The stuff about due process and arbitration was really interesting.
It reminds me of disciplinary foul-ups in other unionized workplaces with similar rights. A LOT of time, some bad employee will skate by for a while because their direct boss is conflict-averse and doesn't want to fire or discipline them - but then it gets up to their boss's boss who yells at them to get rid of this guy, who then hastily fires them without following the rules. They can then unsurprisingly get their job back in arbitration, and the company then has to pay them to leave while taking a PR hit.
I have read a few of your articles so far, and while I appreciate them, I generally find them unconvincing (even though I think I would agree with you on a lot of the facts around policing). But so far, most of your articles seem to be around defense of the status quo in policing. I'm curious whether you a) think that policing needs any kinds of reforms and b) contingent on a, what those reforms would be.
I don't understand this. In the article "accountability starts at the top" Graham seems to be saying that a combination of bad incentives and lack of accountability produces police chiefs that are ineffective at reforming their departments and providing a higher standard of policing.
With this in mind, he seems heavily in favor of police reform, just starting from the top rather than the bottom. This seems like a sensible approach to me, and while you may disagree, I don't think we can accuse him of just wanting to preserve the status quo.
I agree that an article on his ideas for reform would be great though.
I read that one, and maybe I missed it, but it wasn't really about fixing things. It was merely pointing out that leadership bears most of the responsibility for problems. While I would probably agree that leadership is part of the problem, "firing leadership" or "holding police leadership more accountable" are not reforms (in the same way that _just_ firing bad cops isn't reform, _making it easier_ to fire bad cops is reform). That's just relying on hiring good people. Any system that _only_ works when good people are hired is a bad system and needs structural changes to make it resistant/resilient to bad people.
His reply to me does have some ideas for reform (that I had missed in looking over his articles). I think I agree with them on the whole. I'm not sure I agree that those are _enough_, but I think they are good ideas.
It seems like, right now, the "only one time" is rarely true. Maybe that's just the outside impression of only seeing the bad stories.
Would it be a good idea for cops to do something like the military does and have tours? IE, part of the problem is that cops get desensitized and start seeing the people that they have the most contact with as less-than-human, with predictable results, and a way to mitigate that is to get them a damn break?
I think it's probably more often true than the media portrays, but it is definitely not _always_ true. And the current systems, in at least some precincts in some areas, make it far too difficult to deal with problem cops. This is part of the problem with the current discussions around policing. All of the media is national. But policing is local. It doesn't really make sense to talk about policing on a national level, when there is not a single description of policing that applies everywhere. Graham's portrayal of policing is probably correct for the vast majority of precincts. And a lot of the reforms talked about in the media probably aren't that necessary in many of them (although I think _some_ of these reforms should be implemented everywhere, even where problems don't currently exist; power must come with oversight). But they absolutely _are_ necessary in some. I think, from the articles I've read of his, that he tends to focus on the majority of precincts that don't have major problems and ignores the (admittedly rare, relative to media coverage) problems because they are rare. I don't think that just because these things don't happen often means we don't need to change the systems to fix them (although I also think that changes need to be local and very few reforms, with the exception of national problems like QI and drug war reforms need to be done on the national level)
I don't expect perfection. Cops are humans, and will make mistakes. And I'm more willing than most to recognize mistakes that come from policing being genuinely difficult, and am willing to forgive those. If we expect perfection then we just won't have policing at all. But while I would agree with you that those are the kinds of mistakes that happen far more than the media portrays (and lots of things the media portrays as "bad policing" are more nuanced than that), they are not always the case. And cops who abuse their power _do_ exist and don't _always_ only get away with it once, and in some precincts, there are not good mechanisms to deal with those cops. And if the problem is the leadership, as you argue, then it isn't enough to just say "well we have crappy leaders", we need to reform it so that the system gets rid of crappy leaders. Furthermore, I don't believe that "bad leaders" is always the problem (certainly not the _only_ problem), I believe that in at least some precincts, the unions have managed to get in place procedures that make it overly difficult to punish bad cops. Every union in every precinct? Of course not. Most unions in most precincts? Almost certainly not. But in some? Definitely yes. I'll agree that this is part of the problem with the national conversation around policing. Policing is local and problems are local. But the fact that not all police precincts have these problems doesn't mean we should ignore them.
I have my doubts that either elected officials or police chiefs in large urban police departments have much capacity to change the culture of those departments. Let's take the Minneapolis PD as an example as we pass the one-year anniversary of the killing of George Floyd.
“Since 1980, every mayor, including me, has had a reform agenda for the Minneapolis police,” [RT Rybak, the former three-term mayor of Minneapolis] said. “None of us has made anywhere the change that is necessary.”
The Department of Justice has gotten involved at least two separate times over the past 20 years, including one consent decree. The current police chief was himself involved in a civil action about racial discrimination against officers which was settled in 2007 without going to trial. These top-down efforts coming from the brass seem like they are pushing on a string, which should be unsurprising because consciously modifying the culture of any large, bureaucratic organization is a tremendously difficult task. The transmission of that culture from vets to rookies doesn't seem to me to be very legible to the brass and the kinds of reforms they can implement. After all, Derek Chauvin was acting as a trainer for the other three cops on the scene where he killed George Floyd.
Does this imply that disbanding and reconstituting entire departments is sometimes the level of intervention that's required? That's a drastic step but these problems for both black cops and citizens have been tremendously recalcitrant in some cities. I'd be curious to hear about some success stories. I've heard good things about reforms in the Newark, NJ police department going back to Cory Booker's time as mayor there but have not dug into the details.
Fair enough. I do think that any reform will involve a better selection process for the people at the top, but agree there weren't many specific proposals.
I haven't read it yet, but I like Thomas Murphy's "Do The Math" blog and he wrote a textbook that I've heard is good: _Energy and Human Ambitions on a Finite Planet_ https://escholarship.org/uc/energy_ambitions
>Why did European countries develop parliaments much earlier than the rest of the world?
Honest question: did parliaments develop independently anywhere besides Europe? Because the pedantic answer seems to be "Because Europe invented them, and then the rest of the world copied the model after they saw it." But that doesn't answer the real question, which is why Europe in the first place?
Does anyone have a sense if the Nvidia GPUs and AMD CPUs that came out last year are finally available/not subject to crazy price gouging? At this age I just don't have it in me to keep too close of an eye on such things but I'd like to know if I can build a new PC anytime soon. Thanks.
They will. The bottleneck for manufacturing is fab capacity, and many countries have just started large expansion/construction projects for new fabs (to secure strategic supply). The capacity being built up is much higher than overall demand, so when those new fabs come online, there will be a glut and things will be cheap again. That will be in like 3-4 years.
From what I can tell, it will be another 6 months minimum. The demand doesn't look to decrease any time soon and increasing supply amounts to "build and staff another factory".
The prize for Cryptos is still higher than it was at the start of this year. So the demand from the miners will still be high. I don't see a chance for a relaxation on the GPU market as long as the BitCoin prize is above $20,000.
My techy friends have suggested checking the NewEgg shuffle everyday (https://www.newegg.com/product-shuffle) and there's a chance to queue for new GPUs when refreshed stock drops at Best Buy, Fry's etc. Apparently there's a discord for tracking that but I'm not that deep into the chase (yet)
At this point, buying a pre-assembled system might be the best idea, since those companies have long-term contracts, resulting in better prices and availability.
A substantial increase in production capacity seems to come in 2022, at the soonest, so unless demand falls a lot (which may happen if people stop gaming and start doing other things again), shortages may last for a while.
(hope it's ok to write about this again - Scott please delete if not)
I announced New Science (newscience.org) a week ago - a nonprofit the goal of which is to build new institutions of basic science, starting with life sciences.
The board of directors consists of me, Mark Lutter, and Adam Marblestone and we are advised by Tessa Alexanian, Tyler Cowen, Andrew Gelman, Channabasavaiah Gurumurthy, Konrad Kording, and Tony Kulesa.
If the site is exciting and *especially if you do biology*, I would love to talk to you -- alexey@newscience.org
I was excited to read about this initiative when you launched. I'm definitely dissatisfied with my experience in academic science so far (as a grad student), but I'm trying to challenge myself to go beyond this general negative feeling and isolate exactly what I would change.
Here's where I'm coming from. My strongest interests are clustered around concept and analogy formation during learning, in humans and maybe eventually in general/in AI. My strongest skills are engineering-related - software, hardware, signal processing. I entered a neuroscience PhD program because it seemed to be the best/only way to get to a place where I could learn about and work on these problems that interest me - despite the fact that my successes so far had mostly been in open source software development for the neuro community rather than the typical jobs of a scientist, designing experiments, interpreting the results and publishing. Now I'm struggling to develop these skills, and more precisely to justify spending a ton of effort on forcing myself to do things I'm naturally quite bad at when I feel like I could be happily coding away at my old research tech job or something similar, were it not a temporary stepping-stone to a PhD type thing. Or I could even see myself gradually getting to the point of being an independent researcher, but it seems like there should be a much longer period of apprenticeship/working on a team to learn how things really work before one is forced to go off in an independent direction and figure things out by reading papers and trial and error (despite the fact that other people around know how to do these things correctly and could just show you), as PhD students often are.
All this to say that I'm very enthusiastic about your stated goal of creating/expanding alternative paths for people who want to work in science but don't want to go the traditional grad student -> postdoc -> PI route. It seems to me like there should be more people with more specialized roles and more focused expertise working together on fewer problems with more impact. And the public funding system should somehow allow more slack for researchers while keeping standards very high overall (maybe again through division of labor). It sounds like these are things you want too, but I'd be interested to hear if you disagree.
If you want to reply individually, I filled out the form for the 2022 fellowship, so you can find my email there.
Thank you! Yeah, this only adds to my point about the absurd rigidity of the system. Some people started doing lab work when they were 16, some come from completely different backgrounds, but all are forced to go through the exact same path if they want to do research. I'll email you.
I'm a master's student (entering my second year) studying the evolutionary biology of aging! I don't know just what I'll have to contribute in conversation, but think the idea's interesting and admire some of the people involved + mentioned as influences (Andrew Gelman, Tyler Cowen, GWERN). My interest is piqued by the Summer Fellowship, albeit vaguely with no concrete plan of whether I'd apply or exactly what I'd do with it. I'll email!
One of my questions is: the site mentions that it's partly inspired by Cold Spring Harbor Laboratory; since CSHL already exists, in what ways is it different (or is it more of the same, since more is needed)?
The concept can be relevant for different forms of utilitarianism too. Consider rule utilitarianism:
If I run a red light, there's an element of luck that changes whether a traditional utilitarian would say if it's moral (if nothing bad happens) or if it's immoral (if you have a crash). A rule utilitarian would say that the rule is "don't run red lights" because the world where everyone follows the rule is better than the world where no one follows the rule, or even the world where it is inconsistently followed. Thus a rule utilitarian would say that the action of running a red light is equally immoral whether or not you have a crash.
That is all to say, the concept of moral luck is a significant gap between traditional utilitarianism and rule utilitarianism.
I might be wrong, but I don't think a non-rule utilitarian would have no objection to running a red light if it didn't result in an accident. The expected utility calculation would still tell you it is immoral to do it.
Not always! If you can clearly see and confirm that no one is coming from either the left or the right, the expected utility of barreling through the intersection is a bit higher than that of stopping (assuming equivalent emotional valence, making time the only remaining variable of interest).
It depends on how the form of utilitarianism is structured. Some utilitarians don’t follow an expected utility concept and instead judge after the action whether the action was moral based purely on the consequences.
It also depends on how the form of utilitarianism is structured. Some utilitarians don’t follow an expected utility concept and instead judge after the action whether the action was moral based purely on the consequences.
I don’t claim to speak for rationalists as a whole, but my position is that when evaluating how good or bad of a person someone is, the relevant thing to measure is the expected results, as those are what went into their decision making process. The actual results being different from the expected results doesn’t have any impact on their decision making process, so moral luck never plays a significant role. Of course, this does not extend to people deliberately sabotaging their own information.
There are additional layers that should be considered. For instance, due diligence on exploring the potential outcomes of decisions made. If I come up with a course of action and carry it out, thinking it will have a positive outcome, but it's obvious with minimal research that the outcome would be negative, that has to have some weight as well.
Hadn't heard of it, but I got this from wikipedia:
"Driver A, in a moment of inattention, runs a red light as a child is crossing the street. Driver A tries to avoid hitting the child but fails and the child dies. Driver B also runs a red light, but no one is crossing and only gets a traffic ticket.
If a bystander is asked to morally evaluate Drivers A and B, they may assign Driver A more moral blame than Driver B because Driver A's course of action resulted in a death. However, there are no differences in the controllable actions performed by Drivers A and B."
This seems to be mostly a measurement problem. Perhaps it's easy to miss a red light at a vacant intersection, but harder to miss a child in front of your car. If the child had been in the intersection in front of Driver B, would they still have run the red light and hit them?
How can you know that for sure?
In the hypothetical world where we could know the solution to this counterfactual, and also where we could reliably identify everyone in a given negligence class: yes, sure, we should punish all drivers in this negligence class equally.
But in the world we live in, we have to go by the metrics we have. It's inherently very difficult to identify everyone who's being negligent, so we assume there's a correlation between level of negligence and magnitude of bad outcome, and we assign punishment appropriately, and that's an acceptable approximation.
I think you're right about it being a measurement problem. If a driver goes through an empty intersection at 3am, we understand from seeing similar situations that the actual danger was minimal. We're willing to give the driver a pass on a "dangerous" mistake because the danger was not the same as doing it during the day when people may be walking through the intersection. By actually hitting the child, they have proven that they were maximally negligent (barring extenuating circumstances which would have to be evaluated case by case), because hitting a child while breaking a traffic law is pretty much the worst case scenario. Swerving to miss a child and hitting a dog is bad, but we would look at that situation and agree that swerving was justified to avoid hitting the child. Swerving isn't good driving behavior, but we prefer it to the negative outcome that would otherwise have happened.
We hold your driver A culpable because he did the worst driving mistake there is. It's such a bad mistake that we would have gladly accepted the intentional creation of multiple other driving mistakes in order to avoid that result. Driver B may have been just as culpable, but there are lots of reasons to doubt that. There is no doubt about Driver A.
There is a sense of moral luck I think is important, which is the idea that you shouldn't judge someone better because they were born at a time and place when their moral choices were easy. This is a corollary to "You are personally responsible for becoming more ethical than the society you grew up in": you are responsible for that, *not* for being perfectly moral; you get no credit for being moral because you went with the flow of your society, but *neither should you take the blame*. Or at minimum not all of it. Some people who were pretty damn immoral by modern standards, were nonetheless substantially more moral than their society, and we should restrain our condemnation of those people, and maybe even praise them.
My go-to example is Thomas Jefferson. TJ did some terrible things - he owned slaves and abused some of them well beyond the baseline, notably Sally Hemmings. But take a random modern liberal and exchange their birth circumstances - RML born as the eldest son of a Virginia planter in the 1740s, TJ in a modern liberal family - and looking at the outside view, RML would probably have done worse, because most planters did. (And if we stipulate that RML still was part of the Continental Congress and later politics, they almost certainly wouldn't have tried to get slavery condemned in the Declaration of Independence, nor tried to remove its economic incentives as President.) To the extent that swapping people like that is a coherent concept, of course, which is TBF not very far. This implies we should be relatively kind to Jefferson.
I don’t know what a rationalist stance on moral luck is exactly, but I’m quite doubtful that luck, including moral luck, is a coherent concept. You might be interested in my recent book on the topic, The Myth of Luck, here: https://www.bloomsbury.com/us/the-myth-of-luck-9781350149298/
I think of moral luck being related to having some modesty about one's model of the universe. You think some action is expected to have good or bad results, but you might not be completely right, so you make some allowance for actual outcomes.
I suppose that's how things work in the real world, rather than some ideal world where moral luck gets ignored.
I don't understand the distinction you are making. True, Glen dismisses AI risk (in the rationalist sense) and Scott doesn't. This is trivially true but the real disagreement seem to be several levels deeper. My takeaway from their conversation was that Scott was obviously correct everywhere and that I didn't understand Glen at all, but that's expected given my inference distance. The linked post doesn't clarify anything for me.
That was interesting to read, as someone who didn’t previously have much contact with Lanier or that group of people. One thing that struck me a bit oddly about the association that Weyl makes between cybernetic totalism and the rationalist movement was that a number of the points in Lanier’s half manifesto were things that the rationalist movement (to the extent that it is possible to generalize) is also against. The most prominent example was the stuff about evolutionary algorithms and the focus on complexity, which Eliezer vigorously opposed in the Sequences back in 2008-2009.
the point is that "complexity" is a measure of a solution, but not a blueprint for the solution. it tells you that your goal is far away, but it doesn't tell you which direction you need to go.
complexity is a cost you must pay, not something you should intentionally try to increase. saying that something is complex may be a true and useful statement, but saying "I have no idea where to begin, but I know it is complex, therefore let's do some random complex things and hope it works" only means that you have no idea and you are very unlikely to succeed.
(there is no way to edit an existing comment, right?)
even if you somehow know that the complexity of a task is 9000, it doesn't mean that a random project with complexity 9000 is likely to accomplish the task.
there are many people who would say something like: "well, we know that the complexity of the task is 9000, therefore the obvious first step is to construct a project with complexity at least 9000, and then we will use the project to do the task somehow" and it would sound meaningful to them, like they already solved the largest obstacle and the rest is merely technical details the less important people have to worry about.
but this is like being lost in a desert, knowing that the nearest oasis is 10 miles away from you, and saying "well, first I need to walk 10 miles in a random direction, and I will worry about choosing the right direction later". (only it is a desert in an eldritch non-Euclidean universe, where the number of different places with given distance from your starting position increases exponentially with the distance.)
You can edit your comment, there's a button at the top right corner of your comment that opens dropdown menu when pressed, one of the options is "edit". Not sure if there's any time limit on that though.
Thanks. That's rather disappointing, though, since Yudkovsky seems to be misreading what smart people mean by "complex systems" (in Russian, they are called "difficult systems"), which is some combination of non-linearity, sensitivity to initial conditions, feedback mechanisms, the generation of measurably complex phenomena from the interaction of many simple parts, and the study of scaling phenomena.
Here's an example of the kinds of things that people have in mind when they talk about "complexity"
More like "most people use the word [complexity|emergence] carelessly, and that probably includes you, if you are careless".
Does saying that something is complex/emergent allow you to make specific predictions? Or is it merely how you describe in hindsight the things that you already learned using different strategies?
(Thanks for the book recommendation, I may read it later, but of course this comment was written before reading it.)
evolutionary theory is based on some assumptions, such as "organisms make their copies which are not perfectly identical (otherwise we would get no new mutations), but are still very very similar to their parents (otherwise all traits would get so diluted in two or three generations that there would be no meaningful selection of them)". if you use evolution as a metaphor for something where the assumptions do not apply (e.g. nanodevices do not mutate, corporations do not copy), there is no reason to assume it would behave similarly.
not only evolution doesn't care about you, it doesn't care about your species either. the stories about how evolution sacrifices the weak individuals to keep the species strong are still too naive and too anthropomorphic.
evolution is utterly amoral, so if you ask it to solve your problems, get ready for a really horrific solution which you probably wouldn't predict, because unless you are literally a psychopath, your brain is full of self-censoring mechanisms. but evolution does not obey these limits (and neither will artificial intelligence, unless we explicitly program it to).
My summary on finishing: author influenced by both sides and presents a bit of the Lanier/Weyl point of view for a lesswrong audience. Evaluates Eliezer and Scott as not very much the sort of Bad Person who L/W are against; Ray Kurzweil is a better example.
I think that's a reasonable conclusion, though as I already knew much of the background I kind of regret the time I spent reading it. I was motivated to because Weyl's work seems admirable and Lanier seems smart, so it kind of bugs me that I feel repelled by their writing: so a version of their worldview from someone more congenial was worth a try.
I think that listing a bunch of names as Good People one should be on the side of against all the Bad People who have the opposite ideology of the manifesto writer is a poor way to think and that there's already more than enough focus on "Which side are you on?".
Also, many of the Good People listed (basically names in the "tools for thought" tradition) don't rub me the wrong way as Weyl and Lanier do, making me skeptical of them representing the major dimension of difference.
What stabilizes the population of large carnivorous mammals? Pre-humans, how did the population of wolves or polar bears or killer whales change with time? Does it follow predator-prey population cycles? If so, how large are the oscillations? Are the population growing slowly most of the time, to decrease during rare die-offs? If so, how rare are die-offs usually? What causes the stabilization? Is it that predators die of starvation when they can't find food due to competition? Or do they kill each other for territory before food supply can become an issue?
How does this generalize to the ancestral environment of humans?
My problem is that we don't have any data for large carnivorous mammals hat are older than say 100 years. And those population would not be in a pre-human state anyhow. So I guess I want something more speculative. Do we think that e.g. wolf population in North America in 30.000 BC followed the Lotka-Volterra equations? If so, was it lack of prey or competition for territory that drove population decline? Or did the wolf population in North America just grown steadily from the immigration of the first wolves until humans started to hunt them down?
I work in ecology, but not specifically on these types of problems, so I'm relatively qualified, but could potentially be missing nuances or "latest research". With those caveats out of the way, here is my understanding: This is one of those questions that we are unlikely to get very good answers to until human society progresses to the point where we can essentially remove our impact from the ecosystem (if you are a sci-fi fan, think the Nox from Stargate). The realization that we simply do not, and may never, know what ecosystems looked like pre-humans is a fact of the field. My ecosystem in particular is one that was completely re-engineered by humans probably close to a hundred years before _anyone_ was studying it in a more rigorous manner than "Here is what I saw there". We study how organisms interact in the ecosystem with barely an inkling of what the actual dynamics were in the evolutionary historical system.
So, all that being said, our best theoretical ideas are the above linked equations. Prey availability would be the largest determinant. Competition for prey resources could lead to territory disputes as you mention....but in the end it's all going to boil down to prey availability (competition for food is really the main reason why a concept of territory evolves in the first place). Barring some specific niche circumstances or else some new breakthrough in population dynamics theory (which is relatively well understood field)....the Lotka Volterra equations is what we've got. If for some reason you have extremely abundant prey that are able to reproduce faster than predators can eat them (which is only stable for the _very_ short term), then you may get to a high enough predator density that disease starts to strike. But that's sort of similar to a higher tier predator. Basically, long term trends in population dynamics entirely boil down to food availability from the bottom and predator constraints from the top.
Except of course in situations where climate or environmental changes cause issues for reproduction. But usually, on non-geological time scales, we can ignore those.
So it makes sense that prey is the limiting factor in the end. If prey is scarce, bigger terretories are needed. But is what does this "feel" like for a wolf (or an ancestral human) on the downward slope of the Lotka Volterra equations? Does the experience of the wolf become mostly "I can't find any rabbits no matter how hard I hunt" or does it become mostly "I must fight a lot to keep my hunting ground protected"? Is the extra mortality mostly from starvation or mosgly from conflict?
If you are willing to invest some time, perhaps you can dig into the example of wolves and moose on Isle Royal. This is one of the success stories of the Lotka Volterra equations. The data is for a national park in which hunting was forbidden for 50 years, so it's probably as close as you can come to "left alone by humans" as we can hope for. We have data for the number of kills per wolf pack and the size of the wolf packs. The number of kills per wolf does go down with scarcity and it can become really low (0.2 in the worst years, >1.5 in very good years), so I guess they are really hungry. But the authors also mention "aggressive interactions among groups". So probably it's a mix of both.
