picks fights with coworkers costs the company. Someone who sues the company obviously costs a lot more. Someone who inexplicably breaks the machines a few times a week and causes hours of downtime. And then the opposite of all these things - an employee who is always at work, calms down disgruntled employees, and whose machines break down less than company average.
Minimum wages may or may not cause a reduction in employment at a given company. They absolutely cause a reduction in employment for the least employable and may cause an increase in employment for the more employable, including people that were already able to make more than minimum wage. If companies are already paying a higher wage regardless of the person they hire, they may as well hire better employees.
The specifics matter a lot. Some companies will pass on costs to final consumers, but only if they are able to. Some will shut down or hire fewer people (maybe close during their least productive hours). Some will eat into their profits. On the whole, it's very difficult to predict the actual results because the results will depend on a wide range of factors simultaneously. The Econ 101 overall tendencies will be generally correct, but with enough factors that confound analysis as to make that less than universal. Economically depressed regions or industries are more likely to shut down and lay off, while growing places are more likely to pass along costs.
"[I]ncreases in employment aren’t as high as they would have been" *is* "higher unemployment".
Pretending otherwise suggests motivated reasoning. Or willful ignorance.
No it isn’t. It’s the typical scurrying away when the argument is proven incorrect.
There will be unemployment when we raise the minimum wage is clearly not that same as we won’t hire as many people we could have but will still hire lots, enough to handle labour growth.
Of course if economists actually had a formula for this - something like a 10% rise in the minimum wage will always and everywhere increase the actual unemployment rate (not the completely unverifiable woulda coulda shoulda hired more people in a counter factual world we put into our simulations) we could falsify it but they don’t so we can’t.
Anything else is wilful ignorance.
>"There will be unemployment when we raise the minimum wage is clearly not that same as we won’t hire as many people we could have but will still hire lots, enough to handle labour growth."
"enough to handle labour growth" assumes the conclusion, try again.
The problem with such statements is they are counterfactuals: we can't prove or disprove them because they aren't real. It's no different from saying"If Ronald Reagan had't been elected in 1980 the USSR would have won the Cold War". Since we have no access to parallel universes where (maybe) that did happen, there is just no way to know that.
All models are wrong, some models are useful. The supply & demand model is consistently useful across a wide range of contexts, and the conditions in which its fundamental results can be expected not to hold (e.g., monopsonistic power over most/all of a labor market) are rare and not subtle.
FWIW, my current day job includes modeling conterfactual supply & demand conditions for multiple markets to guide various sales & marketing efforts; I'm painfully familiar with the challenges & limitations of drawing useful inferences from them.
Assuming this is true, doesn't this make the claim about raising minimum wages unfalsifiable?
Any time the employment either remains the same or even increases after a minimum wage hike, you can always trot out the same argument that "it would've gone up more without the wage hike". What evidence is required to show this may not be the case, when "it actually not being the case" is not enough?
Reminds me of the politics rock described here:
https://open.substack.com/pub/astralcodexten/p/heuristics-that-almost-always-work?utm_source=share&utm_medium=android&r=5sutf
Two separate points here:
1. Implying, as Peter does, that emphasizing slower increases in employment is moving the goalposts is arguing in bad faith; that's what I'm primarily pushing back on.
2. Differences in rates are harder to detect than differences in levels, but not impossible (thus still falsifiable). Extrapolating from any single instance is akin to rural Minnesotans asserting that climate change is bullshit because they're currently in a blizzard.
read Card and Krueger 1994.
I've read it. It's a very badly done paper. That Card went on to win a Nobel is an indictment of the state of the economics profession
Beware the man of one study.
I don’t think anyone should read Card and Krueger as proof of anything (they should instead read it alongside the entire body of research answering the same question) but it does answer the question “How does a rationalist even arrive at this position in the first place.”
If a single study (especially one from before the replication crisis) provides sufficient evidence for someone to update to "minimum wages don't cause harm" then their prior is too weak; the Law of Supply & Demand is *very* clear that binding price floors create surpluses.
If someone said “how could anyone begin to enjoy reading books,” I might point them to a book I like. It would be silly to respond by pointing out that I haven’t established that the entire hobby of reading is fun - I was just providing a jumping off point!
Beyond that, your comment about the replication crisis makes me think you’re not very familiar with this area of research.
Perhaps overly primed by having seen too many people treating Card & Krueger as the final word on the topic, I interpreted your initial response referencing that paper as a drive-by defense of minimum wage laws rather than an answer to the "how" question.
The replication crisis reference was a bit of a drive-by of my own, and for that I apologize. I had followed the minimum wage research somewhat closely over a decade ago (because the headline results were so contrary to my priors, if the findings were sound they'd've warranted a deeper reexamination of my economic understanding); last I paid significant attention was a meta analysis with a dot plot of (I think) employment elasticities to minimum wage by study, which showed a clear but to-me-surprisingly-small negative effect. The qualitative result was in agreement with my prior (so no soul searching) but the quantitative result was small enough to reduce the priority of the issue for me such that I stopped paying nearly so close of attention.
And yet the evidence is to the contrary.
No it is not
Apparently Javier Milei of Argentina has eradicated his country’s deficit right in the 365 day mark since taking office. I couldn’t find the figures and how he did it? Does anyone know the details?
Milei's interview with Lex Fridman has some good info too. A big one was just cutting a ton of Argentina's welfare and a bunch of other government spending too
https://www.youtube.com/watch?v=8NLzc9kobDk
I’m not sure he eradicated the deficit as they will need to borrow more money soon, he HAS tamed inflation which was the biggest issue. Inflation in Argentina was crazy. You could watch prices climb as a slice of pizza would cost more with each visit. He cut spending from 44 to 32% of gdp which is a great way to put a spanner in the works. Note, the economy is stuck so now that inflation is stable, he needs to get things growing.
“He did so by pushing executive power to its limits with emergency decrees and vetoes, leveraging his direct connection with voters via a flurry of posts on social media to pressure lawmakers, and negotiating fiercely with Argentina’s 23 provincial governors after slashing their funding.”
That’s from this FT article in the big read section. https://on.ft.com/4iqDDAe
I am pretty sure he stopped inflation by creating a massive recession, which is probably the right way to do it, but had costs. He might have created a massive depression which is not the right way to do it. We will see.
Right. Volcker did that here in the early 80s, but I have my doubts the Argentine economy can grow as well as the American one to get them out of it.
Still, like Zhou Enlai never said to Kissinger, "It's too early to tell."
He never said it? I thought the issue was that he was answering a different question because of a bad translation.
>He might have created a massive depression
Doesn't seem likely. Q3 experienced 3.9% GDP growth, already ending the recession in Argentina. While 2024 as a whole is projected to have an economic contraction of about 3%, forecasters expect growth in 2025 of over 4% (with some projecting growth of over 5%).
And other economic indicators have been improving, as well. For example, job numbers are up and the poverty rate peaked from Jan-Jun of 2024 (after rising for several years), and has dropped each month since then.
Agreed. This isn’t victory parade territory - the patient remains in critical condition!
Is there an easy way to know if there's a follow-up of old SSC articles? I'm currently binging it and after sometime I'm curious if he still stands by those articles. The only place I see is /mistakes but it'll not have all the follow-ups especially if the old article isn't really mistaken.
I think he usually puts an edit somewhere on the post if he strongly changes his mind on a topic
...and in the Mistakes tab too.
I came across this recently by the late great Herbert Simon: “If men do not pour new wine into old bottles, they do something almost as bad: they invest old words with new meanings.”
Lately I notice people in public arguments trying to redefine words, presumably in the hopes that their redefinitions will catch on. They often argue they’ve improved some descriptor by replacing its naturally vague conveyance with fresh, precise meaning. Whatever their intent, they must notice how this retroactively fucks up every opposing argument that relied to some degree on the word.
I’ve nothing against stipulative definition or exploring a term’s meaning and see the obvious good in these things but I don’t see it in trying to overturn general usage based on its polysemy. Usage is what determines authority to a general audience. It springs up to fill a need. If you want to talk about something more specific, maybe coin a new term or add a qualifier (‘clinical’ depression, ‘aggravated’ assault).
You see redefinition a lot with conceptually loaded, fuzzy terms like:
Misinformation
Art
Freedom
Cancel Culture
Conservative
Liberal
Man
Woman
These words like all words grow naturally over time without some prescriber getting involved. At the very least, prescriptive redefinition usually just undermines the existing consensus on which shared language depends.
(Strangely enough, attempts to redefine a particular term seem to come in waves, as one redefinition is seized upon by the opposition as reason for a counter redefinition, leading to both parties denying linguistic reality.)
This is one of Scott's earliest essays, about motte-and-baileying people with "feminism is just saying women have a right to exist" and then saying "since you agree with feminism that means you agree with affirmative action."
https://slatestarcodex.com/2014/11/03/all-in-all-another-brick-in-the-motte/
I mean, it works, that's why they try to do it. Look at what they try to do with 'racism' (associating resistance to progressive demands with the opprobrium everyone attached to Jim Crow or the Holocaust), or 'misogyny'...
"So a fantasy about a knight who risks his life to save a princess from a dragon is misogynistic because..."
"It denies the woman agency and reifies traditional gender roles."
"So he should leave her to get eaten?"
"Well, no, he has a responsibility given his position in the power structure to use it for good."
"What if I made the knight female and the princess a prince?"
"Well, that's a step forward, but it's still heteronormative."
"Make them both women?"
"That's not your story to write. You're not a woman. It's not #ownvoices."
"OK, make them both men?"
"Unless you're gay, that's not your story to write either."
"I think you just don't want me to write anything."
"Of course, you didn't see the line saying 'looking for marginalized voices'?"
"Good point. I think I need to self-publish, I think the Gor series could use an update..."
"I think the Gor series could use an update"
Absolutely. You could have a fun, sexy series (if that's your cup of tea) by stripping out the WOMEN ARE NATURAL SLAVES AND WANT TO BE TAMED AND TORTURED guff. I don't care about John Norman's philosophy, it was bonkers. If you want sexy feisty slave girls who need to be tamed by the masterful hero (but maybe cool it on the whipping and branding and starving etc.), that's a standard romance novel trope, just set it IN SPACE and the job is Oxo. That'd be going back to the setting of the first couple of books in the series, which were more like that than the later ones.
Alternatively, you could just make it gay and drop all the women as love objects. His hero, Tarl Cabot, was a lot more interested in the big manly men of Gor as role models/hero worship/crushes anyway 😀
On the gripping hand, you could just dump all the sexual politics and update it to be a fast-paced planetary romance (no, not that kind of romance) because I'd like to know a lot more about the Mysterious Hidden Masters and just what the heck they are up to, for real, those Priest-Kings.
I will admit to having read a very late-period Gor novel when I saw it at a used bookstore and so learned that he was still self-publishing rather than having disappeared entirely when DAW(?) stopped publishing his books. Was curious as to what the series had turned into.
And the man seems to have mellowed in his old age, because he seemed to be trying to do, as you put it, "standard romance novel trope" but in the established Gor setting. Half of the book was written from the "slavegirl" point of view, and yeah, mostly standard bodice-ripper stuff. The other half from the "master" PoV, and he is baffled by the fact that he is actually falling in love with a woman in spite of all the cooties. The setting of course demands that she be a natural submissive and that all the social machinery of formalized sexual slavery be in place to define their relationship, but it sort of actually worked.