(But I don't think that wolves usually kill each other. I am no biologist, but I would assume that the bad thing that happens to a wolf is that you are driven away from prey by other wolves, and your offspring or even yourself starve. Might be different for other species.)
A few years ago, I’ve read an article claiming that oscillations predicted by Lotka-Volterra equations are observed to be in phase even in geographically separated regions (islands vs mainland). This suggests that there are more factors at play besides predator-prey dynamics.
Freddie's comment got me thinking, I can't help but see the synergy between crypto mining, machine learning, and computing generally as putting us in some kind of a paperclip maximizer problem, but we're maximizing compute wafers and integrated circuits. There appears to be no end in sight to how many revenue-generating things we could do with more compute. BTC in particular seems to be particularly problematic because it's just an arbitrary Molochian arms race to dedicating as much electricity and compute wafers towards the task as possible. Would BTC, if actually widely adopted, just be a new form of economic rent? Everyone merely holding onto BTC because someone will buy it from them later at a high price, and you need it because it's the reserve currency?
I should clarify, because everyone needs it but people can profit by just hodling (like land, because it's a fixed supply), wouldn't the same criticisms of Georgism apply to a BTC-denominated monetary system? Where the value of BTC can arbitrarily rise to just below the point at which doing any economic activity would no longer be worth it, and therefore it would become another rent-hole just like land?
And, if this is true, is this actually a bullish take on BTC because civilization is likely to just fall for it and deliver unto ourselves even more lackluster growth BUT pretty decent rents for the BTC hodlers?
I think people would just end up switching to BTC-backed notes, just like people historically switched from using metals directly to using metal-backed notes. So financial institutions might handle their clearing in BTC, but everyone else would be using BTC derivatives.
The explicit limit on the amount of BTC that can exist is analogous to there being a surprisingly small finite amount of gold on Earth.
The economic logic behind the stylized sequence of "X *is* currency" -> "currency is backed by X" -> "currency isn't backed" (i.e., X falls out of the system doesn't require X == gold.
BTC could remain a valuable scarce asset, but its prospects as a medium of exchange are dim.
I think my question would apply to gold as well. Does the gold standard also introduce a monetary layer which is exploited by no-value-add rent seekers? By George, anyone that needs to acquire land upon which to do their business is exploited by a rent-seeking class of land speculators. At small levels of business, this might be negligible (I could run a business through my home, for example, and consider my living expenses separate from my business expenses, and then my business is not being exploited by any land speculators.) But any business that grows large enough needs to acquire commercial/industrial real estate near a labor force and then is exploited by land speculators.
Similarly for gold, perhaps small businesses this is negligible, but if you're a large business you, at some point, needed to start concerning yourself with the medium of value-store for your cash balance. If you're under a gold standard, that means either directly or indirectly (via bank notes) acquiring literal gold. Is gold not like land because it's fungible?
Gold is not like land because it's not an input to production*, just a representation of value.
>you ... needed to start concerning yourself with the medium of value-store for your cash balance
Cash itself is a medium of value store. So are Treasuries, corporate bonds, demand deposits, CDs, accounts receivable, unsold inventory, or IOUs from friends & family. There are legitimate reasons to care about the liquidity of how you store value, but there's no requirement that gold per se ever comes into play even under a fully-backed gold standard.
*Unless you're making jewelry or precision electronics.
I don't think this makes any sense at all. Land matters because it's an actual factor of production, you need a place to put your factory, you need a place near where people live to put your store front etc. A store of value doesn't work like that.
I think I agree with you that I'm probably wrong but would you help me understand why?
It seems like, in practical terms, in order to do any production you need to acquire land, raw materials, talent, and other things, and in order to acquire those things you require *some* store of value, be it food, gold, dollars, or bitcoin, in order to barter/trade/transact with. Why is *some* store of value not also an "actual factor of production" if you practically do require it in order to do any production?
And if you grant that, then by removing one level of abstraction, we can say that for global trade it's basically required that you have US Dollars, simply because the US Navy enforces that everyone does their global trade in US Dollars. Does that not mean that US Dollars are an input to production for anybody doing global trade? Is it wrong to consider trade a part of production, or am I making some other analytic error?
Remember at the beginning of the Georgian post where he talks about how money isn't wealth, it's more like a pointer to wealth. Bitcoin is like that too
The Georgist analogy would be that mining bitcoin is like homesteading. A homesteading regime causes utility loss from people occupying land 'too soon' for the sake of title to it, but eventually all the land is claimed, just as eventually all the bitcoins are mined -- so there's a bounded amount of deadweight loss in the end. Once the bitcoin mining rate is very low, with the network paid for by transaction fees, will there be some other aspect that fits Georgism? I'm not seeing it -- should I?
To be clear, I may be the idiot here, it may not tie into Georgism at all. I do really like Georgism, but one thing I didn't like about it (which may spout from my ignorance) is this treatment of land as this unique thing. What are the specific properties of land that give it this rent-seeking economic power, and do any other things have the same properties? I'm kind of going out of my way to see if any other things in the economy also act like land ownership rights and drag down human economic productivity. Candidates in my mind are cryptos that don't provide useful auxiliary infrastructure (so BTC but not necessarily ETH), domain names (which I think satisfy it exactly, but they're just not as important because search engines act like an intelligent teleporter for the internet), and intellectual property rights.
For example imagine that computer scientists were just generally a lot less smart and no one figured out any good search algorithms like PageRank, just create some convoluted scenario to imagine that there were no search engines, but the internet was still really useful. Domain names would presumably be much more important, because people would probably try to discover new websites by typing in their name. In this world, you would expect returns on an ecommerce business to be slightly better than half of what they are in our world, because roughly half of the profit is wisely eaten up by the hodler of the domain name, who knows he can shut you down at any moment by choosing to stop renting to you and to rent to someone else with a similar ecommerce business. The users just want to buy dog food at pets dot com, they don't exactly care who is selling to them. (This thought experiment is a bit of a stretch but I think it demonstrates well-enough that other things, in theory, could be like land in the Georgian sense.)
I usually just assume that the supply of BTC is already fixed because 18/21ths of it has been. Having thought it through, though, I don't think it applies to cryptocurrencies because of the fungibility.
However, I do think this somewhat applies for intellectual property, and I think we see many of the same pernicious rent-seeking as we do with land. Intellectual property is not fungible, two patents are not interchangeable. We do have entire industries built around hodling intellectual property and charging people rent to use ideas, or suing people to prevent their use at all. And while new ideas can always be created, sometimes you need a particular solution to a problem and that might just have already been patented, and the patent hodler is in a position of extracting maximal rent from you.
I think intellectual property when overextended is another homesteading-like setup with the same problem of a 'land rush'. Nobody's going to write the same novel as you, at the level of granularity where a reasonably faithful translation is 'the same'; but modern copyright law goes quite a way further, and independent discovery of patents is routine. So I'd agree rent-seeking seems like a major problem in IP, but I'd lean more towards dialing back the overbroadness than to taxing it.
Well right, interestingly, taxing IP on some inherent, independently-assessed value doesn't seem to make any sense. If you think about the goal of LVT, though, it's just to make sure that landowners do not have incentives to speculatively trade land. So it would seem you want to apply the same idea to IP. I would go further and argue that we should be incentivized to share IP as far and wide as it is useful, just like a Georgian landowner is incentivized to make his land as economically productive as possible. Not sure what that looks like though.
The special property of land is that we can't make more of it. If I own a factory that turns out to be very valuable because of the things it makes, other people will build similar factories and competition will drive the price down. But if I own a piece of land that turns out to be very valuable because of its location, the amount of land in that location is fixed. Competition is limited.
There are other assets that produce "rent" in the economic sense. In New York City, you need a "medallion" to operate a taxi, and there are a limited number of them. This limits competition, so the extra money you make from the limited competition is "rent".
Bitcoin is a weird example because there's a limited supply, but it doesn't produce any income. Rent is a type of income.
As everyone easily remembers to point out, land has a fixed supply curve, that's special property A. Equally important to it's unique status is special property B, land is a "positional good". That is, an equivalent amount of land with identical physical properties down to the atomic scale might be valued entirely differently due to their relative position in space. The positionality is what let's the land speculators win- because the value of their land can increase greatly based _entirely_ off the actions of uninvolved and unrelated economic agents, without anything ever being added to the land in question.
The fixed supply curve just takes the problem and makes it a disaster. And the fact that land is an input to production takes the disaster and makes it relevant to everyone.
So BTC (and gold) isn't a Georgist good because it isn't positional and doesn't _have_ to be an input to production. So you don't care which gold you've got, and you may be able to do work without it.
And I think you could make a strong argument that BTC isn't _really_ fixed supply. Sure this protocol mathematically is, but I don't see why a new fork or just a new Bitcoin would be limited.
Yeah I think you've set me straight. What then, do you think of intellectual property? It isn't exactly a positional good, at least not in a literalist/physicalist sense, but intellectual property is quasi-positional in the sense that certain pieces of intellectual property are truly positioned somewhere in a graph of possible knowledge. Sometimes you may need to make a discovery only to find out that someone else already made it and now seeks to extract rent. This has always seemed imperfect to me. It is near-costless to share information as widely as possible; it seems like a coordination problem where everyone would be better off if some scheme of information sharing can be arranged. At the same time, treating intellectual property just like physical property that someone has to pay you to use seems to create a working incentive structure. Is some element of the intellectual property regime purely rent-seeking and holding us back, in your opinion?
Yes IP is definitely positional (position in time). Its not enough to develop an idea, you have to develop it _first_. And there is no doubt an element of our (and most) IP regimes that is purely rent seeking.
The difference to land is that the supply of IP isn't fixed- in fact that the whole scheme was devised as an attempt to incentivize the production of new useful IP. The ability to temporarily collect rent is given in exchange for finding new IP. So a georgist analysis probably isn't that applicable.
I think our current system is undoubtedly suboptimal but not necessarily fundamentally inefficient, I'd just have to see the data. In fact our biggest current check on IP rent seeking is that you are only granted this right temporarily, eventually the patents expire. As opposed to land, which currently you can hold in perpetuity. George would probably say that only being given temporary rights to extract rent from your land would work pretty well to remove the DWL and long term ills, but temporary land rights are a bad idea socio-economically, youre better off confiscating the rent.
I can't speculate much on practical policy changes I would make to our current IP system because thats so data heavy and I dont have it. The only thing I can say with reasonable certainty is that weve got a complete and total mismatch concerning the protection of arts and media, these IP are extendable way way too far.
The enormous energy use from proof-of-work is basically due to sticking with one algorithm and refusing to innovate no matter the consequences. Bitcoin is still using the first algorithm that worked. It's as if people were still flying replicas of the Wright Brother's airplane. It's not inherent to cryptocurrency, though, and hopefully competing algorithms will win out eventually.
I don't think the analogy is quite apt. For Bitcoin, the energy consumption is the goal: it can't be done any other way. This isn't necessarily a suboptimal algorithm, it is what it is. I'm not anti-BTC for energy usage purposes, I think that's a Malthusian perspective that we should avoid.
Distributed consensus is the goal. A zero-sum energy use competition is an unfortunate consequence of the first implementation chosen. I don’t know if they will work out, but proof-of-stake and Chia’s proof-of-storage look promising.
Has the rise of SpaceX & other low-cost space carriers finally made Project Thor realistic? The proposed weapons system involves dropping bus-sized tungsten rods from space onto targets- it would have most of the power of a nuclear strike, but without the fallout, radioactivity, and presumably extreme escalation of actually using a nuclear weapon. It seems like it would be particularly useful for hardened bunkers (read- Iran's underground labs, or North Korea's). Unlike a cruise missile, I don't really see how you'd defend against it.
The knock against Project Thor (first proposed I believe in the 50s) is that launching that much sheer weight into orbit was always far too expensive to make sense- especially with advances in cruise missiles and long-range bombers. But costs to orbit seem to have fallen a lot, not just due to SpaceX but other low-cost launchers like India. So- could the US, Russia or China just drop big rods of tungsten outta orbit now? I'm not really clear how 'steering' would work, or even if a guidance system could survive the orbital drop.
Another one thing I always wondered about Project Thor is- could they drop smaller chunks of tungsten, maybe to specifically target say warships? Like a scooter-sized or smaller one? This would require vastly more precision to hit a ship, and I'm not clear if a smaller rod would burn up in orbit. It would be a pretty extraordinary weapon, especially if a satellite launch system could drop say 20-50 precision smaller rods against a navy (like one blockading Taiwan, say)
> without [...] presumably extreme escalation of actually using a nuclear weapon.
I doubt that very much. If you dropped an RFG with nuclear-level yield, i.e. enough to create a mushrooom cloud, I expect that would be treated as an equal escalation to dropping another fission warhead.
And for various reasons, we have decided that the ability to drop that sort of munition from orbit is deeply destabilising. SALT II article IX 1(c) was put in place for a reason.
> it would have most of the power of a nuclear strike
No, it wouldn't. The kinetic energy of something in orbit is roughly comparable to the chemical energy of its weight in *conventional* explosives or rocket fuel: that's why we can get there in two stages (and one stage is within the theoretical feasibility). At that point, you can just drop a bomb on it. You can use a ballistic missile with a conventional warhead as well.
If you solve the targeting problem, RFGs are in fact a replacement for tactical nukes. They don't have the same overall energy output, but the energy density and penetration power is far higher than a conventional explosive. RFGs are bunker-busters well beyond normal warheads.
"At that point, you can just drop a bomb on it. You can use a ballistic missile with a conventional warhead as well" I can't speak as to the rods' kinetic potential, but it seems pretty clear that they'd be much faster to deploy than a cruise missile or bomber. It still takes longer than an hour for either of those to reach some targets on Earth
Note the words "bus-sized" and "tungsten". A bus-sized solid chunk of tungsten weighs 2 kilotons, so "packs its own weight in TNT" actually does put it on par with tactical nukes.
Indeed. I was merely pointing out the maths of it all.
The way I'd do this is to simply build a giant railgun or coilgun that can fire kiloton objects at the needed ICBM velocity, and hook it into the power grid. Blacking out a few million residences for a few hours during DEFCON 1 is not a huge deal, and this way you only pay for the KE when you need it rather than spending excess in a really inefficient way (rockets) and then needing to bleed off some to deorbit.
Unpleasant hypothesis, Possibly suitable for science fiction.
So it's really expensive to put bus-sized chunks of tungsten in orbit. What's the densest material that might be feasible to mine in space? Not that mining in space is cheap.
I mean, depends on what you mean by "feasible". The stuff that's cheapest to mine in space isn't the same as the stuff that'd become profitable first; the most profitable are probably gold, platinum-group metals (including osmium and iridium, the densest metals in existence) and tellurium, as these are depleted on Earth's surface compared to their abundance in the Solar System (they sank into the core during the formation of Earth).
To elaborate - the stuff that's cheapest to mine in space is normal stuff like iron, carbon, water, etc. It'd be worth mining those (and everything else) in space for consumption in space (as anything exported from Earth has a huge delta-V tax), but gold/platinum-group/tellurium is the only stuff plausibly worth importing to Earth. Of course, given that osmium/iridium are in the platinum group, these obviously are the answer to your question.
A Falcon 9 costs about $60 million and can put 22,000 kg in orbit. If we loft a rocket and fuel, and assume the rocket is about 92% fuel by mass (the Saturn V ratio), then that rocket can give about 11km/s delta V to about 1800 kg, for a total cost to change velocity of about $3/(kg*m/s).
Most objects near the Earth will be traveling relative to the Earth at roughly the Earth's orbital speed, which is 30 km/s. So the cost to match velocities (i.e. bring the stuff stationary with respect to the Earth) is about $90/gram. (We're ignoring a lot of factors of 50% or so here, such as the delta-V required to get the rocket to the rock in the first place, as well as clever schemes like gravity assists.) The current retail price of pure gold is $60/gram.
Low Earth orbit has an orbital speed of about 8 km/s, which implies a cost to bring stuff to the ground of about $24/gram, but you can cheat because you can aerobrake and let the atmosphere do most of the work. The only tricky part is that if you're deorbiting something as soft and low-melting as gold you would probably lose a lot of it to ablation on the way down. On the other hand, you would spread a fine mist of pure gold over everyone under your re-entry trajectory, which seems nice.
This is mostly speculation, but here are my thoughts:
We definitely have the capability to launch that much weight into orbit. [1] For SpaceX, this seems to cost about $100 million. [2] I just typed "cost to build a nuclear weapon" into Google and the cheapest ones seem to be a few million dollars [3], so it's still not competitive, although they often put multiple weapons on a single missile / bomber / submarine that is substantially more expensive.
The energy density should be similar to conventional explosives. [4] So dropping a bus sized conventional bomb releases a similar amount of energy. Yes, bus sized conventional bombs exist. [5]
The blast wave of an explosion tends to reflect off the ground. This is why hydrogen bomb test sites don't look like volcanoes, even though they release similar amounts of energy. [6] It can be quite difficult to get an explosion to influence things far underground. This is a potential niche for Project Thor. They also strike with a lot of downward directed momentum, which should allow it hit things farther underground.
To defend against it, you need to deflect it. Hit it with something of similar size going a similar speed (or smaller & faster or bigger & slower). This wouldn't be easy, but building a missile that can deflect one would be much easier than building a missile that can launch one.
Guidance is always really hard during reentry. Even on manned spaceflights, the astronauts just sit tight and there is radio blackout from the ball of plasma surrounding the object. No guidance makes deflection much easier.
Smaller versions might burn up on reentry, but they probably won't entirely. Fairly small meteorites make it to the ground. Guidance becomes even harder because the wind can push around small things more easily. The bigger problems is that it wouldn't hit the ground that fast. The atmosphere would slow it down to its terminal velocity before hitting the ground. For humans, you hit the ground at the same speed regardless of whether you jump out of a 10 story building or if you jump from space. [7] Larger and denser things have higher terminal velocity, so it takes longer for them to reach their terminal velocity. But there is no benefit for dropping small things from space compared to dropping them from an airplane. Really small versions were used during WWI from airplanes. [8]
" But there is no benefit for dropping small things from space compared to dropping them from an airplane. Really small versions were used during WWI from airplanes"
Yeah, that's probably fair. I was actually thinking about it more and the effective range of a 'plane' (that's a broad category) is pretty huge, I believe we could already enter low orbit just with manned planes by the 60s. The X37-B can seemingly go higher than that now, unmanned. So you could probably get most the benefit of Thor with smaller munitions & some type of plane, no need to actually use a rocket.
How do naval ships defend against munitions launched from very high up though? Just thinking out loud, like what's the standard defense for that now? Seems rather difficult to defend against, and a ship is not a fast moving target
You shoot it down. Weapons coming down from orbit are harder to hit, but not impossible. The US Navy's SM-3 is capable of hitting ballistic missiles.
Also, "don't get found" and "don't be there when the missile arrives" are valid defenses. While a ship isn't that fast compared to a plane or missile, it's pretty fast compared to *how long the missile takes to arrive.* A carrier sails at 30 knots. If your Thor weapon takes, say, 15 minutes to fall from orbit, then in the time between "carrier spotted!" and "impact," the target can move anywhere within about 8.6 miles. And you have to search that whole area of ocean with sensors small enough to fit on a missile.
Now, if you have someone constantly tracking the carrier and providing guidance, that's another story, but if you can get a plane or ship close enough to get a radar lock without getting shot, then they could just shoot a missile themselves instead of waiting for the rods from god.
You can get to space on some jets, but you can't get to orbit that way. Space =/= orbit; orbit involves not just the physical location of >100km, but titanic velocities of Mach ~25 (that's twenty-five, not two-point-five). A subsonic ramjet can't go past Mach ~5, and even a scramjet TTBOMK can't get to that speed; you need a rocket to get to orbital speed.
(You also need to circularise your orbit once you've thrust out of the atmosphere, as your initial "orbit" dips into the atmosphere where you exited. This requires thrusting in space, which by definition an air-breathing jet can't do. But this is moot because of the speed issue.)
Sure, but that's not what I was driving at (hash said "we could already enter low orbit just with manned planes by the 60s; that itself is bad terminology regardless of whether it relates to the original point).
(Also, a slight de-orbit will still have way, way more speed than something merely dropped from 150km; you don't need to lop off very much of the sideways speed to intersect the Earth.)
[4] is wrong. "Bus-sized" is volume, and while orbital speed is not immensely more than high-energy chemical reactions in terms of kJ/g (it's roughly 8x TNT; bipropellant rocket fuel tops out at 3-4x TNT), those high-energy chemicals have low densities (~1-2 g/cm^3) while tungsten has a density of 19.3 g/cm^3. You can fit a lot more grams of tungsten in a bus than you can grams of TNT.
What's the point of using on-orbit storage? If you want to drop a 500kg tungsten rod on someone, put it on an ICBM and fire it on a suborbital trajectory. It will get there in just about the same time, maybe less, since there's no guarantee your orbital depot is in the correct location in its orbit to hit your target -- you may have to wait 30-60 minutes for it to come around again.
I mean, there is the direction aspect, and if you have enough of them you can guarantee you'll always have some at the right point in orbit.
As noted above, though, I agree for pure kinetic impactors.
(On-orbit storage is militarily worthwhile for Casaba-Howitzers - they only need LoS and strike essentially instantly. Of course, that would contravene the Outer Space Treaty and be incredibly destabilising.)
Well, sure, you can put tons of satellites in orbit, but, again, what's the point? Your time to deorbit isn't likely to differ significantly from the time for a suborbital ballistic trajectory, which gets you all the advantages of a near vertical high-speed final impact, lets you launch at any time, from your own soil, and with as much oomph as you can pack into your rocket.
You probably get much better terminal trajectory characterisics, actually, since unless you're prepared to add a huge increment of delta-V to your orbital weapons -- massive fuel reserves required -- you're going to come in at a shallow angle, like all re-entering spacecraft, which gives you aiming problems as well as giving your opponent more warning and more opportunity to intercept your warhead.
I don't think a plasma weapon is very practical for space to ground use. Plasma is very high energy but low momentum, so the atmosphere should provide pretty effective defense -- the stuff will basically slow to stop as its modest momentum is absorbed, and then it will heat up a heckton of N2 to no purpose.
Huh. Well, I'm a priori skeptical just because the plasma generated by an ordinary (spherically-symmetric) nuclear blast doesn't punch through much atmosphere, about 150m for a modest 50kt device, 500m for a 1Mt monster. Maybe focussing it really gets you a factor of 500 but I guess I'd have to think about that for a while. I'm also wondering how you keep your plasma collimated to such an exquisite degree it doesn't spread out to near harmlessness on its 400km journey.
Now...on the other hand, if you were to use a small nuke to propel a solid projectile downward, by ablating off the near side of it and using the rocket effect...*that* would generate some really impressive vertical velocities...if we compare to what is achieved in the second stage of a thermonuclear weapon 1000 km/s seems not implausible. That means from, say, the orbit of the ISS you can hit anything in a 1000 mile radius from the point below your orbital weapon with less than 2 seconds travel time, which I think makes defense out of the question.
But you need 1 nuke per shot, which seems very expensive.
I feel I should point out that 500m of atmosphere at 150km is not an equal shield to 500m of atmosphere at sea level; using mean sea-level pressure as an estimator of the total weight of air in a column, and crunching the numbers with sea-level density and Earth's gravity, you wind up with the atmosphere being about 9km worth of sea-level-density air.