I would pay real money for him to have finished the series with his new power couple going on to take down the Priest-Kings and unravel the true mysteries of Gor, because like you I think that part of the setting was the most interesting and underutilized. But I don't think he went in that direction, alas.
Still, it was better than "Twilight" or "50 Shades of Grey", and that is something I would not have expected from John Norman.
How old were you when you encountered them? I was young enough to simply ignore the romance scenes in books, but in the Gor novels it simply wasn’t possible.
"Is Deiseach a kajira?"
"No, Deiseach is not a kajira. She serves the Christ of Earth."
The Anonymous Dude sat behind his keyboard, looking at the screen. He was not sure how to answer. Deiseach's true nature was still uncertain, but that was to be determined. He was not a Gorean, but his nature was not entirely dissimilar...
OK, I'll stop.
Those are fun ideas! I had a couple of my own: (1) Keep the slavery, but make it gender-neutral; anyone can enslave anyone. Bigger audience and you can get the femdom crowd. (2) Make it totally consensual and it's just kink in space. I figured out the smallest group that gives you every potential scene is four: two bisexual doms and subs of either gender.
Frankly with literotica and Archive of Our Own I doubt there's any money in this, even with the obligatory female pen name. I should probably just toss the whole idea and get into EA so I can try to save chickens or something.
I read the first 3 and you do find out what's going on with Priest-Kings. I won't spoil it if you want to go back and read them. There are 37, I'm told they turn into mostly porn after about 7, though there's a second alien race and so on. I am curious to finish The Most Politically Incorrect Series Ever, but 37 books is a lot. Supposedly the guy is still alive at 90 and writing these, I wonder if it's his kid or his wife or something.
> "No, Deiseach is not a kajira. She serves the Christ of Earth."
I don't know what a kajira is, nor am I familiar with the world of Gor, but I just wanted to say that this is a damn powerful line. Let's fund space travel so Deiseach can say this to an alien.
Given that it was just a parody of Norman's writing style, maybe I should write that book after all.
Here's what I was shooting for:
https://mindstalk.net/houseplants.html
The word "refute" is another example. It used to, and should still, mean "disprove", as of an allegation or claim. But sneaky politicians have taken to using it to mean only "deny", as if they hope to smuggle out the subliminal implication that their denial in itself constitutes disproof.
Even "Google define" now lists this alternative meaning. So it looks as if the corruption of the word is complete:
* prove that (someone) is wrong.
* deny or contradict (a statement or accusation).
I think there is a genuine confusion here where people use 'refute' when they should use the lesser-known 'rebut'.
What about "demure"? As in, "very mindful, very demure"?
Not to be confused with demur! Both words are close in meaning though because demure (unless you are thinking of a new sense I'm unaware of!) means reserved or modest, in a sense holding back from putting oneself forward, and the verb to demur (n. demurrer) is a legal word meaning to object to, or figuratively recoil or step back from an opponent's claim.
The new sense came from a MTF trans person on TikTok. Here is a Reddit thread about it. https://www.reddit.com/r/rant/comments/1f05lo2/very_demure_very_mindful/
The person who said it came across as neither mindful, nor demure.
Do you think language is changing faster now than in the past? Perhaps there has been more change over the past couple decades, but I don't really know. More broadly though, it seems like English has been pretty stable for the past 200 years. Other than a bunch of random capitalizations, English books from the 18th century aren't hard to read and understand. Go back another century and it gets much harder.
My sense is that sometime in the late 20th century, political operatives noticed the power that deliberate choice of language has over the national conversation (and thus the electorate) and ever since then we've been in a war over which words are used to describe things and what things are described by any given words.
Some examples are "anti abortion" versus "pro choice", or "undocumented immigrant" versus "illegal alien". Did we have this kind of thing 100 years ago? Maybe a little; "states rights" was used as a euphemism for slavery during the civil war. But I think this sort of thing really took off around the 80s.
> Do you think language is changing faster now than in the past?
Unlikely. Spell checkers, mandated schooling with "standards", rapid growth of writing
I disagree. Now someone can misuse a word, post that online, and it gets taken up and spread around and everyone is using it that way now, and if you try correcting it you are just rowing against the stream.
The Internet has made it much faster, much easier, and much *much* wider audience to reach with "literally means figuratively" than the past, where you had to write things down on paper like a caveman.
I think those newpapers standard guides will effect how English is taught to children for 300 years, with everyone seeing red underlines as objectively true errors. While Shakespear invented several words.
If I intentionally break grammer rules, write some historical long winded sentence that doesnt appear is the newspaper style guides, or maybe use repurposed punction like `,?` `!?` or `~`; im treated worse, maybe do it anyway because bring back anglish...
The corrective mechanism is computerized and objectified and nearly mandated, this is a major change from the 1800's where educated people had different opinions on how to spell words and dictionaries were still new ideas.
That’s a good question. As you suggest, grammar and spelling have remained largely consistent in English over the past 200 years, but it’d be interesting to see how a dictionary from 1824 might differ in its definitions of the same words from a dictionary today. And I do think the rate at which new words are being introduced has increased drastically due to technological dissemination.
Maybe grammar and spelling are changing less, though the timeframe may also be too short as of yet to tell. But word meanings seem to be changing faster. "Irregardless" is apparently now a word, and "literally" is literally its own antonym now. That doesn't even count slang such as "skibidi toilet".
"literally" is literally its own antonym now.
No, it's just wrong. The etymology is quite obvious. How can anybody ever learn if instead of telling them they're wrong we redefine the words to suit their utterances?
"The Sun orbits the Earth" so we make "Sun" an alternative name for the Moon? Really? If only Galileo had realised this he could have saved himself a lot of unpleasantness.
this is literally insightful
If it's not obvious, I agree with you. Websters, however, disagrees, because that is how people use it. I will not, however, use it to mean something like exaggeratedly false. But I will still understand what people mean when they use it that way. And cringe a bit, if only inside.
And the OE presumably takes the same stance, [except most of it hasn't been updated since 1903.]
I understand their postion: I also cringe when I hear it, and ditto lent/learnt used incorrectly. And I'm afraid I also judge, which of course is what it's all about.
Does anyone (1) use LLMs to generate Anki flashcards from PDFs/books/articles and (2) would be willing to share their prompt/preferred model provider with me?
I've been experimenting but am having a terrible time getting the chatbot to produce flashcards I *actually* find valuable (specific definitions, numbers, equations, etc.) as opposed to just quizzing me on anecdotes or asking me to basically complete sentences from the text.
Ok, I tried this in ChatGPT o1. If it's relevant, the pdf is 336 pages.
"I've attached the standard operating procedures for the airline I fly for as a first officer. Can you create a deck of Anki flashcards from chapter 2, as a CSV file, with the first column for the front of the card and the second column for the back of the card? To start with, just ten cards as a test please. The cards should be like typical flashcards, with a question about the material on the front, and a specific answer on the back, not just a reference for where to look up the answer in the manual. It is EXTREMELY important that the answers on the back are correct."
About 80-90% of the cards are great, and the rest hallucinate incorrect answers. I liked this one in particular, might cause a bit of a problem:
How should the thrust reversers be operated under cold conditions?
Cycle between forward and reverse at least once every 60 minutes to prevent freezing
As a fan of anki I am definitely interested in this. I might have a play around with it
Could you give a concrete example, @Tyler?
Of my desired output? It would take the form of a text file with cards separated by linebreaks, and questions and answers separated by semicolons, like so:
question;answer
question;answer
Perhaps with some so-called cloze questions as well, which look like
A {{c1::cloze question}} has multiple terms that can be {{c2::hidden}}.
I did this recently with a book on AI itself. A result I like is:
"How many connections (parameters) are in the human brain?;About 100 trillion"
A result I like less is:
"Why is calculating probabilities important in ML?;ML algorithms assume training data and test data are drawn from the same underlying probability distribution - if this assumption is violated, predictions may be unreliable"
I’ve been thinking about this a bit lately (don’t have a prompt/model provider pipeline set yet, sorry), but was thinking about feeding it bits of Quantum Country or some other Andy Matuschak content to do few-shot, then see how much I can remove while getting comparably-good cards to keep inference costs down.
fine-tuning a frontier model is also an option, if you have access to enough high-quality (pdf, anki cards from that pdf) pairs and have the $/want to turn it into a business
I wrote an essay about a problem that currently seems small, but if model costs continue to increase ($200 for o1 pro?), might be a big issue over the next few years.
https://www.brightmirror.co/p/the-problem-of-intelligence-inequality
I read the essay, and it seems fairly anodyne, with nothing controversial that jumps out at me. But it does not seem to address the hook question: "What will you be doing the day before AGI arrives?".
It seems to focus on the potential inequality gap growing, but the argument is not very coherent or persuasive.
Ironically, the assay would likely benefit from getting Claude critique it and helping with editing. As it is, it reads "too human".
I've been thinking about anti-Orwellian white pills.
Totalitarianism is inherently unstable, bc the totalitarian leader by definition attempts to amass 100% of his jurisdictions resources to himself (or immediate family) at the expense of everyone else, even his regime enforcers. Thus at the margin, the lowest ranking enforcer is only slightly better off or equal to the highest ranking slave civilian, so there is perpetual incentive for bribery and trade across that threshold. In fact the only way their economy avoids collapse is civilians engaging in black market, necessarily involving state agents. There can never be a perfect regime, they will always corrupt themselves, and inevitably get messy enough for revolution.
Your model only seem to work for African-style totalitarianism, imo, where the leader cares about wealth. It doesn't describe e.g. Stalin-era USSR at all: Stalin and the likes of him don't care for wealth all that much. Power is what they want. Money IS power, but they are canny enough to know what kind of money could become dangerous to their position, and ready to share wealth, up to a threshold.
The only real problem is that succession. USSR fell apart in no small part because just had NO succession model, and power was passed to the next generation haphazardly, with the next man in line often being almost-random, voted "least likely to bring back Party purges" by his peers.
But what is we mix a bit of monarchy with totalitarianism? If the chosen successor (and other children of top decision-makers) is properly groomed from the early childhood, it becomes no impossible for such state to survive for longer. It still will not be eternal, but it can endure for longer than seems possible from your model. I think this is a part of what is happening in North Korea: Kim dynasty IS a dynasty. While the family might live in far more opulent style than an ordinary Korean, or even enforces, they train their children to know not to antagonize military or secret service, or even common people too much. Compare and contrast to African and South American dictators, who usually end up deposed in the first generation.
Then again, we now live with a possibility of AI/algorithmic dictatorship. For an AI, wealth might not matter at all.