As I said, though, I'm not the master of fluid dynamics I'd need to be to give a definite, personally-understood Yes or No answer on the overarching question here.
It doesn't solve the latency problem, where if you deploy a weapon in Low Earth Orbit and need to attack a particular target on the Earth you're going to have to wait an average of six hours for it to be in position to strike, whereas the same weapon in a suborbital rocket on Earth can hit pretty much anywhere in half an hour. You'd need on the order of twenty orbital weapons platforms to match the responsiveness of one suborbital ballistic missile, and SpaceX isn't nearly cheap enough for that to be the winning solution.
It doesn't solve the targeting problem, where we don't have any good way to hit point and/or moving targets with a projectile moving Mach 25+ through the lower atmosphere because the plasma sheath surrounding the projectile will block just about any sort of targeting sensor or command signal.
It doesn't solve the problem where people who don't understand the performance or utility of nuclear weapons will mistake orbital kinetic weapons as nuclear weapons substitutes and completely misunderstand what they are capable of. In particular, your "smaller chunks of tungsten to specifically target e.g. warships" are actually *larger* chunks of tungsten; the original was to target tanks. Which is extremely difficult but maybe worth doing if you can pull it off; trying to build nuclear-weapons substitutes out of tungsten was never sensible and never seriously proposed.
It doesn't solve the problem that you are putting major strategic assets in plain view of your enemies, in space that you don't control and that they can easily attack with 1960s technology and maybe subvert beyond your ability to regain control with more modern cleverness.
It doesn't solve the problem that if your talk of "power of a nuclear strike" isn't just plain hyperbole then you've just violated well-established international law and a treaty your own nation has almost certainly signed and ratified, so the people who notice that they can shoot these things out of the sky will be on pretty solid ground doing so preemptively and in peacetime.
If you live on Earth and you somehow do find a good reason to attack people with chunks of tungsten moving at Mach 25, you will find it in every way better to do that using suborbital rockets deployed on large cross-country trucks parked in tunnels or caves somewhere in your nation's heartland, surrounded by your own country's soldiers and commanded by buried fiber optics, than to put them in orbiting satellites. "Project Thor" was a clever idea whose cleverness vanishes if you look at it closely; it's natural habitat is science-fiction stories and pop-science articles for an audience that isn't going to think about it too much. And maybe as a weapons system for people who live in outer space, but really Rule Zero for space warfare is "don't attack Earth from space, ever".
OK thanks. I checked with a couple of people I know who run a satellite tech company, they agreed with your characterization of the number of satellites needed for full coverage. (The one guy thought, conservatively, hundreds of them). He was a bit more optimistic that SpaceX currently has the guidance system needed for re-entry.
"If you live on Earth and you somehow do find a good reason to attack people with chunks of tungsten moving at Mach 25, you will find it in every way better to do that using suborbital rockets" I wonder if an ICBM carrying a shotgun spread of smaller rods could work against, say, Chinese ships blockading Taiwan? The ICBM fires and bursts open high in the atmosphere above the target, a few dozen or more tungsten rods with crude guidance systems rain down on the ships. Tough to defend against dozens of them all falling at the same time
Wait, why is rule 0 "don't attack Earth from space"? Attacking the ground of Earth from LEO is stupid, certainly, and an orbital habitat is almost the definition of indefensible, but I don't see why a(n industrially-established) polity of space colonies (lunar, asteroidal, or on icy moons in the outer system) would be helpless against Earth. Even if they lost the bombardment match - in which they'd have a substantial efficiency advantage for kinetics, and no "strategic surface-mounted energy weapons self-destruct from bloom" problem - it's highly nontrivial to bust bunkers on low-gravity objects due to the extreme depths to which one can burrow (Ceres is probably a bit too big to dig literally all the way to the core, but 50km seems pretty easy).
Good catch on the text of the Outer Space Treaty, though; I was under the mistaken impression that it only covered NBC.
>Wait, why is rule 0 "don't attack Earth from space"?
For roughly the same reason that Rule 0 for 20th century Japan was "don't attack the United States". It's populated by a really huge number of very industrious people who would very much prefer to ignore your entire part of the universe and shouldn't be too much trouble to keep placid but they'll react very very badly and in unison to external attack and there's really an awful lot of them and they're industrious enough to have already built a planet's worth of industry. And all of your imagined martial advantages will not compensate for oh fuck where did all those missiles come from and why didn't I declare my missile-counting variable an unsigned long?
You could imagine a future where this isn't true and offworld humanity outnumbers and outguns Earth and yet somehow considers Earth to be an enemy worth attacking, but basically nobody ever does that in science fiction and it's far enough in the future of any plausible reality as to be beyond the scope of this discussion. If you have to ask, Earth is too powerful for you to attack and you're just going to die horribly if you try it. If you think you don't have to ask, then you probably should have asked. Don't do it.
Leaving aside the sf hypotheticals, my impression was that orbital nuke platforms were never developed (and it was generally assumed in the 50s and early 60s they would be) because by the early 70s it was clear that national technical means were highly stabilizing, hence there was great value to all sides in *not* forcing an ASAT arms race -- which the deployment of offensive platforms would have inevitably done. Since the military benefits of on-orbit storage are a bit marginal, the command and control issues are more painful, and the cost exorbitant, I think the general consensus was...meh, let's build the MX instead.
It did come up again in a complicated way with SDI, of course, but that initiative morphed so quickly, and the fateful year of 1991 arrived so soon after, that I don't think it materially affected the thinking.
But of course all the real discussion happened in secret, so this is just a guess.
Basically means that spy satellites made it impossible to sneakily build up a powerful first-strike capability to defeat an unsuspecting enemy, so arms races were just a way to waste money and increase risk of accidents. Hence people started looking for negotiated solutions to the arms-race problem, and since orbital weapons platforms would open a whole new race track everybody agreed to not do that.
But I think the decision to not build orbital weapons platforms came well before that point. Really, I think there was never a decision *not* to build orbital weapons platforms because there was never really any serious consideration to building them in the first place. The idea was big among SF writers and futurists because drama, and the military studied it seriously because it was part of their domain and the contractors wrote up proposals in case we wanted to give them umpty-billion dollars to build the thing because that's one of the parts of greed. Then the generals came back from their studying and concluded that because of the latency and the vulnerability and the superiority of terrestrial alternatives this wasn't really a useful way of doing anything they wanted to do and it didn't look their adversaries were trying to do it either.
I'm not a UFO buff, but I was super-intrigued by the recent "60 Minutes" episode about mysterious flying objects. Afterwards, I started looking into it, and I discovered some fascinating analyses done by aviation expert Mick West. He argues that the videos probably have mundane explanations, like camera artefacts caused by triangular apertures (which makes out-of-focus objects look like pyramids) and illusions caused by parallax. For some of Mick West's debunkings, see this this video (there are also other videos like it):
Mick West's explanations seem reasonable to me, but I'm not really qualified to judge. I'm wondering if any rationalists here have looked into the UFO sightings and come to any conclusions. I also wonder if the US government might have an ulterior motive in encouraging people to (falsely?) believe in extraterrestrials visiting the earth.
I don't think it's aliens, but if it is I'd say their secrecy is good enough. We can't see them well enough to be sure what they are or figure out what they're doing.
The Navy and Pentagon people have figured it out and reached the same conclusions... The video Mick explains as a gimbal rotation is titled as "GIMBAL" by the Navy. The video he explains as a weather balloon is classified under "UAPs and Balloons" by the Navy.
As for why the government at the highest levels continues to let people believe they have confirmed these are aliens, when they have done nothing of the sort? I don't know. Why *would* they be okay with everyone suddenly more likely to think that strange things in the sky are aliens? Only the US government knows, I guess.
> If true, and they are so easy to photograph, why do all the photos look like smudges on a screen? In an era with so many high def cameras we should have good photos of them by now.
Smartphones can't actually take good photos of distant things. Try photographing a flying bird or a plane with one sometime.
> if secretly observing they are doing a really shitty job on the “secret” part.
Well, they could just be observing and not especially care if we see them or not. We don't care what animals think of us for example.
That article's "ZOMG it's ignored!" contention seems dubious. As the article itself points out, the US military doesn't publicise *everything* it knows, and this seems like the sort of stuff that would be classified due to concerns of egg-on-face and don't-let-them-know-we-know, so official disinterest doesn't mean much.
IMO you need more than camera artifacts to dismiss David Fravor's account. Now he's just one story, and it's sociological data, but I would appeal to his expertise. Navy pilots have four-year degrees, usually STEM, with perfect or better vision, and are highly capable of corroborating instrument data with their eyes. If you don't believe this is a smart guy, you should really listen to his interview on Lex Fridman where he goes all over what flying one of these things is like, his time in the Navy. When you listen to him explain his whole perspective, he seems too knowledgeable, humble, and skeptical to be mistaken, and afaict he's not making any money doing this. Instead it's probably pretty reputationally costly to be doing what he's doing. I'm far more compelled by his story than Bob Lazar.
Something else that I feel like the Mick West viewpoint seems to miss is that this is not just based on the video footage that has been released.
In the Nimitz incident, for instance, not only did several pilots see the same thing the video recorded, but it was also detected by radar from both the planes and the ship. This seems to make it a lot harder to explain away as a camera artifact.
This seems to be consistent with several of the other accounts that have come out - it’s not just grainy video footage from targeting cameras, it’s video footage plus visual observation plus radar.
(I do think he is likely right about the one from a couple weeks ago of the ‘pyramids’ over a ship, though - that really does look like a video artifact.)
?? Mick West argues persuasively that the TicTac video is a weather balloon or something equally innocuous, not a "video artifact". (The on-screen data shows that it's a ~6ft object travelling at wind speed at several thousand feet -- NOT large, close to the water, and extremely fast. Fravor continues to insist the video shows something unusual, IIRC.)
You're arguing against a strawman to say that we skeptics think they're all video artifacts. I recommend actually hearing Mick West out; he makes good videos! And he's interviewed three of the Nimitz eyewitnesses, if you're interested in a more back-and-forth dialogue.
I do think that if his explanation doesn't account for the radar tracking said object moving at extremely high rates of speed, it's not a great explanation.
Cmdr. David Fravor and Lt. Cmdr. Jim Slaight were on a routine training mission 100 miles out into the Pacific when the radio in each of their F/A-18F Super Hornets crackled: An operations officer aboard the U.S.S. Princeton, a Navy cruiser, wanted to know if they were carrying weapons.
“Two CATM-9s,” Commander Fravor replied, referring to dummy missiles that could not be fired. He had not been expecting any hostile exchanges off the coast of San Diego that November afternoon in 2004.
Commander Fravor, in a recent interview with The New York Times, recalled what happened next. Some of it is captured in a video made public by officials with a Pentagon program that investigated U.F.O.s.
“Well, we’ve got a real-world vector for you,” the radio operator said, according to Commander Fravor. For two weeks, the operator said, the Princeton had been tracking mysterious aircraft. The objects appeared suddenly at 80,000 feet, and then hurtled toward the sea, eventually stopping at 20,000 feet and hovering. Then they either dropped out of radar range or shot straight back up.
The radio operator instructed Commander Fravor and Commander Slaight, who has given a similar account, to investigate.
[...]
The two fighter jets then conferred with the operations officer on the Princeton and were told to head to a rendezvous point 60 miles away, called the cap point, in aviation parlance.
They were en route and closing in when the Princeton radioed again. Radar had again picked up the strange aircraft.
“Sir, you won’t believe it,” the radio operator said, “but that thing is at your cap point.”
“We were at least 40 miles away, and in less than a minute this thing was already at our cap point,” Commander Fravor, who has since retired from the Navy, said in the interview.
We don't have the radar data. I wish we did. We just have Commander Fravor's recollection of what it was.
Like I said elsewhere in this thread: the fact that the actual data we do have, the video of a different Tic-Tac incident during the same time, as unambiguously traveling at wind speed (as can be derived from the on-screen data), yet Commander Fravor is always in the media insisting that the *video* is *also* showing something unusual, is concerning to me! How am I supposed to trust his eyewitness recollections (and yes, that's all the radar data is right now, no one has the actual data) if he is constantly out in the media getting *that* part wrong?
Look, the Nimitz Tic-Tac *video* is unambiguously just a balloon on something else innocuous (the on-screen data shows that it's a ~6ft object travelling at wind speed at several thousand feet -- NOT large, close to the water, and extremely fast). So why do David Fravor and some of the others insist the video shows something mysterious and impossible? That calls into question their eyewitness testimony about the other TicTac sighting that occurred in the same place around the same time.
No worries, I just wasn't sure if I missed something since I've been reading about this. Frankly, I'm mystified at people who dismiss actual aviation experts, like the Navy fighter pilots who dispute some of Mick's analyses. No doubt Mick is right about some things, but the personal accounts that accord with infrared and radar frankly can't be accounted for so easily.
I think that there are lots of instances where military pilots and commercial pilots misinterpreted what they were seeing. Philip J. Klass presents examples of this in his book, "UFOs Explained" (published back in 1975). It's been a while since I read the book, so I don't remember the details. More recently, a military flight crew in Chile were mystified by something that turned out to be a commercial plane (probably). https://skepticalinquirer.org/newsletter/curated-crowdsourcing-in-ufo-investigations/
I'm looking for an SSC post in which Scott mentioned how some of his clients told him horror stories from their childhood that didn't seem to affect them as much as Scott would have expected. Something how even though they had been beaten to a pulp as a kid, the reason they'd gone to the psychiatrist was, I don't know, that they had trouble sleeping due to a loud neighbor. The gist of it was that people were more resilient than we think, or that people's expectations color their trauma. "Being beaten is normal, so it was not a monumental event in my life, so I don't count it as traumatic." Am I making this up? I can't seem to find the post.
> I remember, one of my first few months of internship, listening to a patient – not a PTSD patient or anything, just someone presenting with something totally different like bipolar disorder or drug addiction – explain the brutal abuse he suffered as a child. And the whole time, I was thinking “Oh god oh god this is the worst thing I’ve ever heard I want to go home and cry.”
> [...]
> Now if I get a patient like the one who told the child abuse story, I’ll be more likely to just put on my best Concerned Face and ask something like “How are you doing with that now?” And a surprising amount of the time, the patient will say “I’ve put that behind me, it’s not really an issue anymore.”
I think this was indeed what I was remembering, thank you! But, contrary to what I expected, the focus is not on how our priors influence whether or not something feels traumatic. If there is something like that, I'd like to read it.
Van Der Kolk, McFarlane and Weisaeth, Traumatic Stress:The effects of overwhelming experience on mind, body and society, Guilford press, 1996 might have information on what you’re looking for. Robert Scaer, The body bears the burden:trauma, dissociation and disease, also has interesting information.
You and your thirty friends have been imprisoned by a mad logician.
You are going to be placed in a room together, each wearing either a buttercup yellow hat or a hat or a cadmium yellow hat (I like yellow!)
Each of you will then be simultaneously and secretly given the option to guess which shade of hat you are wearing (although you are allowed to demur and not guess if you choose). If any of you guess correctly and no-one guesses incorrectly then you will be released, but if anyone guesses incorrectly, or if all thirty-one of you decline to guess, then you will be sentenced to spend the next twenty years either unable to lie or unable to tell the truth, guarding a door which may or may not have a goat behind it.
Once the trial has started you are forbidden from communicating, but you can agree a strategy in advance. What is the best chance of escape you can manage?
I'm probably missing something, but I don't see any logical way to deal with this. You simply don't have enough information. What could you do besides arbitrarily designating someone to make a wild guess?
100%, because you didn't forbid us to look at our own hats. Except less because I can't distinguish those two shades of yellow reliably in practice.
Otherwise, 50%, by designating one guesser. The hats aren't determined randomly or by a process with known logic, so seeing everyone else's hat is 0 evidence as to the color of your own.
No, you can do a lot better than that. Trying to turn the "you can't do better than 50%" intuition into a formal proof, and seeing where it fails, might give you an insight into how to do so?
Whatever strategy you choose, the mad logician can listen to and choose the pessimal distribution of hats. There is no order to exploit, and if there was it could be turned against you. 50% is the best you can do.
That is, are we allowed to use information about whether some specific other individual has guessed when making our guess?
If so I think the strategy is something like: everyone lines up from tallest to shortest. If the tallest person sees no buttercup hats, he guesses that his hat is buttercup. If the second-tallest person sees exactly one buttercup hat and the tallest person has not guessed, then he guesses his hat is buttercup. And so on.
I'd be curious about the mechanics of our being able to see each others' hats but being unable to communicate. Can I not wiggle my head a bit iff the designated guesser has a butternut yellow hat? What is the room like?
Does the mad logician have access to our pre-game discussion? For instance, if we designate one person as guesser, does she know which person it is?
Uniformly at random! Okay, that's tough - am I right to conclude that there's no relationship of any kind between what hat one person is wearing and what hat another is? Or might "at random" include cases where the logician assigns each person 75% cadmium, 25% buttercup? (If there's a possibility of the latter, then presumably one can do better than average by guessing the hat more commonly worn by others.)
Is a random distribution random? If the 30 other hats you see are all the same shade of yellow can't you statistically infer that the distribution is not likely random?
To be "scientific", you could first look at only 15 hats, hypothesize from that what the distribution likely is, then test your hypothesis on the 15 other hats.
Not true. If the distribution was random (which the problem doesn't guarantee), you could exploit assymetries in who does the guessing. For example, if someone sees that every hat but theirs matches, they guess that their hat is the other color. 1/(2^31) of the time this would generate 31 simultaneous incorrect guesses, but 31/(2^31) of the time it would generate one solitary correct guess. Every individual will make one incorrect guess for every correct guess, but the group makes 31 correct attempts for every incorrect attempt. You can use Shannon codes to generalize that to doing substantially better, though I don't remember how far that can stretch - definitely above 75%. **IF** it's random. Which, as stated, it isn't.
Tatterdemalion said hats were assigned "uniformly at random," in which case the logicians assigns buttercup hats at known but constant and independent probability p. Here's what I'm thinking in that case:
If we don't know what p is, then it seems that you should suspect your hat is the same as the majority of other hats. (Population sample as evidence of the base rate.)
If you DO know what p is, then you should guess buttercup when p>.5 and cadmium when p<.5, obviously.
If you know that p=.5 and you can't do any other sneaky communication then I'm stumped.
I suppose one class of solutions would build on priors about what the mad logician's hat assignment algorithm is. If the logician likes to assign hats according to some secret but deterministic criteria, then the guesser can use information about others to guess how she would be assigned. (For instance, if the guesser is a man and all the other men have buttercup and all the women have cadmium, he should guess buttercup.)
This doesn't apply if the game is adversarial (logician strategically assigns hats with the goal of maximizing you losing) and the logician can know your discussion, though. I think if I were a mad logician I'd like to give my victims a chance, though, and to have deterministic criteria, so you might just want to designate the person who can hold the longest boolean formula in their head as the guesser, so to speak.
No probability distribution over hat configurations is specified, but let's assume that the logician flips a coin when deciding what colour hat to put on each person's head. By agreeing on a strategy, my friends and I are accepting that there will be some configurations of hats where we lose, and some where we win. Call the winning configurations blue and the losing configurations red.
Consider a blue configuration. Blue means that we won, which means that at least one person has guessed their hat colour correctly. If we invert that person's hat colour, then the resulting configuration is red. Therefore, for every configuration that we can colour blue, there must be a red configuration one colour change away. Let's call configurations that differ only in the colour of a single hat "neighbours".
In fact, this property is the only one necessary to devise a valid strategy: If someone gives you a colouring of configurations where each blue configuration has at least one red neighbour, you can devise a strategy that lets you win all blue configurations in the colouring. Each prisoner will look at the the colours of all the other prisoner's hats. This will allow any given prisoner to reduce the possibilities to just two neighbouring configurations: One where her hat is buttercup, and one where it's cadmium. If both configurations are coloured red, then she declines to guess. If both are coloured blue, then she likewise declines to guess. If one is coloured red, and one is coloured blue, however, then she guesses that the blue option is true. (So if her hat would be cadmium in the red configuration, and buttercup in the blue configuration, then she guesses buttercup.)
If the hat configuration chosen by the logician happens to be blue, then all prisoners will either decline to guess (if flipping their hat colour would result in another blue configuration), or will guess correctly (if flipping their hat colour would result in a red configuration). And we're guaranteed that at least one prisoner will make a guess, since every blue configuration has at least one red neighbour.
The trick now is to maximize the chances of winning by maximizing the number of configurations that we can colour blue while still ensuring that each one has at least one red neighbour.
One simple and stupid trick we can do is consider the number of cadmium hats in the configuration. If it's divisible by 3, then we colour the configuration red. If it's not, then we can colour the configuration blue. For any given blue configuration, we can make it red in one flip by either adding another cadmium hat, or removing one so that the total becomes divisible by 3. This scheme colours 715827883 out of 2147483648 possibilities red, giving a success probability extremely close to 2/3.
It's probably possible to do much better, but this at least proves that it's possible to beat a success rate of 50%.
After reading JSK's comment above, I remembered about Hamming codes [1]. Hamming codes have codewords consisting of 2^n-1 bits, and have the property that any 1 bit error in transmitting a code word is correctable. Also, all possible bitstrings of length 2^n-1 are either codewords, or can be corrected to a codeword by flipping 1 bit. That means that if we colour all the codewords red, and all the non-codewords blue, each blue configuration will be neighbours with exactly one red configuration, which is optimal. In this case, we have 31 = 2^5-1, so we want to use the n=5 Hamming code. Each red codeword has 31 blue neighbours, so the chances of success are 31/32, which is 96.875%.
This is very interesting. I'm still working through how Hamming codes solve the puzzle. Are those in the room allowed to communicate in some implicit but not explicit way? Can they que up according to what they see others wearing? Can A signal to B which hat C is wearing?
I'm having trouble thinking it through because the allowed communication rules aren't clear. Obviously one can't simply signal to another what hat they are wearing, because then it would be trivial. But I sense the answer involves being allowed to signal to others what others are wearing, if it can be answered correctly 97% of the time. Or is no communication at all involved once hats are on?
OK, without exactly working out the math, I think I get it from phi's description. Tell me if I am wrong. We all have fantastic memories, and we all gather in a circle after being madly hatted.
Based on the exact configuration of the other hats, we either answer or don't. Pretty sure this means if I'm going to answer, a necessary but not sufficient condition is that I see that more than 15 of the hats are a different shade of yellow than I think mine is, but it is more complicated than that because it depends on the specifics of what every other hat in the circle is.
No, more than 15 can't be right as a necessary condition. The configuration has to be more complicated than that or we would get too many wrong answers.
Okay, I've got a strategy that wins 75% of the time. Consider the three person case: (rot13 spoilers)
Gur fgengrtl vf vs lbh frr zber pnqzvhz ungf, thrff ohggrephc, vs lbh frr zber ohggrephc ungf thrff pnqzvhz, bgurejvfr qba'g thrff.
Guvf fgengrtl jvaf 75% bs gur gvzr, fvapr vg ybfrf va rvgure OOO be PPP, ohg jvaf va gur bgure 6 pnfrf juvpu nyy unir n "gjb naq bar" frghc: gur crefba jvgu gur havdhr pbybe jvyy thrff pbeerpgyl, juvyr gur bguref jba'g thrff.
It seems like the success of this strategy varies depending on the number of people, using that strategy with 4 people would, I think, only give a 50% success rate, so in the 4 person case, it's better to just elect 3 people to be the designated potential guessers, and so you can reduce all cases above 3 to the three person case. (Maybe 3 people and 75% isn't the maximum, though)
Perngr n fbegrq dhrhr ol univat rnpu crefba rkrphgr guvf nytbevguz:
(1) jnvg n enaqbzvmrq nzbhag bs gvzr (gb nibvq zbivat ng gur fnzr gvzr nf fbzrbar ryfr).
(2) vs lbh'er gur svefg crefba gb zbir, fgnaq va gur zvqqyr bs gur ebbz, ryfr: pynvz lbhe cynpr orsber nyy gur pnqzvhzvnaf naq oruvaq nyy gur ohggrephcf. Dhrhr nyjnlf erznvaf fbegrq guvf jnl.
(3) vs lbh pbhag gjraglavar crbcyr va sebag bs lbh, thrff pnqzvhz, ryfr thrff abguvat. Jvaf hayrff nyy ungf ner ohggrephcf.
Fvapr V syvccrq n pbva gb fryrpg gur pnqzvhz/ohggrephc beqre, ybfvat bqqf ner orgjrra svsgl creprag naq svsgl creprag gb gur cbjre bs guvegl, qrcraqvat ba ubj rivy gur ybtvpvna vf
Are you meaning the covid vaccines or vaccines in general? For the latter there are decades of tests and very strict standards on what is allowed with medicine for children, even more so than for adults. So the risks are negligible.
For covid there is a very small chance that there is some kind of side effect as yet undetected that applies with kids but not adults. But I'm not aware of any particular theoretical reason to think there would be, other than the general extra precaution around medicine for children
Sorry, yeah, I meant the covid vaccines. I guess I'm interested in the question of how we know the chances of an as yet undetected side effect are very small. Perhaps relevant background: I'm not a vaccine skeptic in general, and I got the covid vaccine myself. But given that the chances of serious adverse outcomes of covid with kids are so low, it's not obvious to me that the vaccines are worth the risk. I will probably just go ahead and follow the CDC recommendation to get my kids vaccinated, but I'm just trying to do my due diligence, plus it would be good to have some arguments for convincing others.
We find that total people fully vaccinated to date in the US 18 and over is 128 million and the number 12 and older is 130 million, from which we infer that about 2 million people aged 12-18 have been fully vaccinated.
From the VAERS system, which gives you access to (anonymized) data on adverse post-vaccine events reported to the CDC, which you can get into with a bunch of clicks starting here:
...if I've searched the database correctly, I find 85 adverse events of any type reported to the CDC for children aged 6-17 with the Pfizer/BionNTech vaccine, the only one authorized for teens. Bear in mind "adverse event" can mean anything from a nasty headache to death, and can be reported by anybody, not just physicians -- you can dig into the database to get descriptions of the events, sort them by severity, et cetera, if you have the time.
From this one can reasonably surmise a full course of vaccination in teens results in an adverse event of any kind 1 out of every 24,000 times. The original Pfizer Phase III study results for adolescents were announced here:
They had 2260 adolescents in thheir data and reported no serious side effects, so that establishes a priori odds of less than 1 in 1,100 for consequential side-effects.
Note that vaccination is not generally recommended for adolescents for their own sake, but to break transmission between them and someone much more vulnerable (grandmother, 60yo neighbor with cancer, et cetera).
> But given that the chances of serious adverse outcomes of covid with kids are so low
Consider that if the virus can't gain a purchase on older populations, then adaptations that target younger populations will start appearing due to selection pressures. Ultimately, everyone should probably be vaccinated.
Of course, the same argument applies to vaccine protections, ie. perhaps adaptations that circumvent the protections will start appearing. It's an arm's race for sure.
> Consider that if the virus can't gain a purchase on older populations, then adaptations that target younger populations will start appearing due to selection pressures.
I'm not sure that follows.
Say there are 100 copies of the virus in a population, and 2 can spread better among kids at the cost of being worse at spreading among adults. Those 2 would be just as likely to spread among kids whether or not the adults were vaccinated. The virus doesn't "know" to give up on the adult strategy.
The lack of adults to provide breeding ground for mutations reduces the chance of finding that rare mutation that could spread among kids.
There's been million of cases and kids have not been good carriers for whatever reason. There have been over 100 million infections around the world and they still haven't found that mutation that would let it become worse in kids.
Just because there is a plausible mutation that something could evolve to be better doesn't mean that random search plus selection pressure is sufficient to find it. There are bottlenecks that need to be passed through.
> Say there are 100 copies of the virus in a population, and 2 can spread better among kids at the cost of being worse at spreading among adults. Those 2 would be just as likely to spread among kids whether or not the adults were vaccinated. The virus doesn't "know" to give up on the adult strategy.
You're forgetting that vaccinations change the adults' behaviour. Prior to being vaccinated, adults will be more careful to take precautions to limit spread, which limits the kids' exposure to possible mutations that may be better at infecting them.
After vaccination, some adults will relax those precautions and so those 2 of 100 will have more exposure to the populations for which they're better suited.
> Just because there is a plausible mutation that something could evolve to be better doesn't mean that random search plus selection pressure is sufficient to find it.
Right, that's why I said it's an arm's race. Investing more in arms isn't a guarantee you'll have superior military power, but it improves your odds. Analogously, investing in rapid vaccination won't guarantee we won't see adaptations that circumvent their protections, but it improves our odds of avoiding that outcome.
With my made up numbers, 2% of cases mutate into something that would be worse for kids. But we had 100million+ cases, and if 2% of mutations really did end up going into kids, I think we would've gotten it by now.
I still want people to get vaccinated to just close this out. And India still has lots of cases that could mutate into something that bypasses our current vaccines.
> given that the chances of serious adverse outcomes of covid with kids are so low
One should look at the issue systemically. Covid almost never kills kids, but without vaccinations it can spread rather freely among kids (especially after restrictions are lifted). From those kids, it could
(2) spread to the 5-10% of people in whom the vaccine was ineffective,
(3) mutate, possibly creating new strains that can evade vaccines. At the least, this would increase costs to society by creating a need for more vaccines & vaccine drives.
(here I've implicitly leaned in a way of thinking that is big/important to me: the "what if everybody did that?" way of thinking. Realistically lots of kids will be vaccinated so maybe we can get herd immunity without your kids' participation, yet I've imagined the scenario where all parents think "vaccine isn't worth the risk for kids" and all decide not to vaccinate.)
That's probably because the facts lead to a pretty one-sided decision. Vaccines are cheap, side effects are minor and/or rare, and benefits to both the child and society tend to be large. There is perhaps not a more effective, in terms of life-years saved, medical invention than vaccines. It's probably been nearly, if not more, effective than antibiotics and the ideas of sterilization for doctors. Maybe social hygiene (ie, clean water and dealing with waste) are the only things that have saved more lives.
if you are talking specifically about COVID-19, then I'll admit that the benefits are smaller (children are very unlikely to have very bad results from COVID), and, as a new vaccine technology, an argument could be made that the risks are less well defined than usual. However, from everything I've read on the topic, we understand the mechanisms behind the vaccine well enough (we got lucky that the pandemic hit when it did, being able to make the vaccines rests on 50 years of biomedical research....these are not new concepts) that, while not impossible, long term side effects are quite a bit more unlikely than the odds that your child gets severe COVID. Since the government is paying for it, this again seems like the pros outweigh the cons.
Yeah, sorry, I'm asking specifically about the covid vaccines. I would be interested in the background information/data/reasoning behind the thought that the risk of long term side effects is less than the risk of severe covid for a kid. That seems intuitively true to me, but I have no relevant training. My thought process is that the risk with the vaccines comes from the immune response to the spike proteins, but you also get an immune response to spike proteins from getting covid. So that's a tie. But when you get covid, well, you get covid: the infection itself might have or create some sort of adverse effects. So the vaccine is just a complete win-win. My problem is that I haven't seen anyone respectable give that argument, and if it was a good one then it would have entailed that there was no reason whatsoever to pause the J&J rollout (although maybe the mechanics of that vaccine are different--the "immune reaction to spike protein" thing applies to Pfizer, but I'm not sure what others).
"And considering the complete absence of long term understanding, combined with the very mild effects if at all on children and young people, I argue for not vaccinating them."
Where are you getting the long term understanding of children and young people getting covid? Children and young people getting covid could have devastating long term effects, we just don't know.
I think this is a little confused about how immune responses are generated in the first place. It's not that the immune system recognizes the intact virus while it's circulating in the blood or lymph or something. Immune cells need to be "activated" to respond to foreign particles. (It's extremely important to keep them in check so they don't destroy your own cells.)
The way that happens is that the virus infects a cell, the cell then "presents" pieces of the invader on its surface in various holder molecules which are more less painted bright red with yellow letters that say "WHAT I'M HOLDING IS EVIL." Immune cells detect these presented antigens because of the triggering holder molecules, and then generate a response against them wherever else they find them. So the antigens are naturally bits and pieces of the virus coat. Which bits and pieces? This we don't know, but the evidence suggests if they are bits and pieces of the spike protein, this works very well.
Also bear in mind that an mRNA vaccine is rather close in how it behaves to a natural infection. It subvers the cell's ribosomes to make a bunch of foreign proteins, which are (somehow) recognized as alien weirdos and presented on the surface to warn the immune system. The important difference is that the infected cell doesn't get destroyed, because the infection is inherently self-limiting -- as soon as the alien mRNA is degraded (a few minutes) it's over.
So a natural infection would probably generate a better immune response if there are some important immune-stimulating effects that happen late in the cycle of viral infection, when the cell is in much more distress. That's certainly possible, but it's also possible all the really useful stuff happens early, when the cell is still in good shape. I mean, that would be good engineering design, for what that's worth.
Well, we don't know, is the answer to that. We don't know enough about how cells choose antigens to display, or how the immune system is stimulated by them, to know whether having the full capsid to work with (but also laboring under the handicaps of a full infection) means the final result is better. Maybe yes, maybe no, or maybe it depends on the nature of the infection.
It's not an unreasonable instinct to think the answer is "yes" but it's just an instinct, which is a long way from empirical fact. You may be looking for more certainty than is possible with the current state of knowledge. My own feeling is that I'm agnostic on the point: I don't think we know enough to even make an educated guess. But one of the valuable side-effects of the whole pandemic panic is that people are now spending a lot of time and money looking into this, so maybe we *will* known by and by.
My understanding of why experts are not worried about long term side effects is that most of the technical difficulty in creating the vaccine was getting mRNA to last long enough in the body to elicit the desired response. Which means that the active ingredients break down _very_ quickly on their own. A long term impact would require some completely novel biological effects that we haven't observed so far. Possible, but very very unlikely.
I'm not sure I understand your reasoning behind the "tie". That immune response is the entire point of the vaccine: to get that response _without_ having to actually get covid. So I don't understand why that counts as a tie? Other than the sense that yes, getting COVID _also_ gives you that immunity, but that comes with the risk (albeit low for children) of severe COVID, which the vaccine does not.
I think we agree: I should have said "that's a tie so far". Both generate an immune response to spike proteins, with the good and (currently unknown but possible) bad effects that comes with. But there could also be currently unknown bad effects of the covid infection itself. So the overall risk of currently unknown bad effects is greater for getting covid than for getting vaccinated. That "seems right" to me, but I'd be interested in an expert opinion.
Empirically speaking, the numbers I've seen suggest that the adenovirus vaccines give about 70% immunity, actual infection by SARC-CoV-2 gives about 80% immunity, and mRNA vaccines give about 95% immunity.
I have no idea whether anyone knows *why* the different methods of immune training give different levels of immunity, but empirically it seems to be the case, and that difference does seem like it would matter.
None of these studies I'm talking about are measuring immune response. They are looking at how many people with one type of protection end up with new infections, compared to a control population without that type of protection.
I think there are fewer of these, particularly with matched controls, than there are studies of the vaccines themselves, so we have less specific information about the protection provided by the virus itself. But what information we do have suggests that it is at least not stronger than the protection given by the two dose mRNA vaccine protocols.
> the risk with the vaccines comes from the immune response to the spike proteins, but you also get an immune response to spike proteins from getting covid. So that's a tie
This doesn't quite seem to be right. If you get vaccinated, then you get the spike proteins, whereas your alternative to getting vaccinated isn't a deterministic covid infection, but rather the chance of a future covid infection. Depending on the future of the pandemic (both this year, and in future years), one might have different estimates about how likely that is. But unless one thinks it is well above 80% likely, I could see someone legitimately thinking that vaccination is less bad than infection, and still preferring not to get vaccinated (particularly if they think they have good ability to continue other protective strategies for the rest of their life).
I think that's really overstating both cases. The lasting immunity is to *that particular strain*. We don't know to what extent it protects you against variants, or more precisely how "varied" they have to be to evade the original immunity. This is similar to the fact that new influenzas emerge every year and your immunity to last year's version doesn't do you much good with this year's.
Same thing with the mRNA vaccine. The only reason people think you might need booster shots is to protect you against variants that can evade the original immunity, because they are sufficiently different.
It *may* be the natural immunity gives you broader protection (against more variants), but this is something that is still not known and is being actively investigated.
There's a whole lot of theory going on in this set of inferences, and no empirics. We know that a lot of things that happen in the real world with infectious diseases defy any particular theory, and that's why empirics are so important. (Otherwise we would have just started vaccinating people in March of last year.)
So first off, let me say, it seems pretty safe. But I’m going to throw a few points that at least could be brought up in a serious argument.
First off, in adults at least there appears to be a risk of a bad immune response. Some of the vaccines appear to be linked to people going into anaphylactic shock or blood clotting (the shock is, from my understanding, why you are recommended to wait 15 minutes after the vaccine). From everything I’ve read, while both of these can be bad/fatal, if you monitor symptoms they’re easily treatable. Furthermore, people are screened for highest risk of a reaction, so it seems like pretty low odds of something actually bad happening if you wait the 15 minutes and overreact to any of the symptoms of blood clotting. To note (from my non-professional understanding), these severe reactions are not reactions to the “COVID” in the vaccine, rather they are reactions to the chemicals used in the fluids and stuff.
Now then, we can get a label for the Pfizer vaccine to see the odds of your kids being miserable:
In a clinical study, adverse reactions in adolescents 12 through 15 years of age included pain at the injection site (90.5%), fatigue (77.5%), headache (75.5%), chills (49.2%), muscle pain (42.2%), fever (24.3%), joint pain (20.2%), injection site swelling (9.2%), injection site redness (8.6%), lymphadenopathy (0.8%), and nausea (0.4%).
Obviously these don’t seem fun to deal with for your kids. I’d sort of see this first as like a flu shot; they have odds of being sick for a day or two after their vaccine, but also odds of being sick from covid. But the odds of them being sick from the vaccine are real, and while I don’t think it outweighs societal benefits, it’s still worth taking into your calculus.
Finally, I don’t have the numbers on this, but my understanding is that children have low odds of both contracting COVID seriously, and of passing on a COVID infection, so society benefits less from your kids getting vaccinated than from you getting vaccinated.
These are the only three arguments I can imagine against it. For the sake of posting things on the internet, a disclaimer: This is meant as info for someone who is well informed about the benefits and risks of the vaccine and doing due diligence, and not to discourage them or there children from getting vaccinated.
I think this would be exactly the type of website that you would want to use, it is dedicated to providing neutral information to help you make your own decision. Except that it is in German. If your browser has a translation tool, it might still be helpful.
For the "average" person under the age of 18, the cost/benefit result is that it's *probably* not worth it, but it's more complicated than that.
In the aggregate, the fatality risks of Covid-19 decrease smoothly along with age. There's no clear-cut point at which mortality or morbidity substantially jump. So the direct benefits of the vaccine drop as the person gets younger. Likewise, it looks like children are also less likely to spread the disease. This means that the indirect benefits are also lower.
However, the harms of vaccination (injection-site soreness, feeling terrible for a day or so) are pretty much constant. At some point there has to be a cross-over point for the harm/benefits curves. My best guess is that for the typical person, it's somewhere between 15-18.
But individuals are not aggregates, and individual values matter. A person who's healthy, doesn't interact with people who are at-risk, and has a major fear of needles might be better-off waiting. Someone the exact same age who has a few health conditions, really wants to be able to visit their grandparents undergoing chemotherapy, and has continual nightmares about dying from Covid might be better off getting vaccinated.
The best thing to do are to evaluate the life circumstances of your child, their particular interest and values, and contact your child's pediatrician for professional advice.
I'm curious to know what others think about the reporting on Bill Gates following his divorce. Here's one from Vox: "Bill Gates Will Never Be the Same" - https://www.vox.com/recode/22441627/bill-gates-scandal-divorce-epstein and an earlier article from the New York Times: "Long Before Divorce, Bill Gates Had a Reputation for Questionable Behavior" - https://www.nytimes.com/2021/05/16/business/bill-melinda-gates-divorce-epstein.html. That loaded word "questionable" implies bad judgement on his part and a call for a larger social condemnation by these journalists, who insinuate that there's more to the story. I say it seems gossipy, prudish and opportunistic.
If somebody hangs out with Epstein that seems newsworthy (if not particularly surprising from someone of Gates' echelon) and if he does I'm not sure of a more polite word for it than "questionable."
If the mayor has dinners with the local mob boss, I would be in favor of newspapers opportunistically publishing the story - that seems to me the absolute baseline of the kind of thing you'd want newspapers to do. If the paper didn't have anything on the mayor's relationship *other* than those dinners, or if the dinners themselves were from disputed sources, or whatever, expect a lot of cagey language about "questionable." But the implication isn't just an artifact of the language.
Depends on whether you see Epstein is "rich pedo guy" or "honeypot CIA plant designed to entrap wealthy people with blackmail material" or some combination of both.
I suspect they're about as reliable as the NYT piece on Scott - Bill Gates is exactly the sort of person nyt/vox would have it in for and go out of their way to paint in an unflattering light. Whatever your priors were on Bill Gates, you should probably just stick to them (or even slightly improve your image of him, since if there were any real smoking guns they probably would have reported those instead).
> since if there were any real smoking guns they probably would have reported those instead
Alternatively, given that he's one of the richest men in the world and therefore has access to a bear infinite supply of lawyers a reluctance to say anything concrete may be more out of fear of being sued, or otherwise facing negative consequences. There is a big gap from things that journalists are confident of on the balance of probabilities, and things that they can prove in court, particularly in cases where witnesses would be reluctant to talk.
Right. It's weak evidence - there's a lot of potential reasons the media wouldn't report a smoking gun and only one of them is "it doesn't exist" - but this non-reporting is still more likely in a world where he did nothing wrong than in a world where he did, so I'd say overall the media reports should be (weak) arguments in his favour (but mostly just stick to your priors).
I think nobody will really care in about six months, especially if he makes a pretty aggressive PR push and effort to get some new venture out there. Parts of the media might, but that's because a lot of online media has a leftist bent and dislikes billionaires on principle.
I read those articles, and I didn't think either one was a "call for a larger social condemnation". They seemed a lot like the article about Scott - it uses words like "questionable" but is overall not really negative.
Scott, any chance of something like the old classified threads making a comeback? I have a niche job posting I'd like to list, but I don't know if that would be allowed in these regular open threads.
since the main optimizer behind human achievement isn't individual competence but the evolution of behaviors we call culture, the ability to discern and harvest the most useful memes of today's multibillion megaculture is in itself invaluable.
rationality is mostly useful not for creating novel solutions (leave that to evolutions, idiot), but because it offers some good heuristics for selecting existing ones.
I gotta think about this. I’ve had the same first thought, I tend more toward: “Meme conversations are a very dense ways of communicating and hence valuable. There’s important meta information in memes and people who understand may have an advantage.” I think that signs with what you state.
The next part I don’t follow so well: Rationality doesn’t create solutions it just best aggregates and selects them?
the connection may be that I think knowing whether a meme is useful is *hard*.
There's so many supposedly useful memes - more than you can ever absorb - and a huge portion of LessWrong style memes is basically tools for identifying BS, or for deciding when versus when not to rely on established solutions / social consensus:
Chesterton's Fence, Dissolving Questions, the Algernon Argument, Absurdity Bias ( / most of the biases), Mysterious Answers, The Toxoplasma Of Rage, Change Is Bad, Simulacrum Levels, etc
I'm curious about your model, what do you mean by memetic meta information?
We might be talking about different things. You're talking about more formal memetic ideas distilled into shorthand, I think. I’m talking about the shorthand.
I really like internet memes, not because the existing set are intelligent or deep but because they are dense packets of information. Take “Motte and Bailey” for instance. Many people have a tough time recalling which is which, yet I could show you a picture of a stone tower in a field with the word Bailey on the field and motte on the tower and you’d never again forget which is which. Then I can show you a second picture with some arguments in place of the names and the concept is solidified. Two pictures, maybe 20 words instead of hundreds of words to communicate a complex idea. Additionally, I can reach people with varying reading skills and possibly across language barriers. It’s an amazing brain hack but I think we’re only in the early days, still developing the language.
As the language and use grow, savvy memers will be able to add complexity to their memes adding symbols, links, slogans, branding, etc. to add further context. The best will be able to supercharge them. Maybe this is all obvious stuff, but I find it amazing.
One thing I’ve kind felt was actually missing from much of the Rationalist, IDW, online-intelligentsia is an ability to condense information in ways that make an impact on people in such a way that they can easily recall and communicate the concept. Scott is actually pretty good at this, but things like slogans and phrases that act as placeholders for deep concepts are important but perhaps not sufficiently dense.
Some ideas are really easy to get and have taken off, for instance, “steel man,” which I see everywhere now. People get it because it carries the phrase itself is a meta contextual reference to “straw man,” which is a known concept. Others, like, say, “Dunning-Kruger” can often be misused or misunderstood because the concept is sufficiently deep and the term not informationally dense enough to convey the nuance. Someone, somewhere could likely create a single panel meme to convey this concept better than the millions of words already written about it.
I think this is important for very smart people to figure out because if done well it can act as a bit of a corrective for the abbreviated thinking that will inevitably happen when dealing with deep concepts and can aid in the spread of important ideas. Historically, I think this was done with songs, poems, jokes and powerful anecdotes. internet memes are just the newest (also oldest?) tool.
As for your original question I think I understand better what you were asking, which is are the memes (commonly used ideas) we discuss here specifically geared toward sense and decision making and I’d say yes. New ideas are still going to come from original and creative thinkers (who can also be rationalists) but over all, rationalism and it’s memeplex is going to be about justifying decisions and orienting toward a better way of acting in the world.
I've seen any number of explanations for which is the motte and which is the bailey, and it doesn't stick. Maybe repetition will help, or maybe I can come up with a mnemonic that works for me.
Meanwhile, I think there's a lot to be said for tower and field. Maybe fort and field.
The unfortunate thing about motte-and-bailey in particular is that using archaic words as a metaphor to descibe a concept is obviously counter-productive, and doesn't do any favors regarding the stereotype of rationalists being pretentious nerds.
It contains posts on a wide variety of topics. The ones I think this community would find most interesting are the ones on book reviews ( http://thechaostician.com/category/books/ ) and on the philosophy of science ( http://thechaostician.com/tag/philosophy-of-science/ ). I also write descriptions of science for a popular audience, posts about Mormonism, and occasionally posts on history.
I am currently updating every Monday and Thursday.
There are some injuries which seem like extreme bad luck (slipped on pet poo and hit one's head on the bathtub, or stepped in a deep narrow unmarked hole and broke bones), but a lot of them are more like "sneezed wrong and screwed up my back". "Something mysterious happened while I slept and now my shoulder is fucked up."