In 1984, the dictatorships were run by oligarchies, not single leaders. They were also highly technocratic and understood the science of maintaining power, so they probably would have avoided making the mistakes you describe.
when corona hysteria hit, I grew a spine a month into the "2 weeks", telling people I started refusing to wear a mask got me 3 verbal abuses online that I feel crossed the line; it was 5 seconds to stare down the one person who ever asked me to wear a mask irl a grand total of *once*
While the internet is the world culture and I wont ever discount it as irreverent, real life has far more constraints on peoples sanity, gun rights suggest a guerrilla war I dont believe anyone is willing to stomach any time soon; I will never be doing even two weeks again
you can find arguments from the ... grumpy ancaps, about the impossibility of enforcement if .01% of the population has a spine, cops being wildly outnumbered and all that; we need to more worry about that 1/3rd of people were informants to the secret police in the historical examples
Well yes, but democracy is also inherently unsustainable, so... Change is inevitable, entropy takes all, etc etc.
Though, I do think that a stable dictatorship is possible in theory. It just requires the entire population (including the leadership) to be in thrall to an unchanging concept. Religion is probably optimal, but other types of ideology can potentially work as well. Though at that point, I guess it doesn't really matter whether the government is technically a democracy or a dictatorship, since the mechanism through which it stays stable is by robbing *everyone* of agency...
Yep, dictatorships are pretty stable as long as the ruler is alive - there are many examples of that in the past and today. I think the tricky part is when the ruler dies and you need a change of government. In my country (Portugal), we had a dictator Salazar that ruled between 1933-1974 - a pretty long time. Old people like my grandparents still have fond memories of him. A few years ago, there was a big vote on public television for the greatest Portuguese, and Salazer won. What happened when he died in 1970 (he fell off a chair) was that the government pretended he was still alive for a long time, and it lasted for another 4 years. It would have lasted longer, but the military was fed up with the colonial wars in Africa, so they deposed the regime peacefully. Otherwise, it may have continued.
Some dictatorships have solved this "change of hands" problem by making them hereditary (monarchies), but still, history has plenty of cases when this becomes unstable. Democracies seem to be better at this - maybe. The USA has been going for 235 years now.
Wikipedia has a somewhat different version:
"After the incident [falling off a chair or in the bath], Salazar's life went on normally. Sixteen days later, Salazar admitted he felt sick and he was admitted to Hospital de São José two days later. On 16 September, he went into a coma.[172] With Salazar incapacitated, President Américo Tomás considered that the 79-year-old prime minister would die soon; on 25 September, he dismissed Salazar and replaced him with Marcelo Caetano.
Salazar lived for a further 23 months. After he emerged from over one month of coma[172] and unexpectedly recovered lucidity, his intimates did not tell him he had been removed from power, instead allowing him to "rule" in privacy until his death on 27 July 1970.[16]"
I agree with 10240. The masses don't have to be starving or even poor. It's not hard to imagine a future USA in which we have about the same level of wealth as now but with zero privacy, limited freedoms and zero political voice. No reason a legal market economy couldn't still exist. Look at China.
yes things can randomly walk in a worse direction for some duration. I am arguing against the dystopian idea of eternal steady state regimes conquering humanity. there will not be "a boot stamping on a human face forever".
Worth noting that even in 1984, O'Brian is canonically wrong there. The book ends with an essay on Newspeak written from an in-universe perspective describing the Ingsoc regime in the past tense. For some reason people usually ignore that and act like the story actually describes a regime that persists indefinitely.
The same boot stamping forever is different than a boot stamping forever. 😶🌫️
Judging by history, no form of government is overly stable, so I agree that totalitarianism isn't either.
> Totalitarianism is inherently unstable, bc the totalitarian leader by definition attempts to amass 100% of his jurisdictions resources to himself (or immediate family) at the expense of everyone else, even his regime enforcers.
I wouldn't include that in the definition of totalitarianism. I'd define it as total *control*; the enforcers may be better off than others in some ways without having more freedom.
but "control" costs resources, so attempting total control necessarily means attempting total control of resources.
the lowest rung enforcers are offered the bare min to secure loyalty, but there will always be variance in ambition, so inevitably some % of least paid regime employees will take the risk to profit from whatever the most crafty slave class members can accumulate on the sly, and agree to bend the rules. If its possible one instance of this can occur, then there's nothing stopping critical mass of "illegal" transactions occurring that might all be self interested, but eventually lead to downfall of the regime.
So I've now come around to the EA community with two separate proposals for world-changing science, seeking funding.
One is a totally novel, biologically and epidemiologically sound hypothesis on the origins of T1D that needs investigating. The other is a cure for hypercholesterolemia, the #1 killer of humans—the kinds of things you would think people would be interested in, if they're trying to do the most good with their money.
I damn near wrote a whole grant app around the T1D thing on Manifund, and stated my case on the heart disease thing succinctly last open thread. No bites! At least not from anyone with money.
At this point I can only conclude that the actual rich ppl among you are not actually interested in GoodMaxxing, but are mostly interested in forming polycules and discussing Yudkowski and pontificating about how Autocomplete will unify/destroy us all. I will be living in my mom's basement and funding this research from my own personal finances, and I will openly condemn your asses from the stage in Stockholm.
Thank you and goodnight.
Flouncing off in a huff because nobody rushed to throw money at you on the basis of "Trust me, I've got a killer idea that will cure everything! Yeah, I'm just some guy on the Internet, but so what?" is not going to win you points or get you investors.
Who are you? What are your bona fides? Why should I believe this is genuine and not an elaborate con where you run off to Antigua with the funding raised?
Before you start plotting your revenge fantasies of "I'll show them, I'll show them all!!!!", how about you show us *why* this isn't just another perpetual motion machine proposal?
See, these are precisely the kinds of question I was hoping for when I made the posts! The Huff & Flounce stemmed from the radio silence I got in response. I wasn't expecting "here's a million dollars" based on an open thread comment, just to start the conversation.