They do tend to happen more often to people over 30, though not always.
While this is a good place to share more stories, I'm also interested in what might be going on. Coordination becomes less reliable with age? Connective tissue is weaker? Those little muscles which specialize in support and stability are weaker?
Is this sort of thing less likely to happen to athletic people? Do sports injuries make up the difference? Is anyone even keeping track of weird injuries that happen for no apparent reason in normally risk-free activities?
There's a lot of talk about life extension. Should we assume that life/health extension will simply lead to at least some degree of rejuvenation?
I agree that an uploaded human brain might well need human-style sensory input. Or maybe it could adapt to some other sort of input that's at least of comparable density.
Some of those things may be simply damage accumulated through decades.
Over 40 I have a problem with my spinal disc. Almost certainly a consequence of the sedentary lifestyle (software developer). But it doesn't necassarily imply than these days I sit *more* that I used a decade ago. It might mean that some small damages in the spinal disc tissue were added perhaps since I was a teenager, and at some moment it was too much and something broke. But the thing that "made it break" could be the last 1% of the damage, with the previous 99% caused by other things.
In my case, one day I woke up, stretched my legs... and started screaming with pain. Of course it would be silly to say that stretching my legs *caused* the problem.
> There's a lot of talk about life extension. Should we assume that life/health extension will simply lead to at least some degree of rejuvenation?
You can extend the *average* life span a little by fixing some of the causes, while not fixing others. Though extending to 200 years would probably require some rejuvenation, because many parts of the human body seems to have a shorter warranty period.
There are people who live into their nineties in good health and die after deteriorating for a little while. It's genetic, and I'm hoping there will turn out to be a way to bottle it.
So it should be possible to extend the average life span and health span something like 10 or 20 years-- that's a lot from one angle and very little from another.
Accumulated damage is worth adding to the theories about those weird accidents, though it might also be plausible that if people's aging were stopped, damage would accumulate more slowly.
Stress. As you age, your body learns to store stress in obscure parts of the body where it can't interfere with everyday life. But one day you stretch, take an odd step, or turn at an odd angle, and suddenly, that obscure part of your body is called upon to bear an unaccustomed weight or motion. Stiff and unprepared, it does not respond flexibly by transferring the weight or motion to the parts of the body that can handle it. Instead, it takes the whole load upon itself, disrupting the body's fine balance of forces and causing injury.
So for instance, 99 percent of the time, you can stretch your arms and enjoy a huge, gaping yawn. But one day, your body may be storing stress in some in a tiny pair of muscles in your jaw, and when you open wide to yawn, your jaw painfully locks.
Most commonly you are lifting furniture, a box or some other weight that your body is perfectly capable of lifting without injury, even if you lift it "wrong", but on this particular day, you resent having to move furniture or lift boxes, or your boss or spouse is criticizing your work, or you are tense about something else and your body happens to be storing stress in those back muscles you don't ordinarily use while standing or sitting, and those muscles are suddenly called upon to bear the weight of whatever you're lifting - you will soon be lying on the floor howling.
I've been wondering how much of this is actual new weakness/damage to older people, and how much is just new awareness of existing weakness/damage that younger people just ignore. Surely some of the changes are real. But some of it is probably also just that after decades of your body being largely a fixed shape and size, you notice little changes more than you do just two years after you stopped growing.
The accidents I started this thread with are presumably the revelation of weaknesses which have been building up slowly, but it's not that the weakness is directly perceptible.
I think this is from general decay of health of muscles and connective tissue. With connective tissues, you often won't notice anything is wrong until an acute injury.
While you're young and growing, tissues tend to recover fast and stay healthy and flexible. As you age, the tissues tend to stiffen and degenerate, and accumulate wear-and-tear.
Being active helps: it can help stimulate recovery processes in the tissues, and having a surplus of strength can be helpful for preventing injuries (e.g. being able to catch yourself from a fall without getting hurt). However, even being active, wear-and-tear accumulates in a lot of tissues: cartilage, tendons, and ligaments heal slowly and incompletely. Weightlifting is a somewhat serious hobby of mine, and even though I've done my best to train safely and have almost no acute injuries in 15 years, my body doesn't recover the same at 40 as it did at 25.
I think life extension is a multifaceted problem, and rejuvenating connective tissues is one of those facets. It's quite possible we'll figure out how to regrow cartilage before we can extend lifespan, and vice versa.
Carrying on from the Eurovision thread, I have to link to the Icelandic jury results from Saturday night because this is life imitating art https://www.youtube.com/watch?v=aocIlMp-WWY
Italy were the eventual winners, if anyone is interested 😀
I'm interested. It was the first time I ever saw it.
Was very surprised Italy won. I now know they were the favorites going into it, but I didn't know that while watching and didn't think they had a chance. I thought it would be between Iceland and Russia. Was very surprised Russia didn't score higher. All I can figure is that the contest is about producing maximal irony and a 1980s style glam rock band winning in 2021 was maximally ironic.
Or was it simply a good strategy to put a rock band against all those dance numbers? The dance number fans are going to divide their vote, whereas the rockers only had to choose between Italy and Finland.
Or am I trying to make to much sense of it? The jury voting seemed random. Was that mainly politics? I was totally confused by the other votes showing up on the board while the juries were voting. What were those?
Why did the juries like Switzerland so much? My girlfriend and I guessed that was one of the acts that would get zero votes because it sounded horrible to our ears, yet obviously the juries thought differently. Many of the others made sense. England and Germany were genuinely horrible, although at least Germany was a bit funny. France was fine, but that performance sure didn't seem to be in the spirit of the contest. No originality, no choreography, no fun.
I suppose Iceland was too much in the spirit of it all to win it all, too on the nose, particularly after the movie. I suppose I can see why all the 13-year-olds voting wouldn't vote for what all us Olds liked best. Yet Iceland did get a lot of votes, just not compared to the Motley Crue of Italy.
Is the British jury still upset over the cod wars with Iceland? I read what you wrote in the last open comments about Eurovision and how political the voting is, and noticed how Cypress and Greece voted for each other. Were all those votes for Switzerland due to politics or did people actually like that?
The jury vote isn't only the 12-point awarded verbally. The juries also award 1-8 and 10 points, but instead of being announced those just show up on the screen. Could that be what you mean by other votes during jury announcement?
I'll just be thankful that they've not taken that opportunity to make the already excruciating jury vote announcements even longer! Especially when they could at least be pre-recorded (the juries vote the night before) so as to remove delays andess-ups. Though Iceland jury announcement was worth the waits!
Thanks for the explanation about the jury votes. I thought that may be the case but wasn't sure after reading about the scoring system and was confused while watching whether those were jury votes or phone votes.
To bring this back to a Scott theme. I honestly and naively thought the feminist power of the Russia song would go over well. It was a good song. I'm no fan of Russia politically but this song was no pro-Putin thing (unless you take a very Cowenian (Straussian)) look at it. Yet it had no resonance, despite being a great performance.
That leads me to believe the feminist thing is over, for now, internationally. Perhaps Putin even allowed it because he realized it didn't matter.
More countries with low numbers of black people had black performers than I expected. Sweden, The Czech Republic, Netherlands... but nobody seemed to care. The Swedish song wasn't horrible but it wasn't anything. The individual countries seemed to be engaging in affirmative action, in terms of the acts they put forward, but the juries didn't give a shit. Whatever was driving the politics of the jury votes, it wasn't wokism.
Israel had a black performer who I first thought was a ringer because she was so good, but it turns out she is not ringer. I was surprised Israel didn't score higher, because I thought that song and performance was really good.
San Marino, wtf? I didn't realize that was a country, the song was horrible, and when I looked it up, it's an American singer??? They should not have made the finals. But neither should have England or Germany.
I have read that black, in the American sense, isn't really a thing in most other counties. Other minorities fill that role. In France in the Algerians, in Germany the Turks . . . those are the only two I have actually. I know a famous black writer, James Baldwin I think, talked about how freeing to be able to put down the burden of being BLACK for a while and just live.
In The Netherlands there is a huge American influence and the Dutch woke are definitely trying to make it all about black people. We do have a decent number of migrants from the Carribean part of the Kingdom, but there are far fewer of them than black Americans, so the claim that they should be way more prominent is really only popular where they 'pool,' which is mainly in Amsterdam.
The Carribean migrants have relatively high criminality and poverty, but not really that poor of a reputation. The ethnic group with the worst reputation are the Moroccans.
People liked the Russian song, but not enough. And I saw some commentary while liveblogging the final about "Good song, but fuck Russia" so there is definitely a political angle going.
"What kind of song will win Eurovision?" is a perennial guessing-game. There tend to be a lot of ballads anyway, for some reason, but if a ballad wins this year, next year most of the entries will be ballads.
Since a (glam)rock band won this year, I expect to see a lot more rock-type acts next year. Nobody knows what the magic formula is, which makes the Swedish interval act by the hosts in 2016, about what the perfect winning song would be, so funny: https://www.youtube.com/watch?v=aMgW54HBOS0
"More countries with low numbers of black people had black performers than I expected."
Austria - singer is Filipino-Austrian (born in Vienna to Filipino parents)
Czech Republic - singer is Angolan-Czech (born in Czechoslovakia to Czech mother and Angolan father)
Israel - singer is Ethiopian-Jewish (born in Jerusalem to Ethiopian mother and Israeli father)
Malta - singer is Nigerian-Maltese (born in Malta to Maltese mother and Nigerian father)
The Netherlands - singer is Surinamese, a former Dutch colony
Sweden - singer is Congolese refugee living in Sweden since a child
San Marino - singer is Eritrean-Italian (born in Italy to Eritrean parents); Flo Rida (the American rapper) got involved with the production and because San Marino is one of the smaller countries, they're allowed bring in outside singers as there are no rules, technically, about where the singer has to come from (they don't have to be from the country they're representing).
They have links to the countries they're singing for, I don't see why this is a huge surprise (given the amount of African-American singers and musicians, acting as role models for black youth in other countries, going into pop music doesn't seem that unlikely).
Okay, now we delve into the history of Eurovision! (The following is a personal take, don't treat it as Gospel).
There are two really big divisions: Eurovision up to when all the Eastern European countries entered, and afterwards.
(1) The History
Eurovision started out as a relatively high-minded cultural exchange notion, as well as a test of the live broadcast technology of the day - members of the European Broadcasting Union https://en.wikipedia.org/wiki/European_Broadcasting_Union transmitted the first contest, where seven nations took part, in 1956. The ideas of the time were still working on post-war hopes of preventing yet another conflict by increasing mutual understanding and appealing to a common identity, and this was one way of doing it - through music!
The winner of one year is the host country for the next year. There's been a little bit of controversy over languages - there was a rule introduced that countries should send an entry in their native language, but then in the 70s this was relaxed and many countries choose to enter songs in English. Numbers of countries competing increased to around 20 every year, with some countries dropping out and new countries entering. This is how we get Israel, Turkey and Australia competing, even if technically they're not in Europe as such. Each competing nation holds a public song contest to select their Eurovision entry which is voted on by the national public (or however the country decides to select) and that act is then sent on to Eurovision.
The 70s were the hey-day of the first version of Eurovision, with ABBA the biggest success story, but not the only one to later have successful music careers. However, by the end of the 80s/start of the 90s, it started to feel stale. The divergence between commercial pop/rock music and the kind of songs that went to Eurovision was very great at this stage, and it didn't seem relevant any more.
Enter the second stage! After the end of the Cold War, break-up of the Soviet Union, etc. suddenly there were a lot of Eastern European countries that were anxious to be part of the whole pan-European experience, and Eurovision was one way of doing this. They started entering, and they took it very seriously. So this acted as a rejuvenation in one way, but on the other hand, a lot of the new acts were stuck in musical genres of 20 years or more before. https://eurovision.tv/countries
Eventually it settled down that everyone adjusted their expectations, it was taken semi-seriously (but neither too seriously nor not seriously enough) and it has transmogrified into a continent-wide night of extravaganza, rather than a song contest to find the best song/most representative European culture/new pop star/whatever.
So many countries were now entering that there had to be a selection process. It has now settled on two semi-finals, where the public vote for who goes forward to the grand final, and the grand final thereafter. However, there are countries that automatically qualify for the final, without having to go through the selection process:
(a) the host country
(b) the Big Five (because they pay the bulk of the funding to host Eurovision) - France, Germany, Italy, Spain and the UK
And that brings us on to the second part:
(2) The Voting
This is commonly taken to be the best part of the entire show, and again there is the old version and the new version.
You asked about "the other votes showing up on the board while the juries were voting". Traditionally, especially when the numbers competing were lower, the jury would announce every single vote cast from "one vote for X" up to the highest "12 votes for Y". (The famous "nul points" or "zero vote" was never officially announced, but you can always see by the scoreboard who gets no votes from anyone https://eurosong-contest.fandom.com/wiki/Nul_Points).
As more countries entered, this became much too long and drawn-out. So they tweaked the voting system, as well as experimenting with televoting. Right now it's a mix of "first the professional/national jury give their vote, then the public vote by phone or text or app is given".
Now the only result formally announced is "the Polish (or wherever) jury gives 12 points to North Macedonia (or wherever)". The other votes (10, 8, 7-1) are put up on the scoreboard at the same time.
You can see this year the difference between the professional (music industry, broadcasters, journalists and so forth) jury votes - they tended to go for Switzerland - and the public votes which resulted in Italy winning. There's often a lot of controversy about this, and it's often political voting.
Cyprus will *always* vote for Greece, and Greece will *always* vote for Cyprus. This goes back to the 1974 coup d'etat carried out by the Greek military to make Cyprus part of Greece, and the Turkish invasion in response, and even before (there was a UN intervention in 1963 to keep the peace between the Turkish Cypriots and the Greek Cypriots) and the island is still divided between North (Turkish) Cyprus and the Republic (Greek) of Cyprus.
Malta used to reliably vote for the UK, as it is a former British colony, but not even they gave them a vote this year.
Scandinavian countries tend to vote for each other, but not exclusively. Ditto Central and Eastern European countries. Neighbours vote for neighbours and against old enemies.
UK got nothing this year; not alone was the song not really a Eurovision winner, but you know, Brexit means Brexit 😀
The French entry was *incredibly* French; an impassioned ballad sung by a gamine in the vein of Edith Piaf/Jacques Brel, in black and white (as near as dammit)? https://www.youtube.com/watch?v=Unj9WbeLzRU
I just want to recommend the work of comic artist Anders Nilsen to everyone. I think that this is probably a pretty weird environment to recommend his work, but what the heck? This thread is about everything right?
Anders draws comics but to say that they are nothing like what you might think of when you think of comic books is a wild understatement. He's an artist who never ceases to shock me with his humor and unutterable artistry.
I got the fourth comic in his book TONGUES, his current ongoing project, and just read it today. It's just next level work.
The last really big work he completed is a tome of a book called BIG QUESTIONS. Like it sounds, it evokes really big topics.
He takes big archetypal imagery and weaves them together into dreamlike stories that can be unpacked for years.
He's always been a brilliant artist but his current work feels like he's deconstructing the whole form. Like you're seeing the future.
I don't have much agenda here. I've been sitting here all afternoon kind of in the hum of how good his most recent entry in TONGUES is and just wanted to say it somewhere. I saw this thread open and thought: why not.
This is the one I just read, but you'd want to start at the beginning:
Huge comic nerd here, though almost no Marvel or DC stuff. I haven’t heard of this guy before, so I’m excited to check him out. Thanks for the recommendation!
Other question: has anyone ever written about how the current use of the term "rationalist" doesn't really square with like
How they talked about it when it was a question fo Descartes (a rationalist) v. Hume (an empiricist)
As I see it (and look I could be wrong)
I see today's rationalists as more like empiricists, but not quit so pure about it as Hume. Like they see reason itself as a decent little backup to observation, but they would probably put observation above reason.
Descartes tried to shake everything out JUST from reason and Hume basically said you can't get anywhere with reason (right?)
So it seems like the current conception in the SSC/Galef/LW end of today's culture is sort of a marriage of the two?
Popular use of the words "rational" and "empirical" are much closer to synonyms than antonyms.
Since Kant (who is probably the origin of this potted history of the 17th and 18th century), nearly all philosophers have wanted to retain elements of both the rationalist and empiricist worldviews, while rejecting elements of each as well.
Statements like "China is keeping the Yuan undervalued" confuse me. What does that even mean?
Sounds like it should mean that you can buy more stuff with Yuan than with the exchange-rate equivalent of Dollars. But if that's the case, can't I just arbitrage that away by buying stuff with Yuan, selling it for Dollars and exchanging back?
You could, but that would mean buying it in China and selling it elsewhere, which is exactly what the Chinese government wants - they're basically subsidizing the exchange rate in order to promote exports.
Somewhat. Although this can also have a market distortionary effect that can hurt them (subsidizing exports costs them access to both consumer goods and intermediate industry inputs), so it's also not a riskless strategy even if you don't mind making consumer goods more expensive.
I dunno if you can arbitrage it away, because your source of yuan is the Chinese government, and they can print and sell to you infinity yuan. So it seems to me you will run out of dollars before you can successfully bid up the price of yuan.
Right. But if you (or a lot of you) want to bid the price of the yuan up to its "real" value you need to keep repeating this cycle, over and over again, and trust that your increased demand for yuan at step #2 eventually results in a higher price for yuan. Normally it would, but when the "seller" (the Chinese government) of yuan has essentially a zero marginal cost of production, they can sell you infinity yuan (as far as you're concerned) without raising the price. So long before you succeed in raising the price of the yuan, you run out of dollars. Well, more precisely, the price of your goods in dollars falls as you increase supply until step #4 no longer gets you >X dollars.
your last statement proves the point. in the above, i sell goods for >X dollars, so i will never run out. if the price of goods in dollars falls such that the cost is equivelant in dollars and yuan, then i have succesfully equalized the value.
I don’t know if this has come up here in the past. For people with strong feelings that they know the ‘best’ political/economic system to live under, they should consider Warren Buffet’s ‘Ovarian Lottery’ thought experiment before they go all in on it.
Yes i just read a quick summary of VOI and they do seem to be getting at the same idea. I think the ovarian lottery take might be a more viscerally experienced and easier to understand way to drive the idea home with someone not open to or experienced with a more academic take.
This has the same fundamental problem as Rawls' Veil of Ignorance: it depends on metaphysical dualism and falls apart entirely if humans are biological organisms instead of souls trapped within flesh.
There is no lottery involved in being your parents' child as that is the only possible option. Your laudable traits are inherited from them, including "mental" traits like intelligence or conscientiousness which depend on a physical brain, so anything they can be said to have earned through their laudable traits likewise should apply just as much to you. And vice versa for their condemnable traits.
If there is luck involved, it is at a different stage from heredity which is not a tremendously random process. Recombination smooths things out and of course there's always the chance of de novo mutations, aneuploidy, etc. but overall you're much more similar to your parents than our culture is willing to admit.
The Ovarian Lottery isn't supposed to be a real thing that exists. It's an idea to help people think altruistically.
As humans, we should care about the well-being of all humans. But it's easy to fall into only caring ourselves and those like ourselves. The point of the Ovarian Lottery idea is that it helps you consider the plight of everyone in the world. If there's a class of people that make you think "Oh, I'd hate to be one of those", then you should consider how to fix that.
I’m working at a job that is really dragging me down emotionally. The pay is not bad but the work is stressful and not at all stimulating or meaningful (I’d like to have at least one of those two). I’m stuck when it comes to deciding what to do with my life and I don’t think I have the financial luxury of hopping from job to job until I find something satisfying. I think I’m pretty intelligent and willing to work hard but my educational background is narrow enough that it probably won’t help me do anything I would actually want to do. Anyone have any ideas/suggestions?
You could educate yourself in something you want to do. E.g. I know someone who wanted to be a medical doctor but had an undergrad degree in english so they did a post-baccalaureate in pre-med for a year and then applied to med school. I know someone else with an undergrad degree in music doing a 2 year post-baccalaureate in CS.
I'd make general job satisfaction a relatively high priority. It's where you spend ~half your waking hours and I know several people who got crushed mentally by enduring a job they hate. That said, there may be many ways to solve this without quitting your job, including
- talking to your boss or co-workers (generally speaking, employees who enjoy their work tend to perform better and people have different things they enjoy / hate)
- reducing your financial needs and your work hours (or split work between two jobs)
- finding something worthwhile in your free time (if you fell in love tomorrow, would the job still suck so much?)
If none of this is possible and you find yourself in a downward spiral, please. Quit. Before. It's. Too. Late!!! It you're reorienting, the 80.000 hours guide on finding fullfilling work might help.
Dr Fauci also said today that he considers it an open question that should be investigated.
What do people think about this? Are there any virologists in this community who can give an informed opinion on it? 60% sounds pretty high to me, although I don't view the lab-leak hypothesis as being a fringe conspiracy theory anymore.
A bit of both. I've changed my mind on the lab leak hypothesis (I now think it is likely true), mainly due to evidence that I happened to see recently (but may have been available some time ago). The fact that we can't find a wild host population is the kind of evidence that only become stronger with time...
Yea, that was the case for me as well. Technically none of the evidence have been news, but they haven't really made it into the mainstream until quite recently. Kudos to the people who led the charge when nobody wanted to listen.
I thought the tweet quoted in the link above was quite significant new evidence ("if it spread from the lab, why are there no cases of employees there having it early on" previously seemed like a pretty compelling argument against the leak hypothesis but is no long).
Nate's tweet implied that he went from 40% before to 60% after, implying that he saw the new evidence (3 WIV employees visiting hospital in November) as actually somewhat weak (only 2.25 times more likely with lab leak vs. no-lab-leak). That's kind of where I am on this as well: I saw the lab leak as highly plausible before the new report, given:
1. the location (near a lab doing research using the most-similar known virus but 1000 miles from a SARS spillover zone)
2. new viruses that have just jumped from one species to another usually undergo rapid evolution as they adapt to their new host; SARS-CoV-2 did not, suggesting it was already well-adapted to infecting humans.
3. the lack of concrete evidence for natural origins, which was identified much faster in the case of SARS-1
4. lack of transparency from WIV (hiding databases, hiding the origin of RaTG13, obfuscating its connection to pneumonia cases in mojiang miners) that suggests they have something to hide
The new report is additional circumstantial evidence, but somewhat weak given that this could have been "common seasonal illness", and given that it was already reported that WIV workers were ill, just not that they visited a hospital.
60% has been a reasonable estimate from the start; I'd consider anything in the 40-90% range to be reasonable and I wouldn't consider it reasonable to spend too much time trying to pin it down. Not reasonable to spend too much effort to pin it down because A: it's unprovable and unfalsifiable barring a highly implausible outbreak of transparency in China and B: it's not actionable; everything we'd do if we knew it were true is something we should be doing just on the basis of it's being plausible.
What's interesting is that two months ago it was politically maximally incorrect to suggest this possibility, and now it's somehow OK and everyone's doing it. But that probably does get us into culture war politics in some way I haven't figured out. It's weird.
But, a priori, there's two ways for a zoonotic virus from a distant source population to show up in Wuhan. One is for a random walks of natural infection and bushmeat trade to proceed unnoticed across ~500 km until it shows up in that one wet market and no other, having left no trace of prior infections that anyone can find. The other is that people we *know* were in the business of going forth from Wuhan to collect strange new pathogenic life and bring it home to their lab in Wuhan, did exactly that and screwed up on the "don't let it out of the lab" part. This was clear from the start.
There's also the question of what people in the lab might have done to make COVID-19 more dangerous than it naturally would have been, but since the plausible answers mostly involve attempts to tease spoiler alerts out of Mother Nature as to what the virus would have evolved into on its own in another ten years. that too falls mostly into the unprovable-and-unactionable category.
Why not actionable? It seems that if we better understand the risks associated with certain form of research, that may inform us on how to better protect ourselves, and even if the benefits of such research outweigh the costs. I heard Josh Rogin saying that they're working on sextupling the funding for that kind of research, presumably as a measure to improve pandemic response. But if it turns out that kind of research actively caused the pandemic, that would probably be good to know.
I think the idea is that if the research changed the chance of this happening from 0.1% per year to 1% per year, that's quite significant whether or not this incident actually occurred that way, and the evidence about this one incident isn't likely to actually tell us much about the change in probability.
If we find out with 100% confidence that *this* time it was a wholly natural event, are we going to rationally respond with "OK, gain-of-function research is totally awesome and we don't have to be picky about whether it's done at BSL-4 or BSL-2 or BSL-I'm-too-lazy-to-follow-the-rules"? There's a set of behaviors, involving both laboratories and wet markets, that is rational based on what we already know and isn't going to change if we add detail or confidence to what we already know. So the details aren't actionable.
Why was it maximally politically incorrect two months ago but not today? I think this article has a good explanation of why the media initially treated it as a debunked conspiracy theory, even though it had some plausibility: https://www.washingtonpost.com/politics/2021/05/24/fix-china-lab-leak-0524/ (warning, politics plays a role). Once the hypothesis was placed in the land of "debunked conspiracy theory", it was hard for respectable people to be publicly seen to take it seriously, but this also meant the idea had much higher private support than it did public support. This created the perfect environment for a respectability cascade, which was further accelerated by increasing circumstantial evidence pointing in that direction, and a complete lack of any new evidence for purely zoonotic origins in over a year.
> Not reasonable to spend too much effort to pin it down because A: it's unprovable and unfalsifiable barring a highly implausible outbreak of transparency in China and B: it's not actionable; everything we'd do if we knew it were true is something we should be doing just on the basis of it's being plausible.
I don't think either of these are true in practice. China could be pressured by every other nation that suffered from their possible negligeance, so we could get more insight, and protocols for gain of function research should be significantly tightened.
I suggest reading the best article that reviews the available evidence to see why your B is likely false:
So, if we find the smoking bat corpse that conclusively proves that Covid-19's origin was wholly natural, would you and Young then reason, "...and so my priors are now shifted towards US labs doing their work safely and we don't have to worry so much about lab safety"?
In order for a data point to be actionable, there has to be something you'd do *differently* if that data point were different. Preferably something more substantial that "use/not use this incident as a talking point when debating an issue that I'm not really going to change my mind on".
Rootclaim has a site on this - https://www.rootclaim.com/analysis/What-is-the-source-of-COVID-19-SARS-CoV-2 - with full quantitative reasoning and Bayesian likelihood calculations. Their reasoning may or may not be correct, but it crucially is *transparent* - you can see where the hypothesis starts, how and why various pieces of evidence affect their estimate, and what likelihoods they end up with.
Precisely. GoF is basically trying to speed up the natural process of jumping host, so I do not see how it could leave specific clues in the genome, contrary to editing (which specialist find highly unlikely, because engineering a efficient viral genome is much harder than let natural selection do its stuff (try a lot of variants and keep what work better -rinse a repeat). Some genetic histories may provide clues, especially the viral evolution speed after first observation, but if anything this plead for significant evolution before being observed in the wild, i.e. GoF. For example, many siginificant new variants may be linked to a passage through mink farms....Which is basically, from what I understand, an out-of-lab GoF...
My opinion about the likelyhood of an accidental lab release jumped to "damn likely" as soon as I learned that Wuhan lab did GoF studies and had gathered (so had access) to a genetically-close-to-covid bat coronavirus able to infect humans (the closest found yet). If the lab escape was wrong, those two facts would be a crazy coincidence...
There is well known, impossible to cover up albeit indirect evidence that SARS-COV-2 leaked from the lab: the fact that it was immediately so well adapted to infecting humans. Virologists have conjectured that it must have circulated among a sub-population of humans first to gain this ability, but no evidence of this population has been found.
I was hoping the the other poster had some other, more direct indicators.
That fact is also consistent with a natural origin if we also hypothesize that viruses which are mediocre at infecting humans in rural Guangdong are never noticed because nobody cares if a few Guangdongees farmers or guano-miners die and with only mediocre infectivity the occasional infected farmer visiting the Big City isn't likely to trigger a visible outbreak. So the virus gets to evolve in rural Guangdong for years until it *is* well adapted to infecting humans, and *then* the first carrier who visits the Big City starts a significant outbreak.
The bit where the Big City in question is Wuhan rather than Guangzhou is still quite suggestive, of course.
Posted this a bit late in the last OT and got no replies. While it contains the word "politics" I don't think it's actually political, so reposting:
If you think AI takeoff is likely to happen in the next 20-30 years but are useless at coding/politics/business management/finance, how do you do something useful with your life?
I mean, aside from literally stopping people from dying before singularity (which could actually be bad in case of Roko-style or MMAcevedo-style Virtual Hell), there's not much you can do for people that won't be rendered irrelevant by a bad *or* good AI takeoff, and without access to one of those you have ~0 control over the time and goodness of takeoff.
Barring anti-aging, everyone alive now is going to die within ~100 years anyway. If everyone is going to die (or ascend) within 30 years, I'm not sure that changes the calculus too much. If you were going to try to save/improve someone's life despite them dying in a century anyway, it's probably worth saving/improving their life when they'll die in 30 years.
Or if you really want to focus on the AI thing you can work a normal job and donate some percent of your income to charities trying to improve AI safety.
My passion is teaching. One could hope under normal circumstances that this would do some lasting good (i.e. helps people help other people etc. and at future time X for X < some large number the world is distinguishably better). However, if we all get eaten by Universal Paperclips in the year 2040, then the state of humanity is effectively erased at that time; I could be the best teacher in the world or the worst, and my students would still all be dead in 2041. And if we all get uplifted into a proto-super-Culture in the year 2040, then all my students could download all the knowledge of humanity into their brains instantly and would have essentially-identical living conditions regardless of my input (the only exception being if they're dead by then). Again, the information of what I've done is effectively chucked into a black hole.
If I thought AI were 100 years off, then I wouldn't be so worried, as I'd be more confident that I'd have some butterfly effect on the nature of what happens. But 20-30 years is rubbing up against "too short a timescale for this" (I think you might have said something along these lines at some point, or linked to such?).
And I can route around little problems like nuclear war; I can go into regional education, and I'm more than capable of teaching people to Duck and Cover. But a bad AI is under the category of "storms we cannot weather", along with alien invasion or divine/simulator intervention.
As far as charity goes... well, that has to do with why I mentioned finance. I'm not especially likely to get rich enough to be more than a drop in the bucket there, and of course this assumes charities are of positive use here (it's never 10^-67, but it's sometimes 0 or -0.001 - for instance, if an AI-safety charity has differential ability to affect Doctor Good and Doctor Amoral, they might end up giving the keys to the kingdom from what would have been a Doctor Mostly-Good 30%-shot to a Doctor Amoral 5%-shot). I mean, sure, that's something I'll consider, but it feels pretty limp.
“The doctrine of the Second Coming teaches us that we do not and cannot know when the world drama will end. The curtain may be rung down at any moment: say, before you have finished reading this paragraph. This seems to some people intolerably frustrating. So many things would be interrupted. [...]
“A man of seventy need not be always feeling (much less talking) about his approaching death: but a wise man of seventy should always take it into account. He would be foolish to embark on schemes which presuppose twenty more years of life: he would be criminally foolish not to make, indeed, not to have made long since his will. Now, what death is to each man, the Second Coming is to the whole human race. We all believe, I suppose, that a man should ‘sit loose’ to his own individual life, should remember how short, precarious, temporary, and provisional a thing it is; should never give all his heart to anything which will end when his life ends. What modern [people] find it harder to remember is that the whole life of humanity in this world is also precarious, temporary, provisional. [...]
“Frantic administration of panaceas to the world is certainly discouraged by the reflection that ‘this present’ might be ‘the world's last night’; sober work for the future, within the limits of ordinary morality and prudence, is not.” (C. S. Lewis, “What If This Present Were the World’s Last Night?”)
“It is like a man going on a journey, when he leaves home and puts his servants in charge, each with his work, and commands the doorkeeper to stay awake. Therefore stay awake—for you do not know when the master of the house will come, in the evening, or at midnight, or when the rooster crows, or in the morning—lest he come suddenly and find you asleep.” (Mark 13:34-6)
I think the principles are very similar in the case of the singularity, although without the expectation of Judgement that gives a particularly sharp moral character to the whole question for theists. You cannot know when the paperclip maximizers will dissolve the earth, but then again you cannot know when you will be diagnosed with cancer or hit by a passing car—we all invest our lives in projects we may not see to fruition. All we can do is the work in front of us, the tasks that our social roles, obligations, talents, and resources, as well as the needs around us, fit us to pursue. I think (though I don’t know the right philosophical term) that this can even be made utilitarian: if you had perfect knowledge you might calculate that doing something very unusual would maximize utility, or that action on your part is useless and you might as well give up or do as you please, but human knowledge is never perfect and most people who believe themselves to be in these exceptional cases are wrong, so it is very likely best for everyone to stick with “sober work” and “ordinary morality and prudence” even when they believe their circumstances are exceptional. I would say, teach and serve in the time given to you; if the singularity comes, let it find you at your post, and you may after all be pleasantly surprised with more time than you’d planned.
If you're a genre fiction author, do you think of your novels also as story treatments for potential movies or series (in other words, are you actively hoping to get your books adapted)? If so, how does this affect your writing?
You'll probably get more relevant responses than mine, but I was once a wannabee fiction author and wannabee screenplay writer.
If you want your novel to one day be a movie, you should think cinematically, which mainly means thinking visually. Exposition and interior monologue are not cinematic. (Voice overs should be a last resort in movies.) All visual detail is good. The more granular the better.
Dialogue should be tight and minimalistic. Hemingway basically invented movie dialogue. His exact words didn't make it to the movies because the studios back then had a philosophy that everything needed to be rewritten, but all the good movie writers stole his style. Hollywood today would still want to rewrite your dialogue, but if you want it to have a chance keep it minimalistic.
The main difference between a movie and a novel is dramatic structure. A novel can be any length and follow whatever rhythms it wants. A screenplay generally needs a 3-Act structure. It's probably easier to turn a short story into a screenplay than a novel, but we no longer live in the era of the popular short story. You should probably think of what the short story version of your novel would be like and actually write one, if you want to imagine it as a movie.
Ultimately, though, the more commercially successful your novel the more likely you are to sell the movie rights to it for a lot of money.
I realize that I didn't answer your question, but I answered the question I was able to. For all I know you are trying to avoid the influence of the movies on your novel writing. If so, more power to you!
I value being an independent thinker -- it is one of my highest values -- yet I realize I go too far with it sometimes. I hate to go with the crowd. I was raised with the idea that the crowd is mad and hate the stink of the herd, yet these days the educated wisdom is that the crowd is wise not mad.
Emerson says "One recognizes in the genius of others one's own aborted thoughts." I believe in the wisdom that statement, so I try to see my own thoughts through, even if they seem anti-social.
Perhaps the Internet has made a major difference in how one should weigh their own thoughts vs the crowd. As a pre-Internet person, "the crowd" was the high-school halls. You were a fool to follow that crowd. Or very clever.
Now you can choose your own crowd. You can follow a crowd that you know is smarter than you. That was rarely the case for most people for most of human history.
Perhaps independent thought has less value these days. Perhaps there are fewer $20 bills on the sidewalk simply because information travels so much faster.
My question is -- I usually have a question -- are there good signposts for determining when your independent thinking has gone astray?
Yudkowsky spends a lot of words trying to address that question, I think. In his Inadequate Equilibria book he talks about what basically is holes in the common wisdom, areas where you can hope to beat it.
Other than that, I have a feeling this boils down to 'assessing the quality of your thinking', which is basically the same as (the bottleneck in) improving it.
When evaluating my own thinking in comparison to others', I've found a helpful test to be "assuming other people are reasonable, can I explain how they arrived at their beliefs?" This is similar, but not the same, as steelmanning the other side--sometimes the other side is simply wrong, but there are social or cultural reasons why their beliefs persist in reasonable people. In particular, "other people are dumb/evil" is not an acceptable explanation. If you can't explain the reason for why the consensus views are such, then you're probably missing important information.
> yet these days the educated wisdom is that the crowd is wise not mad.
A position that's been perverted by poor science reporting. The crowd is wise when ts members are not all exposed to the same information and opinions, and thus must make up their own minds from a plurality of non-ovrelapping data.
Any bets on whether that describes the current media landscape?
Since my earlier post was linked in this open thread as “Russia strong”, I should clarify that “Russia mediocre” would be more accurate and less interesting.
One time a boss told me something that has stayed with me. He said "It's easy to say 'This is fucked. What's hard is contributing something better.'"
There's truth in that, yet I think more and more people should be saying "This is fucked." at their job. I read endlessly on the internet about people complaining about how woke or whatever their workplace is. What happened to saying "This is fucked."? Granted, I'm a Gen-Xer and that's all we said, so it's easier for me to say. I just don't get why people can't tell their corporate bosses these days how they are fucked.
Corporate culture has so much power today because people worship at its altar. We used to want to die before we turned 30. If you work at Google and complain about it's culture, well, you had other options than selling out. Your politics on the internet don't matter: You are the problem. You support our woke power structures if you work for corporate America.
Well, some argue that the main reason for the wokeness push is exactly so that it would sideline economic leftism. Incidentally, one of its main features are labor unions, the very institution supposed to compensate for the powerlessness of a single employee in the face of general fuckedness of the workplace.
“ Any freedom that right-wingers claim for themselves they will happily grant every american has.”
That’s very true of the libertarian wing of the right. It’s very much not the case for the (much larger) non-libertarian leaning majority of the right.
This is interesting - I thought the entire idea of woke-ism was people starting to say "this is fucked" without any idea of what to do better, and finally getting around to saying it, instead of just ignoring it like we did for the previous several decades.
Yep, that was very sudden, but then again the manga industry is both very secretive and extremely exploitative, so maybe it shouldn't be too surprising. Still, as far as "dark fantasy epics to remain unfinished by their author" go, I certainly expected ASOIAF to be confirmed in this status first.
I never even realized Miura was suffering from stress, since I didn't look much at his work aside from Berserk (I was on the boat for 4 years since I started reading late). It's sad. I seriously hope Martin takes care of himself.
Ugh. I crunched some Metaculus predictions against actuarial estimates yesterday, and my ballpark estimate is that we have a ~22% of getting A Dream of Spring before Martin's death, and that number roughly halves every year.
*every additional year from 2021 that passes without publication of The Winds of Winter, that is. Still a guesstimate, but that's what it's conditioned on.
Agreed. He was pretty young too as these things go -- 54. I recently saw an interesting take on the timing of his death with the plot of Berserk, which is that Miura's death came at the one point in the story where most of the characters have found some kind of peace. (Obviously a lot of bad stuff still going on, but relatively speaking, I think the main cast is at a pretty high point compared to usual.) Even though it's kind of silly, this makes me hope that Miura himself -- clearly being a guy with many personal demons -- found some kind of peace near the end. Maybe wishful thinking, but I've had a bad week, so I'll take it.
Sorry for posting so much. I'll go to sleep soon and won't post again for a few weeks. Just getting all my ideas out now.
Why do people in this crowd care so much about future humans? I can understand caring about your grandchildren, but your grandchildren's grandchildren? Really? I don't care about them.
The future of the human race? Is that even going to be a thing? We're not crocodiles. Our environment is changing and we are changing with it. H.G. Wells probably had the right idea in The Time Machine: the future of humanity will not be humanity, it will be some bifurcation of troglodytes and geeks. Why care about the future of your genes? They will be a different species than you. We are Cro-Magnons and Neanderthals, to our descendants, or just dump apes. Or maybe our descendants will be more like rats. Intelligence isn't everything. Why do you care about our great grandchildren rats so much? You want to fill the universe with them? We'll be the species that fills the universe with rats. I know I will feel proud about that.
The Sex Pistols had it right, there's no future for humanity. Our genes may have a future, but it won't be human.
What I do find remarkable is how volatile human terminal goals are. Or at least they feel like terminal goals, until some supporting assumption breaks.
Why should an agent change their fundamental decision theory just because of some philosophical insight? Humans seem to do that. Nihilism as a class of glitches.
Makes me wonder how our goals are really organized. What is the meta-utility-function? What do we truly want?
That's because terminal goals are unimportant from evo-psych perspective, they matter only insofar as they influence behavior. And it takes a very special kind of human to substantially change their day to day life after experiencing a philosophical insight, so rare that it's statistically insignificant.
Do you expect some complicated argument involving acausal game theory and simulation capture or something? Too bad, I just want to continue experiencing hedonic pleasures. I don't care whether the information-theoretic distance between "me" right now and "me" two weeks ago is within some epsilon-level of tolerance, "my" ability to experience things get taken along on the ride anyway.
I'm not sure I care tremendously about future humans, but I at least find the idea of caring about them attractive.
My assumption is that genetic engineering will lead to an explosion of different sorts of humans. Some of them will be somewhat like us, some of them will be very different.
I've wondered about heritage from a different angle. There's a big deal made about ancient art being kept in the possession of the modern country which is on the territory where it was made, but I can't imagine the ancient Greeks caring about whether modern Greece has the Elgin marbles. Not that hypothetical mental states of ancient people should get a vote, but possibly modern nations project themselves into the past too much.
As for The Time Machine future being "troglodytes and geeks", it's funny but perhaps not entirely wrong to think of geeks as Eloi, or at least the production-consumption-and-discussion of art as Eloi behavior.
I don't think it's that people are thinking in terms of past mental states. The connection shared by members of a tribe, transcending both space and time, is generally understood to be spiritual at least as much as psychological. As most people perceive things, if I'm defending my nation's past, I'm also defending its present -- if nothing else, its present reputation and dignity.
They aren't thinking in terms of past mental states, but I've made some art and I thought it might be worth thinking about a maker point of view.
What I've done is unlikely to survive or be remembered for 50 years let alone millennia, but in my imagination, I wouldn't care about it being in the hands of whatever unimaginable culture holds what is now the middle Atlantic states. It would be nice if humans cared about it.
Because we're wired to. A billion years of ruthless natural selection have ensured that PROPAGATE THE GENOME is graven into every organism above a virus.
I don't think PROPAGATE THE GENOME is the right level of abstraction. For humans, it seems to be HAVE SEX and RAISE CHILDREN and possibly HAVE GRANDCHILDREN, with the strength of the first two varying a lot among individuals. My impression is that the desire for grandchildren is connected to the desire for grandchildren.
Anyway, we've got the tech for sex without fertility, and it turns out that a lot of people prefer that. Evolution doesn't care.
If it really was PROPAGATE THE GENOME, there'd be a lot of guys sneaking their sperm into sperm banks. Instead, while it happens, it's pretty rare.
I think you're thinking too narrowly. The same basic motivation underlies people who want to "make a difference to the future" as well as people who contribute to child-rearing in a more general way than having their own children, e.g. people who go into K12 teaching, child-care, child psychology, pediatrics. It's rare that jobs that relate to improving the lot of children -- any children, not just your own -- need to pay the kind of wage premium needed for less desirable jobs, like collecting the trash or negotiating divorce settlements. These jobs are "underpaid" (relative to others of similar qualification and responsibiity) precisely because part of the "wages" are the satisfaction of internal drives (and external validation of same).
None of these things can be reduced to "have sex" and "rear children," they are all more abstract urges to forward the success of genomes related to, but not necessarily the same as, your own. For that matter, evolutionary biologists have speculated that one reason whatever genetic predispositions to homosexuality exist haven't been suppressed long ago by natural selection is that the homosexual member of a family or tribe can provide a significant *advantage* to genomes related to his own, relative to another parent. A gay uncle, say, who makes a fortune and/or becomes socially powerful (because he's undistracted by child-rearing) and who then provides wealth and social standing to his nephews is doing a big service to the genome, much more than if he simply supplied another few copies of it.
The same kind of thinking underlies the "grandmother hypothesis" for the existence of menopause: at some point, the contributions of a woman to the prospects of her genome at a more abstract level outweight the simple creation of more copies of it (i.e. having more children of her own). She can provide wealth, guidance, wisdom, and these things enhance survivability more than a few extra copies.
The original question was along the lines of "Why do we give a damn about what happens to humanity after we are dead?" My answer is "because we're wired to, because most genes that coded for purely YOLO self-centeredness got edited out umpty millions years ago." I'm aware that most people are uncomfortable with thinking their noble aspirations can be reduced to biochemistry, but that's probably just our egoism. If you asked "Why do horses give a damn about the welfare of foals?" we would not indulge in similarly elegant flights of fancy about the philosophy of horses. For that matter, the fact that we *almost all* come up with very similar motivations (and even rationalizations for them) is rather suspicious, if the origin is presumed to be original philosophical introspection. That's like imagining we (almost) all like ice cream because we've all independently analyzed its nutritional value and logically concluded it's good food -- rather than that we all have taste buds which the DNA have caused to like being bathed in sucrose.
Well, OK. I'm not entirely sure I understand the difference, honestly. (Nor do I think "plausibly" belongs in your rephrase, as Mother Nature and Father Evolution never act on merely plausible -- it either does or it doesn't, and it only gets encoded in the DNA if it actually does.)
My theory explains why sperm fraud isn't extremely common.
I admit I don't understand why it happens at all.
Perhaps we should be thinking about the pleasures of trickery. For some people, it seems like a goal in itself, and for others it might be entangled with other goals, but still a strong drive.
1. What works tends to be kept, what doesn't tends to disappear.
2. Evolution is stupid - it's just mutations. Variations on whatever currently exists. (and btw most variations are detrimental)
So the phrase "do things plausibly related" means "do things that (1) were correlated with gene propagation in the past, provided that (2) those things were discovered via mutation in the first place (e.g. a perfect memory, or a built-in biological compass, or an instinct for complex mental arithmetic, or extremely accurate toxin detection in the tongue, or the ability to hibernate, or a cryptographically secure immune system, might all be helpful for human gene propagation, but we don't have these features because mutations didn't achieve those things, and for some of those things mutation seems very unlikely to ever be able to achieve them.)
Any thoughts on the role of fundamental energy balance in mental health?
Mental illnesses can often lead to poor sleep, which in turn leads to both general misery and an energy debt that prevents the tackling of the issue in the first place.
Similarly, declining physical fitness can reduce a person's ability to handle stress and the energy they have to resolve the stressors.
Basically I'm wondering if concepts from sports science have relevance to psychology.
In Time Enough for Love, Heinlein has it that rejuvenation takes highly skilled professionals. I suspect this was at least partly because he wanted to have something for some of his characters to do, but does it actually make sense?
Also, though it was the story Heinlein wanted to tell, you might get a fairly dystopian society where whether you age and die depends on what you can afford.
I can believe rejuvenation is harder than extending life and health starting with a young person, but how hard would it be?
I talked with someone decades ago who was keeping records of his bloodwork so that when rejuvenation was possible, the doctors would know what his baseline is. He said that blood factors can vary a lot from one healthy person to another.