Several points that you might not be aware of: (a) biology, especially basic science-type biology, is difficult to evaluate on the importance-tractability-neglectedness framework used by EAs; (b) there's a lack of EAs that have the expertise or knowledge sufficient for evaluation of biological research; and (c) your admitted lack of biology expertise and unprofessional tone to your writing means that EAs can't simply provide funding based on your competence in the field (i.e., supporting the individual, even if they can't evaluate the project).
Your T1D hypothesis detailed in your substack post, specifically, isn't rigorously supported and your BLAST result isn't strong evidence due to mismatches, insertions, and incomplete coverage (I would also recommend checking the protein sequence against a T cell epitope:MHC predictor to check if it is likely to be presented (it would be much more compelling if HLA fit correlate with T1D incidence)). It's interesting and maybe worth looking further into (e.g., maybe check wastewater data and possible correlations with minimum water heater temperature laws/regulations), but there wasn't much that makes it dramatically more compelling than the very large number of hypotheses you can read in discussion sections of the primary biology literature.
Based on your overall vibe, I think you might be interested in https://hardea.org/
Thank you! This is definitely the subsect I'm looking for.
Even without ideological preconceptions getting in your way (which absolutely is a thing, don't get me wrong), grant writing is difficult and often frustrating and pointless. Even when the system's working right good ideas don't often get funded.
Though I admit the polycule/Autocomplete thing made me laugh out loud.
Is there a reason you haven't pursued traditional biomedical funding? Genuinely novel and promising hypotheses for diabetes, heart disease, and cancer are generally eminently fundable for reputable scientists via the NIH. Novel data for T1D origins or cholesterol treatment would be slam-dunk for a million dollar R01 grant. How exactly are you going to do biomedical research from your mom's basement on human disease, on personal funds? For all the trouble, you might as well just get a PhD and do the research that way. You can actually *get paid* to write grant proposals, to people who actually know enough to take them seriously! And who have giant coffers of money!
>How exactly are you going to do biomedical research from your mom's basement on human disease, on personal funds?
To be clear, I'm not doing the research in mom's basement. I'm doing the research at a bioscience incubator attached to a hospital in Indianapolis, because rent on a full lab suite in Indy is less than a one-bedroom in the Boston area.
My manufacturing inputs are essentially bouillon and offal, the molecular-mechanistic endpoint (fecal coprostanol/cholesterol) is a thin-layer chromatography assay that I can run for about $2 a pop, and the clinical endpoint is a lipid panel, which costs, what, $5 on the backend?
> Genuinely novel and promising hypotheses for diabetes, heart disease, and cancer are generally eminently fundable for reputable scientists via the NIH
I am not what you'd call "reputable scientists". I'll probably apply for an SBIR one of these days, but that is weeks of work now for *maybe* money in 9 months, and grad school is even more of a raw deal—four years of grunt work for the kind of Glinda that rises to the top of the heap in modern academia, in exchange for a "reputable scientist" certificate.
In those four years, a hundred million people will die of heart disease.
>I am not what you'd call "reputable scientists". I'll probably apply for an SBIR one of these days, but that is weeks of work now for *maybe* money in 9 months, and grad school is even more of a raw deal—four years of grunt work for the kind of Glinda that rises to the top of the heap in modern academia, in exchange for a "reputable scientist" certificate.
Putting it delicately...that kind of attitude will become an impediment to solving the problem you claim to care about. Virtually all biomedical advances are, at the very least, born from close collaboration with real, actual, reputable scientists (and usually developed by them), who really do care about the causes they work on. Even maverick heartthrob Lumina comes from traditional academic biology research, and the company has several advisors with traditional academic qualifications.
Please don't get me wrong here. I am a real, actual, wet-lab scientist. I learned anaerobic cultivation microbiology from some of the best in the world. I singlehandedly built out and operated an anaerobic lab at the Swiss Institute for Allergy and Asthma Research. That is where I discovered this antihypercholesterolemic bacterium. I am less maverick than Lumina. But I have no patience for the careerists who are more interested in publication count than in saving lives.
I sent you an email asking you if I could help with your heart disease thing and you didn't respond.
Bro what
https://imgur.com/a/vqLXsV7
LMAO okay yeah you're right there in the spambox between some predatory journals and a place I once bought my mom a giftcard for a massage.
I'll hit you back by email. I've been through the whole NIH grants rigamarole before—successfully, I might add—so I know it's a process and wouldn't expect anything less.
For now, I retract my denunciation!
You don't mean to tell me the Neurology Congress 2025, the Infection 2025 in Rome (Italy, even!) and the job offer aren't legit?
I'm shocked, shocked I tell you! If you can't trust random guys on the Internet offering you scientific and medical advice, then who can you believe? 😁
Deep in the comments of an open thread, a tantrum post actually fixed a missed connection. I want this to work now for reasons that have nothing to do with EA
Squeaky wheel and all that, eh?
Maybe it went to spam? Or - you're not the only person who's had this problem lately, maybe the system I use to send via my @slatestarcodex address is buggy.
I also think that just posting something on Manifund is not a very realistic way to get things done - people in your position have spent years trying to convince people, get venture funding, etc. I would like to help you short-circuit that process - hence the email - but if you don't believe it's a process and you publicly denounce people who don't give you everything you want immediately (or who did try to help but not in a way you noticed), then you're going to have a bad time.
For email, [have your tech expert] make sure that SPF, DKIM, and DMARC are properly set up - these allow recipient systems to see that your domain is cromulent. I had to do this for my email address at a custom domain (via Google Workspace) since I had set it up eons ago before those things existed.
Didn't expect to see someone posting here that I know from elsewhere. (We met ages ago at BoostCon.)
LLMs have read the internet, meaning they have read every out-of-copyright book on Google Books. Is there reason to believe they have also read most of the books still in copyright? If not, isn't that the obvious frontier for new quality data?
I don’t know but relatedly, HarperCollins just partnered with Microsoft to allow select nonfiction backlist titles to be used for training AI models. Authors can opt in to this agreement and get $2,500 per title for a three-year license.
There is reason to believe they've already read the copyrighted books, and the Feds are looking into them because of it. https://www.copyright.gov/ai/
Thanks
It's spelled "copyright"; it's the (exclusive) *right* to *copy*.
Thanks.
I’m genuinely scratching my head as to why an EA in 2024 would work on anything other than these 3 broad directions:
- PauseAI and other efforts to shut it all down
- Alignment work
- e/acc, YOLO to ASI
Why give money to malaria nets when the future of humanity is right there in front kr us?
AI apocalypse is a fantasy, while actual living and breathing humans can be helped by a malaria net. Helping a local dog shelter needing food and blankets is infinitely higher on my priority list than all three causes you listed combined. But then I lay no claim to be an "EA" so there's that...
>PauseAI and other efforts to shut it all down
This seems like an odd one. How would you imagine we stop everyone in the world from doing certain types of math on their computers?
Look into the successful efforts to stop child pornography. If working on AI carries a 25-year prison term, 99% of such efforts will immediately cease. Enthusiasts training models in their basements won't get us to AGI.
Are efforts to stop child pornography actually successful? Do you have a link to data showing the success metrics/outcomes? I'm not trolling, it's a genuine question.
Extremely successful but the lack of something is always hard to prove. But it’s pretty much non existent now on the clear web (used to be a big problem 20 years ago if you look at the news from that era) and dark web site operators are constantly arrested and have their websites shutdown.
And that’s just about sharing images which is six orders of magnitude easier than building AGI.
I'd like to agree because it'd mean world is a slightly better place at least on one axis, but I worry that your statement about difficulty of proving a lack carries quite a bit of weight. Is there truly less child porn or is it just less visible to normies? They constantly arrest drug dealers too...
Try to look for it online. You'll find *nothing*. Then try to find how to buy drugs online and it'll take you like 10 minutes to figure that out, tops.
The problem with narrowing your potential work down to a very small subset of careers is that it assumes that people are capable of operating at 100 percent of their capacity in any given task if they have the training when this is almost certainly untrue. Speaking personally, I don't really have the skills or personal interest in AI stuff, so even if I think it's important, I don't think I'd be effective enough at it for my contributions to actually mean anything serious.
This. A talented musician will almost certainly do more good in the world by working to develop a new musical style to uplift or even just entertain, than by learning to code and becoming the 8,375,312th person trying to solve the AI alignment problem. And of course an even faster path to human extinction than any plausible AI apocalypse, is for every farmer on Earth to give up their day job to become sub-mediocre AI researchers.
Do what you're good at, in a way that does the most good. AI alignment may be necessary but it is not sufficient; all the other stuff has to get done by *someone*, so if it's your thing, it might as well be you doing it.
If you are the talented musician, you may want to take a stab at a nice feisty protest song to kick off the Butlerian Jihad or whatever, but don't go overboard on that sort of thing.
I don't consider myself an EA, but I still give money to malaria nets. I just don't believe the singularity is that near.
It's increasingly clear the current technology will not (by itself) bring us to AGI (much less ASI), so there's nothing to shut down at this point.
It's not at all obvious what will, and therefore how to even begin aligning it.
On the other hand it's we should probably be scrutinizing AI companies a bigger amount than the current near-zero.
Because the tractability of any of this is extremely questionable. The 'safety' in a lot of AI safety has ended up being like the "Democratic people's republic" of the northern Korean peninsula. It's not obvious to me that any work done in this field is even directionally helpful, and meanwhile children are dying of trivially preventable diseases and there's a moral catastrophe in factory farming.
As a general rule, if a problem impacts rich people, I don't think it's a good candidate for EA work. Rich people have enough agency, children, the global poor, and animals don't.
I'm confused that you think "everyone dying" is a rich person only problem. The reason why people take AI alignment seriously isn't because they think it'll only affect them, but they think that *literally and not metaphorically everyone would die.
https://volokh.com/2008/07/29/world-to-end-women-minorities-hardest-hit/
I didn't say 'only'. This would, of course, impact everyone, in the same way that a nuclear war would. Given that rich people would be amongst those impact, however, we can expect the issue to have substantial resources dedicated to it relative to its importance and tractability.
But your argument applies at *all* levels of investment in any extinction level issue, including zero investment in it. ("Why aren't rich people trying to bring about the apocalypse? Because they would die! Therefore, it implies that the optimal level of apocalypse preparedness is to try and bring it about".)
I understand that you said that it's not clear if it's directionally helpful, but it's also not clear to me if that's an artifact of AI safety sucking or of low interest in the field in the first place.
Reasonable people can conclude that the confidence interval on the expected effect for any given AI-related intervention might include zero -- or negative values. For global health there's far higher confidence in the magnitude and direction of effects, and even in cases where there isn't ("oops, there's a new drug-resistant parasite now") are generally reversible and tractable ("let's try a new antiparasitic next year").
If a third of the EAs put their time and money into accelerating AI and another third put theirs into slowing or shutting it down, that will cancel out and we'll be where we started, which is maximally ineffective. They may as well coordinate to put it into a neutral cause they agree on, like malaria nets.
You seem Yud-pilled, and while most of the EA probably also is, they at least consider other possibilities. Humans are known to convince themselves of "truths" that turn out to be wrong. Definitely worth not focusing on one idea, no matter how scary.
Maxing "calculated EV" of all your actions is not always optimal if you might calc value wrong, or your EV error bars are high, ad they often are. Also think Kelly Criterion with regard to risk and reward.