"There is no simple effective against death", or something like that. People grow old and decrepit for two dozen reasons simultaneously, and most of those reasons are side effects of biological processes that have good reason for existing. Most any possible countermeasure will have its own side effects, mostly ones we don't want. Trying to balance all of that even in the not-growing-old sense, so that we get a centenarian who is youthful and healthy and not e.g. tumor-ridden, is going to involve a great deal of applied medical expertise that probably isn't going to result in the same one-size-fits-all answer for everyone and probably isn't going to be automated this side of the strong-AI singularity. Adding in the requirement that we reverse aging that has already occurred, I suspect won't make it *lots* harder, but it will be hard enough to begin with. So, yeah, "anagathic internist" will be a recognized medical specialty if we get that far before the Singularity, and rejuvinative gerontology will be a more exotic specialty.
I think your answer is very solid, but just for fun let me throw out an alternate perspective, not because I think it has a greater chance of being true, but just because I think it has a nonzero chance of being true.
What if there *is* a fairly simple reason for the ravages of age? For example, it might be located in a programmed (one assumes) degradation of the immune system. We know senescent cells that escape recycling by the immune system seem to have some outsize influence on aging effects. We suspect age-related failures of the immune system are one reason cancer risk rises dramatically with the onset of late middle age. There is the Longo-group hypothesis about the connection between calorie restriction and immune system regeneration that might start to provide a mechanism for why calorie restriction extends life (at least in lower mammals).
In short, what if the only thing you really need for a substantial rejuvenation effect is to jazz up the immune system so it starts aggressively pruning senescent and abnormal cells, so you rebuild those tissues from stem cells? That might not be exceedingly hard to implement, and maybe stuff like CAR-T therapy is pointing the way. Still needs about 100 years of basic research, but it could conceivably end up being some fairly standardized therapy -- we take out a sample of your blood, do X and Y to assorted immunological cells, reinject them, and hey presto in a few months you look and act 40 years younger.
That doesn't mean you have an anti-agathic, of course. You might not even extend life by all that much -- there is plain evidence that cells have some ultimate reproductive limit when functioning normally (although the existence of immortal hybridoma lines might suggest that limit can be subverted relatively easily). But we might want to separate "rejuvenation" = "feeling 35 when you're 75" from "life extension" = "living to age 200 instead of 100." And there's definitely value in the former, even if the latter remains difficult or out of reach.
Maybe. If there are several biochemical trainwrecks happening at roughly the same time, the effect size will be unimpressive even if you manage to solve one.
The Hayflick limit is mostly irrelevant to aging, cells can produce telomerase in a variety of circumstances. AFAIR by Hayflick limit we'd live around 250 years, and by that point we'd figure out how to do away with it for good.
If there were a fairly simple reason for the ravages of age, there would fairly quickly (on evolutionary timescales) be many reasons for the ravages of age. Designing something to last centuries is *hard*. So if there's one simple thing that causes everyone to die at ~100 years, then the evolutionary process for every *other* thing is going to optimize at "designing this particular thing to last more than a hundred years is too hard and offers no payoff whatsoever, so we're not going to do that".
You're going to get e.g. joints that form with a hundred years' worth of cartilage and no good way to add more, and your One Weird Trick of aggressively pruning senescent cells isn't going to change that, and your rejuvinative gerontologists are going to need to learn a separate weird trick for adding new cartilage to joints. And many more separate weird tricks for other things that break down around the same time.
Progeria suggest to me that there's a mechanism which causes aging. I want know what happens if that mechanism is shut down, though I grant that there will still be a lot which doesn't last indefinitely.
Assuming it'll be some kind of complex gene therapy (which I'm giving 90% odds), probably a medical professional to administer it and monitor that nothing goes south will do. Can be downgraded to an intern who pretends to know what they're doing, if you relax the safety requirements.
Unimpressed on comment #1. Of course Tim Cook, CEO of a company that tries to ensure its products are perceived as luxury (i.e. not cheap), under tremendous US domestic pressure to onshore some of their production given their absurd profit margins, would try to deflect "cheap" accusations and instead suggest American incompetence. It's hard to defend massive profit margins in an interview if you're just "cheaping out on labor".
The funny thing about tooling is you don't need to use it where you build it - the overwhelming majority of tooling globally is used, manufactured, and designed at different sites, often continents apart. Tooling is just as global a market as any other product. Makes you wonder, if China is such a high cost market, why those assembly lines don't make their way somewhere cheaper?
How quick we are to accept excuses when cost optimization pressures will suffice!
I'm sorry, what pressure? Tim Cook has $200 billion cash on hand, and his products sell like ice cream on a hot day. If you can think of an effective way of pressuring a person in such a situation, I'd love to hear it. Maybe if 250 US Representatives and 60 US Senators *and* the President were all pissed at him *and* couldn't be bought off with a few $hundred million in campaign cash, there'd be some pressure that he'd actually feel.
He mentions the possibility of an AI system working great, until someone factors RSA-2048 (https://en.wikipedia.org/wiki/RSA_numbers#RSA-2048), at which point the AI realises it's in the real world and not a simulation and begins acting in line with its *real* goals.
Made me think of some of the fiction you've written, and seems like something you'd like in any case.
A small group of anecdotes, shared by my friend group across the nation in various urban/sub-urban areas.
I got robbed, door kicked in, various items taken, including credit cards and devices with logged in accounts.
Police called. Officer arrives 10 hours later. Looks around. Says "We'll get right on it". Leaves.
The same day, I locate logged in devices at a residential area near me, turned on, with ip address. Cops aren't interested, tell me to check in pawn shops for my stuff.
The next day, credit card is used at gas station. Gas station personnel say that the police can access security footage, and they have the plates of the car the stolen card was used on. Police categorically refuse to access the footage, require that I do it. I am told it is illegal for me to access the footage, I either need to get a lawyer or have the cops do it.
I find some stuff at pawn shop, owner will release copies of the probably fake id to the cops when they call. I tell the cops. Nothing. I call the pawn shop 2 weeks later; he hasn't heard anything from the cops. They didn't bother to call.
I cut my loses and just have my homeowners insurance comp me for everything, and anchor a jobsite box to my slab to lock laptops and guns and such in while I'm out and rout my NAS cables through the punch outs.
This is every story, with the details switched.
East coast friend was mugged at gunpoint; managed to get security camera footage of the dude using his card at a 7/11, cops weren't interested.
NW friend got his fancy bike stolen, got the security cam footage from his building, and the cops didn't even want it from him.
The only time I've ever seen cops do anything was being available for a guy that got fired at a jobsite, and clearing people who were too un-aesthetic from in front of big box stores.
While anecdotes are anecdotes, this has given me the feeling: The cops won't help. Calling the cops for anything other than a guy with a gun RIGHT FUCKING NOW is a waste of time.
I don't understand how 570 million for a medium metro area gets you about 20% above nothing. I'd expect to at least see cops out and about wasting my money for that amount.
I've seen versions of this comment from the Defund the Police maximalists, who actually want to defund the police. I think the counterargument/explanation would be that the cops figure your insurance will make you whole, so they're just there to provide documentation. But yeah, it's pretty weak tea.
I had a cop repeatedly gaslight me that I wasn't harmed when something was stolen from me. The thieves stole something from my bank card, and since the bank has refunded me the money the cop was "well it sounds like the bank was the only victim." The thieves charged something to a credit-card vendor (in an easily traceable way) and the cop was "well it sounds like that vendor was the only victim." I had to keep on reminding him that cash was stolen as well.
I don't like the "Defund the Policy" narrative but damned if the cops don't make it appealing. I just bitched about the susceptibility of schools being captured but at least we can opt out there. What's the escape for a do-nothing police system?
Same here (EU). Police is largely there to scare the population into obeying laws by existing. Their actual ability to enforce is nonexistent, unless the perpetrator is really careless.
It mostly works because those most likely to realize this are, in fact, really careless, and the rest is a rounding error in the grand scheme of things.
Reading Scott’s anger at people who want to deny people a chance to do better for their kids, I wanted to take some time to address why some of us who’ve spent a great deal of time in education oppose not just charter schools, but alternative forms of education outside of government run programs generally.
The basic idea is that these alternative (and sometimes improved) forms of education allow the upper class to “opt-out” of the general public system. This removes a strong direct incentive on those with the most political power to actually address educational problems and indeed you’ll often see people who use these alternative, and gated, forms of education work to reduce funding to public schools making things worse for those less well off.
To make an example of this dynamic, let’s visit a location near Scott and myself - the cities of Sausalito and Marin City inside of Marin County (near the Golden Gate bridge in California). In 2019, the city got hit with what LA Times reported was the first desegregation order in 50 years:
My family has been directly involved in the school system since the 90s so I’m aware of many of the public and not-so-public facts. The basic story was as follows:
1. The school district covered Sausalito (primarily white) and Marin City (primarily black and brown). The class makeup of Sausalito has continued to shift up over the years, from middle towards upper, while Marin City has not changed significantly from being lower class.
2. A charter school was set up that functionally only took students from Sausalito and took basically all of them, while children from Marin City continued to go to the public school.
3. Having a greater section of the population in the school district, Sausalito residents voted for officials who would allocate more resources to the charter school than the public school. The charter school would not only receive more money per student but many additional services both recorded and not (volunteer drives, waiving of fees, various services etc. can be granted without a paper trail).
I’ve watched this struggle one step removed for my entire life. Much has been made of the actual racism involved, and I can confirm it was there and it was ugly, but many of the parents who supported these policies were really just trying to get the best education possible for their own child damn the consequences. So if they could take a dollar per student from the poorer kids in the public schools and get away with giving it to their own kids, they did.
I’ve spent enough time in schools and education at this point to see that similar scenarios play out through a lot of the country, though generally the class lines are a little more pronounced and the racial ones a little less. Those with time and money available free themselves from the system, pool their substantial resources to provide an improved experience, then try to free themselves of obligations to those with less. We all end up significantly poorer for it. (I recognize some of this is institutionalized in property tax funding of schools, good essays have been written on that and I won’t repeat them here).
Rich people care deeply about their children’s education, allowing them to not encounter the same problems as everyday citizens allows them to not have to care nor expend power to improve the system for themselves in a way that also helps others. I think that’s a substantial loss for the society as a whole.
I’m sure I’d enjoy a lively debate with Scott on some of his other views on education, but that stood out to me as unaddressed.
How does the charter school filter it's applicants to mostly take the ones from Sausalito? If this part were fixed would you still see a problem? Do you think, besides demographics, the charter school is doing something special which is of benefit to its students?
Probably a different school, but Steve Sailer shared this anecdote in the Cult of Smart review comments. (If you don't know who he is, google him).
"When I had my son apply to a charter high school founded by teachers at his middle school, I was told that admission was by lottery only, but the parents had to show up in person to find out if their child was picked at random.
When I showed up to find out if my son's name was on the list, the person holding the list was his old middle school math teacher. I nervously asked if he'd been picked in the lottery? His teacher said, without looking at the list: 'Of course he did.' I asked if he could check the list just to make sure. The charter school teacher said: 'Don't worry about it. He's in.'
> The basic idea is that these alternative (and sometimes improved) forms of education allow the upper class to “opt-out” of the general public system
Damn right.
I've heard your complaint many times: people should stay in the broken system and try to fix it. Usually told to me by the people that have captured the broken system and will decide how to portion out the gains of any effort I put into fixing it.
Lots of systems are subject to capture by insiders. Escape is the safety valve.
If you spend your time worrying about people escaping East Berlin, your solution should not be adding more barbed wire to the wall. No wonder people feel anger.
1. People should not be allowed to live where they want. They should be assigned to class-mixed housing projects - without even separate apartments, maybe you can have your own bedroom, but must share kitchens/bathrooms/living rooms etc - so that the rich people have an incentive to make sure the poor people have good living conditions.
2. People should not be allowed to eat where they want. They should have to go to class-mixed cafeterias, so that the rich people have an incentive to make sure the poor people have good food.
3. People should not be allowed to work where they want. The government should assign them jobs, with some poor people and rich people on each team, to make sure that the rich people can fight for good working conditions.
4. People should not be allowed to raise their own children. The government should raise all children communally, so that rich people have an incentive to make sure the communal child-rearing conditions are good since their own children are in them.
Like, I see where you're coming from - what you say makes some sense as a mechanism of ensuring quality for poor people - but I don't understand why the poverty concern outweighs the autonomy concern for schooling but not for anything else.
In 2019, an interesting study came out about school district secession, a somewhat more drastic version of what wpow describes in Sausalito. Because Southern states have tended to organize their public school districts on a county basis due to being more thinly populated and therefore have tended to have more integrated schools post-Brown than other parts of the country, the effects of these secessions are easier to see because they're less bound up with other social phenomena like suburbanization.
One of the more toxic inheritances that we American millennials are receiving from the Boomers is the _de facto_ racial segregation that is inscribed in our urban geography. As you say, trying to control where people live is no solution to this problem. You could try to create broader school districts that cover entire metropolitan areas but that brings with it a whole host of problems from greater scale and increased bureaucracy.
This is the main point that Ta-Nehisi Coates was making in "The Case For Reparations". You don't need to go back to 1865. The harms done by postwar redlining to black Americans, which stemmed in part from the federal policy that created the 30-year fixed-rate mortgage, have continuing effects to the present day. The racial segregation that postwar suburbanization created is only getting worse because individual families' responses to these perverse incentives exacerbate segregation in the aggregate.
I agree with you that we're not going to make any reforms that constrict families' autonomy to choose where and how they want to live, nor should we. However, poor people have less autonomy. They can't move into a better school district if they can't make rent. If we can find effective means by which to close the racial wealth gap and make it easier for _any_ family to vote with their feet and move from dysfunctional jurisdictions to functional jurisdictions, then the aggregate effect of individual families' choices about where and how to live could potentially tip from exacerbating inequality to reducing inequality.
Certainly here in Illinois, the school districts' heavy reliance on property taxes tends to create positive feedback loops where losing a part of the tax base through a plant closure causes the schools to lose funding, which makes the district less attractive to young families, which suppresses residential real estate prices, which causes the schools to lose funding to inflation, u.s.w. That's true in both the suburbs and here downstate. Of course, nobody should be modeling public finance on anything we're doing here in Illinois.
Scott, this is really strawman-y. Housing, food, work and children are all private. And no one (or close enough to no one to round off) is proposing anything like ending the parallel private system. No matter what the government says, if you don't want your kid to go to school with poors, you can send them to private school.
But the charters are part of the public system, which is supposedly established by us all and for us all. So when the charters are being used as a publicly funded way to keep well-resourced upper middle class kids away from the poors, well, that's different. If you want exclusive private schools you should just send your kids to those. But I don't see any reason to pay for it with public money, and I think that we have the right to ask the public system to do things differently.
Or, to answer your points:
1. People should be allowed to live where they want, but if they want to live in public housing, they can be asked to live in certain places and even assigned neighbors according to public policy goals.
2. People should be allowed to eat where they want, but if they want to eat for free the gov't could provide cafeterias. If you show up to one of these, there are no private dining rooms if you make over a certain amount per year or can show up at an inconvenient time to make a reservation. There's just the common room that everyone shares.
3. People should be allowed to work where they want, but if you can't find work you can register at the re-established WPA, where you'll be assigned a job and a team. But you may not get to pick your job or teammates.
4. People should be allowed to raise their own children, but if they'd like to avail themselves of free daycare and afterschool programs, they can register at a central office and be assigned to providers.
While you could certainly implement any of the suggested programs differently, each of those seems reasonable to me, and each seems a better analogy than the ones you're using.
> 1. People should be allowed to live where they want, but if they want to live in public housing, they can be asked to live in certain places and even assigned neighbors according to public policy goals.
Does the principle change if my town government pursues a zoning strategy that just so happens to drive 90% of the population out of the private housing market, and into public housing?
The justification of using public services to pursue policy goals via the existence of private alternatives develops problems when the government that provides those public services is *also* the one who gets to dictate terms to its competitors. It creates a direct path to increased power by creating costly obstacles, and generally making the world worse in ways it can weather or ignore entirely.
If you force all kids to go to the same terrible schools, everyone will suffer, and the kids who actually want to learn, regardless of the family income, will suffer most. But the well-off parents will buy extra education outside of school, whereas the poor parents will just have to let their kids suffer and not learn. So the only thing you'd be achieving would be denying poor kids, no matter how talented and willing to learn, the chance to get an education.
There are way not enough charter schools, so this is already happening over most of the country anyway.
>Rich people care deeply about their children’s education, allowing them to not encounter the same problems as everyday citizens allows them to not have to care nor expend power to improve the system for themselves in a way that also helps others.
Yes, we get that you want to hold rich people's children hostage to their parents using their wealth in the manner you deem most socially beneficial. We've always understood this; you don't need to explain it.
It doesn't work. "Rich" people, and even more so the upper-middle-class people you're really going after, don't each individually have the power to substantially improve educational outcomes across the nation, or even a modest city. If you hold their children's education hostage to general educational improvements nationwide, or citywide or whatever, they will each and all of them hope that all the other people will act in a manner to generally improve educational outcomes.
And then each of them will privately defect and devote their own personal efforts to whatever it is they can do to improve the education of their own particular children. Having the wife stay home and do after-school tutoring, moving to a small town with good schools and no practical way to bus the students to the big city, paying a doctor to diagnose their kid with some bogus disability that only UMC children seem to get so that they can go to the "special" school that teaches only not-really-disabled UMC children, whatever.
Parental love is a powerful resource that can easily be channeled to improve educational outcomes in *many* schools, leaving you with the hard problem of fixing *some* broken schools. If you try to make the love of UMC parents fix all the schools, you break the connection and the parental love is refocused elsewhere and now you've got to fix *all* of the schools with the weaker tools remaining to you.
Also, it's morally wrong and abhorrent of you to decide that this is something you get to "allow" parents to do, or not. It's their choice, not yours.
Also, there's a matter of time. Children are going *through* the school system, so the the problem parents see in first grade isn't going to be solved while the kid is in first grade.
I'll make an exception for the condition of school buildings, but even that is unlikely to be fixed while the kid is still in the same school.
This comment is so uncharitable as to be non-productive. I think it fails to be either necessary or kind, particularly the first and last paragraphs.
On the substantive side, I am unclear about your claims. Are you claiming that well-off parents would help their children only in ways that did not help other students or in addition to doing so in ways that would help other students? If it is the latter, how does that run counter to wpow's argument? If the former, wouldn't that also suggest that such parents would currently not contribute to the school at large? If not, what am I missing? If so, would the theory then be falsifiable if we could show that those parents do contribute to the school at large (I'm thinking about things such as being more likely to volunteer, paying higher rates of school taxes, and contributing in forms like bake sales)? I do not have data on hand to show that this is true, but I strongly suspect that it is.
The best recent popular-science book out right now on modern biology of ageing research is Andrew Steele's (2020) "Ageless: The New Science of Getting Older Without Getting Old." I'm only a few chapters in but already confident saying that. (I study the evolutionary-theory side of ageing as a master's student.)
It's accessible, leaving unnecessary technical details unspecified (but also not giving faux-precise "lies-to-children" substitutes for them*), but covering all the right stuff at the right level for pop science: e.g. the first chapter introduces long-lived species and talks about the demography of aging (increase, roughly exponential, in the risk of death over time) and how life expectancy and the age distribution has changed over the 20th century. The second chapter talks about the evolutionary theory (Medawar's "mutation accumulation" theory, Williams' "antagonistic pleiotropy" theory, the "disposable soma" version of the antagonistic pleiotropy theory), and also corrects the fallacy in thinking aging is "just" entropy. The third chapter tells about how modern biogerontology was born out of studying the mechanisms by which life was extended in model organisms by the Calorie Restriction diet (first observed in rats in the 30s) and singe-gene longevity mutations (first seen in C. elegans nematodes in the late 80s / early 90s). Etc.
Its account of the mechanistic side of ageing (introduced in chap. 4) is based on the mainstream "hallmarks of ageing" framework (cf. https://dx.doi.org/10.1016%2Fj.cell.2013.05.039), also giving a shoutout to its dad, Aubrey de Grey's SENS approach: this categorizes those age-related changes, that we have good reason to suspect are causal, into a modest number of categories based on what kind of process they are and how they might be treated. This gives a semblance of order to the fact salad that was the biology of aging, at least useful for learning, and also likely useful if we do ever manage to treat aging in humans. Steele uses a slightly different list of categories than the usual hallmarks list (merging two and introducing two, for a total of ten hallmarks). Steele's list is:
1. Trouble in the double helix: DNA damage and mutations;
2. Trimmed telomeres;
3. Protein problems: autophagy, amyloids and adducts;
4. Epigenetic alterations;
5. Accumulation of senescent cells;
6. Power struggle: malfunctioning mitochondria
7. Signal failure;
8. Gut reaction: changes in the microbiome
9. Cellular exhaustion;
10. Defective defences: malfunction of the immune system.
I'm partway through chapter 4 now, and excited to read the rest.
---
*e.g. he talks about ageing as a roughly exponential increase in the risk of death w.r.t. age, over much of the adult lifespan. The precise version of this would mention the "Gompertz equation," and that "risk of death" in the Gompertz equation is defined as the "hazard rate" or "force of mortality" (instantaneous relative rate of decrease of the survival function; i.e. on a graph of a newborn's probability of still being alive at each age x, the hazard rate is the slope, divided by the height, times minus one); the hazard rate is not a probability, and in principle has no upper bound, though its growth in fact may or may not turn out to asymptote in some or many species. The hazard rate is sometimes conflated, either accidentally or as a lie-to-children, with the per-interval probability of death, which is a probability and has an upper bound of one. Even some textbooks accidentally say that the Gompertz equation says the probability of death grows exponentially (it couldn't possibly, at least not forever)—but Steele avoids that mistake.
I noticed a while ago that the people who got eulogies in Tai Chi Magazine tended to live into their late 80s or early 90s. Not bad, but not impressive for a philosophy which (at least in some versions) aimed at greatly extended lifespan.
In a recent Bruce Frantzis workshop, he mentioned that people who study bagua and hsing-i tend to live about 10 years longer than people who study. He attributed it to them working with their tendons as well as their ligaments.
Anyone have a sample of people who've done serious study of bagua and hsing-i?
So, Fermi paradox time. Could an inhabitable planet have a much higher proportion of mountainous terrain than earth? Or a much lower proportion of mountainous terrain?
Would that affect the odds of a technological civilization which could get out into space?
If there are lots of aliens, it's quite possible that lots of them are trying to get into space from planets they can't launch from, for various reasons. Bad geography, bad gravity, bad atmosphere, whatever.
But for it to explain the Fermi Paradox, it has to apply to all civs.
Is there any way to indicate blockquotes in a comment? Markdown's `>` prefix doesn't work, and googling variations of "substack comment syntax" is failing me.
Who are the really fascinating people you follow, who are not related to each other? To keep an open-mind and a broad perspective on things? I'm looking for real quality material.
During WWII, the U.S. shifted almost all of its clothing production towards making military uniforms, creating shortages of new civilian clothing. At the same time, worn-out or damaged uniforms at the front lines were thrown away. Why didn't the U.S. ship those garments back to the homeland again to clothe civilians? Old pants, shirts and jackets could be cleaned, patched up, and dyed to look almost new. There was no shortage of available shipping for this since the merchant ships went west across the Atlantic practically empty.
Idiocracy? Every cell just mindlessly does what its genetic code tells it to do.
The stuff about due process and arbitration was really interesting.