One Utility now is worth more than one utility in the future, because it's guaranteed.
And once we get ASI why should an EA ASI waste 1 second saving anyone either. It should exterminate humanity to allow maximal exponential growth toward a utopia where it can resurrect humanity/eudamonia nanobots.
In theory any investment now into non exponential growth is a "waste". But it hedges our bets. If humanity is wiped out in 10 years, you still gave some people 10 years of life.
Note that if PauseAI succeeds, the world does not end and people will still need bednets, so the amount of charity money going into PauseAI should be under 100% (if not necessarily by much).
1. Some people don't have many AI related skills or aren't very interested in AI. You've got to do your job for your entire life, if you're miserable the whole time you'll burn out quickly, and if you're only mediocre at it you might be net negative.
2. For whatever you're working on, you can always come up with an argument that there's something even more speculative which is higher utility in expectation if you don't mind Pascal's Wager, different people have different tolerances for this sort of thing, and some people choose to stop before devoting their life to something about superintelligent AI.
3. It's not obvious that we as a society know how to make the AI transition better, and there are lots of ways that ham-fisted intervention could potentially make it worse. I think the least backfireable of the three things you mention is alignment research, but Eliezer thinks most alignment research is net-negative (because it's poorly done and crowds out better alignment research), and alignment is also the one that requires the most specific and rare talents to pursue.
Re: 1. You don't really need any skills for PauseAI, right? At the very basic level you just need to go chain yourself to the OpenAI/Anthropic office door and otherwise cause civil disobedience to attract attention. If 10 thousand people in SF started doing that on a daily basis to every major AI company, this might make their operations pretty inconvenient.
I mean, it deals with the skills component, but it's a really miserable way to actually live. A lot of people do that sort of stuff once, but only really hardcore zealots take up that type of civil disobedience over a long period of time because it functionally precludes stuff like having a normal career, family life, and friend network.
One big issue here is that "decadence" and "degeneracy" are pretty poorly-defined concepts that I'm generally suspicious of. If we try to work backwards from the general trope that "decadence is the thing that leads powerful empires to eventually decline and fall," we arrive at the conclusion that the proper definition of decadence is roughly "optimizing for winning power struggles within the state at the expense of weakening the state as a whole," which isn't really what people who use the term usually seem to mean.
How could anybody possibly answer those questions definitively? They're so weirdly specific. Like, what would the index of 1920s European decadence be?
A more serious attempt:
Per their 1925 census, a bit under 1% of the Weimar Republic was Jewish. Surely that in some sense limited their cultural contributions.
And, y'know, the fact that soon thereafter they tried to kill every single one of them I think also suggests a lack of influence.
Timothy McVeigh just had Terry Nichols. Not exactly a bunch of Klansmen
If colonists came to America, created wealth through technology and labor acting on the land, then shared that new wealth with the natives on a equal basis, then it would be a different story.
Instead wealth was shipped back to the old world and squandered by elites in the form of luxuries and senseless wars, and the natives got screwed over even more than the colonists, the vast majority of whom had a dismal experience as well, not including the plenty of those who were literally human trafficked and enslaved.
Tying this back to Christianity, to my naive understanding, literal communalism was the ideal, of course this was not a viable or sustainable model of political economy in the time of the early Christians or American Colonialism.
In fact at the time they were inventing new interpretations of Christianity to justify the current and emerging forms of exploitative political economy.
Christianity says you should let them do whatever even if they're not Christians, really. Turn the other cheek and all that. Virtue is not meant to be easy or fun.
Putting that aside, it is bad for a bunch of people to steal your land regardless of whether they are your co-ethnics or co-religionists, because stealing is wrong. If you are imagining an alternate universe where Anglo Christians all just showed up in the Americas, did not commit rampant violence, and peacefully purchased land under the laws of their nation, then why are you referring to Anglos and Amerindians? Perhaps the defining basis of that relationship - and why it is remembered so poorly - was constant perfidy and violence on the part of the Anglos.
A pity the original comment was deleted. I hope it's relevant to point out that even if the European colonists had nothing but peaceful intentions, there still would have been the issues of introduced diseases, and Native American tribal politics, which would've complicated relations.
Luke 6:34-35 New International Version (NIV)
And if you lend to those from whom you expect repayment, what credit is that to you? Even sinners lend to sinners, expecting to be repaid in full. But love your enemies, do good to them, and lend to them without expecting to get anything back. Then your reward will be great, and you will be children of the Most High, because he is kind to the ungrateful and wicked.
Wisdom. It would be so corrosive to ask the few friends or family member who have asked me for money to get it back.
I think you’re deconstructing an identity more so than is justified. Every identity is artificial, as it’s fundamentally a collection of human, not natural norms, and you can imagine a million scenarios where that identity doesn’t make much sense.
What the Babylon Bee is trying to do is probably define the American identity as broadly Christian. The American identity can be argued to have always been Christian, but it has previously excluded basically all denominations that weren’t Protestant. Catholics, Orthodox, Mormons, etc. have all been previously excluded from the American identity, and now they’re not, largely because you can’t really have a national identity consisting of a minority of persons. Protestants stopped being a majority in 2009 or so, but Christians (especially with large numbers of mostly Catholic Hispanic immigrants) are still a large majority.
It’s not a cynical ploy to try to construct, or identify a national identity, as arguably without one your country is liable to collapse. I don’t think it’s a coincidence that democracy generally doesn’t function well in the Middle East, where tribal identity often takes precedent over national, and Africa, where completely different linguistic and ethnic groups have been grouped together somewhat arbitrarily thanks to colonialism and few centralized states existing historically over much of the continent.
It seems to me that ANY kind of nation-state is struggling to stay solid in these regions where tribal identity comes first.