It reminds me of disciplinary foul-ups in other unionized workplaces with similar rights. A LOT of time, some bad employee will skate by for a while because their direct boss is conflict-averse and doesn't want to fire or discipline them - but then it gets up to their boss's boss who yells at them to get rid of this guy, who then hastily fires them without following the rules. They can then unsurprisingly get their job back in arbitration, and the company then has to pay them to leave while taking a PR hit.
I have read a few of your articles so far, and while I appreciate them, I generally find them unconvincing (even though I think I would agree with you on a lot of the facts around policing). But so far, most of your articles seem to be around defense of the status quo in policing. I'm curious whether you a) think that policing needs any kinds of reforms and b) contingent on a, what those reforms would be.
I don't understand this. In the article "accountability starts at the top" Graham seems to be saying that a combination of bad incentives and lack of accountability produces police chiefs that are ineffective at reforming their departments and providing a higher standard of policing.
With this in mind, he seems heavily in favor of police reform, just starting from the top rather than the bottom. This seems like a sensible approach to me, and while you may disagree, I don't think we can accuse him of just wanting to preserve the status quo.
I agree that an article on his ideas for reform would be great though.
I read that one, and maybe I missed it, but it wasn't really about fixing things. It was merely pointing out that leadership bears most of the responsibility for problems. While I would probably agree that leadership is part of the problem, "firing leadership" or "holding police leadership more accountable" are not reforms (in the same way that _just_ firing bad cops isn't reform, _making it easier_ to fire bad cops is reform). That's just relying on hiring good people. Any system that _only_ works when good people are hired is a bad system and needs structural changes to make it resistant/resilient to bad people.
His reply to me does have some ideas for reform (that I had missed in looking over his articles). I think I agree with them on the whole. I'm not sure I agree that those are _enough_, but I think they are good ideas.
It seems like, right now, the "only one time" is rarely true. Maybe that's just the outside impression of only seeing the bad stories.
Would it be a good idea for cops to do something like the military does and have tours? IE, part of the problem is that cops get desensitized and start seeing the people that they have the most contact with as less-than-human, with predictable results, and a way to mitigate that is to get them a damn break?
I think it's probably more often true than the media portrays, but it is definitely not _always_ true. And the current systems, in at least some precincts in some areas, make it far too difficult to deal with problem cops. This is part of the problem with the current discussions around policing. All of the media is national. But policing is local. It doesn't really make sense to talk about policing on a national level, when there is not a single description of policing that applies everywhere. Graham's portrayal of policing is probably correct for the vast majority of precincts. And a lot of the reforms talked about in the media probably aren't that necessary in many of them (although I think _some_ of these reforms should be implemented everywhere, even where problems don't currently exist; power must come with oversight). But they absolutely _are_ necessary in some. I think, from the articles I've read of his, that he tends to focus on the majority of precincts that don't have major problems and ignores the (admittedly rare, relative to media coverage) problems because they are rare. I don't think that just because these things don't happen often means we don't need to change the systems to fix them (although I also think that changes need to be local and very few reforms, with the exception of national problems like QI and drug war reforms need to be done on the national level)
I don't expect perfection. Cops are humans, and will make mistakes. And I'm more willing than most to recognize mistakes that come from policing being genuinely difficult, and am willing to forgive those. If we expect perfection then we just won't have policing at all. But while I would agree with you that those are the kinds of mistakes that happen far more than the media portrays (and lots of things the media portrays as "bad policing" are more nuanced than that), they are not always the case. And cops who abuse their power _do_ exist and don't _always_ only get away with it once, and in some precincts, there are not good mechanisms to deal with those cops. And if the problem is the leadership, as you argue, then it isn't enough to just say "well we have crappy leaders", we need to reform it so that the system gets rid of crappy leaders. Furthermore, I don't believe that "bad leaders" is always the problem (certainly not the _only_ problem), I believe that in at least some precincts, the unions have managed to get in place procedures that make it overly difficult to punish bad cops. Every union in every precinct? Of course not. Most unions in most precincts? Almost certainly not. But in some? Definitely yes. I'll agree that this is part of the problem with the national conversation around policing. Policing is local and problems are local. But the fact that not all police precincts have these problems doesn't mean we should ignore them.
I have my doubts that either elected officials or police chiefs in large urban police departments have much capacity to change the culture of those departments. Let's take the Minneapolis PD as an example as we pass the one-year anniversary of the killing of George Floyd.
“Since 1980, every mayor, including me, has had a reform agenda for the Minneapolis police,” [RT Rybak, the former three-term mayor of Minneapolis] said. “None of us has made anywhere the change that is necessary.”
https://www.theguardian.com/us-news/2021/apr/25/minneapolis-police-race-violence-justice-department-investigation
The Department of Justice has gotten involved at least two separate times over the past 20 years, including one consent decree. The current police chief was himself involved in a civil action about racial discrimination against officers which was settled in 2007 without going to trial. These top-down efforts coming from the brass seem like they are pushing on a string, which should be unsurprising because consciously modifying the culture of any large, bureaucratic organization is a tremendously difficult task. The transmission of that culture from vets to rookies doesn't seem to me to be very legible to the brass and the kinds of reforms they can implement. After all, Derek Chauvin was acting as a trainer for the other three cops on the scene where he killed George Floyd.
Does this imply that disbanding and reconstituting entire departments is sometimes the level of intervention that's required? That's a drastic step but these problems for both black cops and citizens have been tremendously recalcitrant in some cities. I'd be curious to hear about some success stories. I've heard good things about reforms in the Newark, NJ police department going back to Cory Booker's time as mayor there but have not dug into the details.
Fair enough. I do think that any reform will involve a better selection process for the people at the top, but agree there weren't many specific proposals.
I really like _Climate Shock_
I really like Weather, Macroweather, and the Climate by Lovejoy.
Here is my book review of it:
http://thechaostician.com/book-review-of-weather-macroweather-and-the-climate-by-shaun-lovejoy-2019/
I haven't read it yet, but I like Thomas Murphy's "Do The Math" blog and he wrote a textbook that I've heard is good: _Energy and Human Ambitions on a Finite Planet_ https://escholarship.org/uc/energy_ambitions
>Why did European countries develop parliaments much earlier than the rest of the world?
Honest question: did parliaments develop independently anywhere besides Europe? Because the pedantic answer seems to be "Because Europe invented them, and then the rest of the world copied the model after they saw it." But that doesn't answer the real question, which is why Europe in the first place?
Does anyone have a sense if the Nvidia GPUs and AMD CPUs that came out last year are finally available/not subject to crazy price gouging? At this age I just don't have it in me to keep too close of an eye on such things but I'd like to know if I can build a new PC anytime soon. Thanks.
No, the answer is no, and wait another year or so.
I don’t think they come down like… ever again.
They will. The bottleneck for manufacturing is fab capacity, and many countries have just started large expansion/construction projects for new fabs (to secure strategic supply). The capacity being built up is much higher than overall demand, so when those new fabs come online, there will be a glut and things will be cheap again. That will be in like 3-4 years.
From what I can tell, it will be another 6 months minimum. The demand doesn't look to decrease any time soon and increasing supply amounts to "build and staff another factory".
No. Semiconductor shortage in swing, expect things to by fucky well into 2022.
The prize for Cryptos is still higher than it was at the start of this year. So the demand from the miners will still be high. I don't see a chance for a relaxation on the GPU market as long as the BitCoin prize is above $20,000.
I'm in the same boat and man is it frustrating.
My techy friends have suggested checking the NewEgg shuffle everyday (https://www.newegg.com/product-shuffle) and there's a chance to queue for new GPUs when refreshed stock drops at Best Buy, Fry's etc. Apparently there's a discord for tracking that but I'm not that deep into the chase (yet)
Given the results in https://dolphin-emu.org/blog/2021/05/24/temptation-of-the-apple-dolphin-on-macos-m1/ , buying an M1 Mac Mini is maybe the cheapest(!) way to get a "decent" GPU right now.
At this point, buying a pre-assembled system might be the best idea, since those companies have long-term contracts, resulting in better prices and availability.
A substantial increase in production capacity seems to come in 2022, at the soonest, so unless demand falls a lot (which may happen if people stop gaming and start doing other things again), shortages may last for a while.
(hope it's ok to write about this again - Scott please delete if not)
I announced New Science (newscience.org) a week ago - a nonprofit the goal of which is to build new institutions of basic science, starting with life sciences.
The board of directors consists of me, Mark Lutter, and Adam Marblestone and we are advised by Tessa Alexanian, Tyler Cowen, Andrew Gelman, Channabasavaiah Gurumurthy, Konrad Kording, and Tony Kulesa.
If the site is exciting and *especially if you do biology*, I would love to talk to you -- alexey@newscience.org
I was excited to read about this initiative when you launched. I'm definitely dissatisfied with my experience in academic science so far (as a grad student), but I'm trying to challenge myself to go beyond this general negative feeling and isolate exactly what I would change.
Here's where I'm coming from. My strongest interests are clustered around concept and analogy formation during learning, in humans and maybe eventually in general/in AI. My strongest skills are engineering-related - software, hardware, signal processing. I entered a neuroscience PhD program because it seemed to be the best/only way to get to a place where I could learn about and work on these problems that interest me - despite the fact that my successes so far had mostly been in open source software development for the neuro community rather than the typical jobs of a scientist, designing experiments, interpreting the results and publishing. Now I'm struggling to develop these skills, and more precisely to justify spending a ton of effort on forcing myself to do things I'm naturally quite bad at when I feel like I could be happily coding away at my old research tech job or something similar, were it not a temporary stepping-stone to a PhD type thing. Or I could even see myself gradually getting to the point of being an independent researcher, but it seems like there should be a much longer period of apprenticeship/working on a team to learn how things really work before one is forced to go off in an independent direction and figure things out by reading papers and trial and error (despite the fact that other people around know how to do these things correctly and could just show you), as PhD students often are.
All this to say that I'm very enthusiastic about your stated goal of creating/expanding alternative paths for people who want to work in science but don't want to go the traditional grad student -> postdoc -> PI route. It seems to me like there should be more people with more specialized roles and more focused expertise working together on fewer problems with more impact. And the public funding system should somehow allow more slack for researchers while keeping standards very high overall (maybe again through division of labor). It sounds like these are things you want too, but I'd be interested to hear if you disagree.
If you want to reply individually, I filled out the form for the 2022 fellowship, so you can find my email there.
Thank you! Yeah, this only adds to my point about the absurd rigidity of the system. Some people started doing lab work when they were 16, some come from completely different backgrounds, but all are forced to go through the exact same path if they want to do research. I'll email you.
I'm a master's student (entering my second year) studying the evolutionary biology of aging! I don't know just what I'll have to contribute in conversation, but think the idea's interesting and admire some of the people involved + mentioned as influences (Andrew Gelman, Tyler Cowen, GWERN). My interest is piqued by the Summer Fellowship, albeit vaguely with no concrete plan of whether I'd apply or exactly what I'd do with it. I'll email!
One of my questions is: the site mentions that it's partly inspired by Cold Spring Harbor Laboratory; since CSHL already exists, in what ways is it different (or is it more of the same, since more is needed)?
Thanks! I'm inspired by the early days of CSHL but right now it's just a fairly ordinary research institute...
Fair point!
When do you plan to branch outside biology?
Uh, in 10-20 years?
What is the rationalist stance on moral luck?
I’m a rationalist and it reminds me of the eight vicissitudes… that’s all I’ve got. But you can search LessWrong and get an idea as to if other rationalists have discussed it before. https://www.reddit.com/r/Buddhism/comments/6htkkp/sutta_the_vicissitudes_of_life_world_the_8_things/?utm_source=amp&utm_medium=&utm_content=post_body
I think it's a deontological concept, and rationalists are usually consequentialist.
The concept can be relevant for different forms of utilitarianism too. Consider rule utilitarianism:
If I run a red light, there's an element of luck that changes whether a traditional utilitarian would say if it's moral (if nothing bad happens) or if it's immoral (if you have a crash). A rule utilitarian would say that the rule is "don't run red lights" because the world where everyone follows the rule is better than the world where no one follows the rule, or even the world where it is inconsistently followed. Thus a rule utilitarian would say that the action of running a red light is equally immoral whether or not you have a crash.
That is all to say, the concept of moral luck is a significant gap between traditional utilitarianism and rule utilitarianism.
I might be wrong, but I don't think a non-rule utilitarian would have no objection to running a red light if it didn't result in an accident. The expected utility calculation would still tell you it is immoral to do it.
Not always! If you can clearly see and confirm that no one is coming from either the left or the right, the expected utility of barreling through the intersection is a bit higher than that of stopping (assuming equivalent emotional valence, making time the only remaining variable of interest).
If you make a habit of barreling through stop lights when you think it's clear the expected number of accidents you will get in increases
It depends on how the form of utilitarianism is structured. Some utilitarians don’t follow an expected utility concept and instead judge after the action whether the action was moral based purely on the consequences.
It also depends on how the form of utilitarianism is structured. Some utilitarians don’t follow an expected utility concept and instead judge after the action whether the action was moral based purely on the consequences.
I don’t claim to speak for rationalists as a whole, but my position is that when evaluating how good or bad of a person someone is, the relevant thing to measure is the expected results, as those are what went into their decision making process. The actual results being different from the expected results doesn’t have any impact on their decision making process, so moral luck never plays a significant role. Of course, this does not extend to people deliberately sabotaging their own information.
There are additional layers that should be considered. For instance, due diligence on exploring the potential outcomes of decisions made. If I come up with a course of action and carry it out, thinking it will have a positive outcome, but it's obvious with minimal research that the outcome would be negative, that has to have some weight as well.
Hadn't heard of it, but I got this from wikipedia:
"Driver A, in a moment of inattention, runs a red light as a child is crossing the street. Driver A tries to avoid hitting the child but fails and the child dies. Driver B also runs a red light, but no one is crossing and only gets a traffic ticket.
If a bystander is asked to morally evaluate Drivers A and B, they may assign Driver A more moral blame than Driver B because Driver A's course of action resulted in a death. However, there are no differences in the controllable actions performed by Drivers A and B."
This seems to be mostly a measurement problem. Perhaps it's easy to miss a red light at a vacant intersection, but harder to miss a child in front of your car. If the child had been in the intersection in front of Driver B, would they still have run the red light and hit them?
How can you know that for sure?
In the hypothetical world where we could know the solution to this counterfactual, and also where we could reliably identify everyone in a given negligence class: yes, sure, we should punish all drivers in this negligence class equally.
But in the world we live in, we have to go by the metrics we have. It's inherently very difficult to identify everyone who's being negligent, so we assume there's a correlation between level of negligence and magnitude of bad outcome, and we assign punishment appropriately, and that's an acceptable approximation.
I think you're right about it being a measurement problem. If a driver goes through an empty intersection at 3am, we understand from seeing similar situations that the actual danger was minimal. We're willing to give the driver a pass on a "dangerous" mistake because the danger was not the same as doing it during the day when people may be walking through the intersection. By actually hitting the child, they have proven that they were maximally negligent (barring extenuating circumstances which would have to be evaluated case by case), because hitting a child while breaking a traffic law is pretty much the worst case scenario. Swerving to miss a child and hitting a dog is bad, but we would look at that situation and agree that swerving was justified to avoid hitting the child. Swerving isn't good driving behavior, but we prefer it to the negative outcome that would otherwise have happened.
We hold your driver A culpable because he did the worst driving mistake there is. It's such a bad mistake that we would have gladly accepted the intentional creation of multiple other driving mistakes in order to avoid that result. Driver B may have been just as culpable, but there are lots of reasons to doubt that. There is no doubt about Driver A.
There is a sense of moral luck I think is important, which is the idea that you shouldn't judge someone better because they were born at a time and place when their moral choices were easy. This is a corollary to "You are personally responsible for becoming more ethical than the society you grew up in": you are responsible for that, *not* for being perfectly moral; you get no credit for being moral because you went with the flow of your society, but *neither should you take the blame*. Or at minimum not all of it. Some people who were pretty damn immoral by modern standards, were nonetheless substantially more moral than their society, and we should restrain our condemnation of those people, and maybe even praise them.
My go-to example is Thomas Jefferson. TJ did some terrible things - he owned slaves and abused some of them well beyond the baseline, notably Sally Hemmings. But take a random modern liberal and exchange their birth circumstances - RML born as the eldest son of a Virginia planter in the 1740s, TJ in a modern liberal family - and looking at the outside view, RML would probably have done worse, because most planters did. (And if we stipulate that RML still was part of the Continental Congress and later politics, they almost certainly wouldn't have tried to get slavery condemned in the Declaration of Independence, nor tried to remove its economic incentives as President.) To the extent that swapping people like that is a coherent concept, of course, which is TBF not very far. This implies we should be relatively kind to Jefferson.
I don’t know what a rationalist stance on moral luck is exactly, but I’m quite doubtful that luck, including moral luck, is a coherent concept. You might be interested in my recent book on the topic, The Myth of Luck, here: https://www.bloomsbury.com/us/the-myth-of-luck-9781350149298/
Thanks for all these fascinating replies. This is my take, https://whatiscalledthinking.substack.com/p/are-people-with-the-right-views-lucky?utm_source=substack&utm_medium=email&utm_content=share&token=eyJ1c2VyX2lkIjo0MDQzMjQsInBvc3RfaWQiOjM2NzQ3NzY4LCJfIjoidzRGQy8iLCJpYXQiOjE2MjE3ODIxMDAsImV4cCI6MTYyMTc4NTcwMCwiaXNzIjoicHViLTkzODM5Iiwic3ViIjoicG9zdC1yZWFjdGlvbiJ9.udWbD1rAjxEV47wVKS58TidZCZGJjYjeVkMTK3kDy3Y
I think of moral luck being related to having some modesty about one's model of the universe. You think some action is expected to have good or bad results, but you might not be completely right, so you make some allowance for actual outcomes.
I suppose that's how things work in the real world, rather than some ideal world where moral luck gets ignored.
"You make your own luck." I don't want to habituate myself to running red lights because I'll revert to my training at the worst possible time.
Some commentary on the Glen Weyl exchange: https://www.lesswrong.com/posts/Panh7EwfLDqRtbnEw/it-s-like-poetry-they-rhyme-lanier-and-yudkowsky-weyl-and-s-2
I don't understand the distinction you are making. True, Glen dismisses AI risk (in the rationalist sense) and Scott doesn't. This is trivially true but the real disagreement seem to be several levels deeper. My takeaway from their conversation was that Scott was obviously correct everywhere and that I didn't understand Glen at all, but that's expected given my inference distance. The linked post doesn't clarify anything for me.
I appreciate the RedLetterMedia reference in the title.
That was interesting to read, as someone who didn’t previously have much contact with Lanier or that group of people. One thing that struck me a bit oddly about the association that Weyl makes between cybernetic totalism and the rationalist movement was that a number of the points in Lanier’s half manifesto were things that the rationalist movement (to the extent that it is possible to generalize) is also against. The most prominent example was the stuff about evolutionary algorithms and the focus on complexity, which Eliezer vigorously opposed in the Sequences back in 2008-2009.
For those not well-versed in the Sequences, would someone give a tldr on what Yudkowsky had to say about evolutionary algorithms and complexity?
on complexity, the relevant chapter in Sequences is: https://www.lesswrong.com/posts/kpRSCH7ALLcb6ucWM/say-not-complexity
the point is that "complexity" is a measure of a solution, but not a blueprint for the solution. it tells you that your goal is far away, but it doesn't tell you which direction you need to go.
complexity is a cost you must pay, not something you should intentionally try to increase. saying that something is complex may be a true and useful statement, but saying "I have no idea where to begin, but I know it is complex, therefore let's do some random complex things and hope it works" only means that you have no idea and you are very unlikely to succeed.
(there is no way to edit an existing comment, right?)
even if you somehow know that the complexity of a task is 9000, it doesn't mean that a random project with complexity 9000 is likely to accomplish the task.
there are many people who would say something like: "well, we know that the complexity of the task is 9000, therefore the obvious first step is to construct a project with complexity at least 9000, and then we will use the project to do the task somehow" and it would sound meaningful to them, like they already solved the largest obstacle and the rest is merely technical details the less important people have to worry about.
but this is like being lost in a desert, knowing that the nearest oasis is 10 miles away from you, and saying "well, first I need to walk 10 miles in a random direction, and I will worry about choosing the right direction later". (only it is a desert in an eldritch non-Euclidean universe, where the number of different places with given distance from your starting position increases exponentially with the distance.)
You can edit your comment, there's a button at the top right corner of your comment that opens dropdown menu when pressed, one of the options is "edit". Not sure if there's any time limit on that though.
I've never seen any such button. I'm using Chrome.
What are you using? I've never seen that.
Thanks. That's rather disappointing, though, since Yudkovsky seems to be misreading what smart people mean by "complex systems" (in Russian, they are called "difficult systems"), which is some combination of non-linearity, sensitivity to initial conditions, feedback mechanisms, the generation of measurably complex phenomena from the interaction of many simple parts, and the study of scaling phenomena.
Here's an example of the kinds of things that people have in mind when they talk about "complexity"
https://www.routledge.com/Complexity-Entropy-And-The-Physics-Of-Information/Zurek/p/book/9780201515060
There's also Yudkowsky's complaint about "emergent phenomenon":
https://www.lesswrong.com/s/5uZQHpecjn7955faL/p/8QzZKw9WHRxjR4948
In short, these amount to "some people use a word carelessly, so I propose not to use the word at all". Not very useful.
More like "most people use the word [complexity|emergence] carelessly, and that probably includes you, if you are careless".
Does saying that something is complex/emergent allow you to make specific predictions? Or is it merely how you describe in hindsight the things that you already learned using different strategies?
(Thanks for the book recommendation, I may read it later, but of course this comment was written before reading it.)
on evolution and evolutionary algorithms:
https://www.lesswrong.com/posts/XC7Kry5q6CD9TyG4K/no-evolutions-for-corporations-or-nanodevices
evolutionary theory is based on some assumptions, such as "organisms make their copies which are not perfectly identical (otherwise we would get no new mutations), but are still very very similar to their parents (otherwise all traits would get so diluted in two or three generations that there would be no meaningful selection of them)". if you use evolution as a metaphor for something where the assumptions do not apply (e.g. nanodevices do not mutate, corporations do not copy), there is no reason to assume it would behave similarly.
https://www.lesswrong.com/posts/gDNrpuwahdRrDJ9iY/evolving-to-extinction
not only evolution doesn't care about you, it doesn't care about your species either. the stories about how evolution sacrifices the weak individuals to keep the species strong are still too naive and too anthropomorphic.
by the way, the Dark Lord also doesn't follow our preconception of evolutionary "progress" and will happily evolve dogs to cancer https://en.wikipedia.org/wiki/Canine_transmissible_venereal_tumor or humans back to apes https://www.lesswrong.com/posts/xNqeGssARAYwbNCgH/link-back-to-the-trees
https://www.lesswrong.com/posts/QsMJQSFj7WfoTMNgW/the-tragedy-of-group-selectionism
https://www.lesswrong.com/posts/KE8wPzGiX5QPotyS8/conjuring-an-evolution-to-serve-you
evolution is utterly amoral, so if you ask it to solve your problems, get ready for a really horrific solution which you probably wouldn't predict, because unless you are literally a psychopath, your brain is full of self-censoring mechanisms. but evolution does not obey these limits (and neither will artificial intelligence, unless we explicitly program it to).
Halfway through, my main reaction is "very long post needs a summary; now I jump to the end: not there either."
My summary on finishing: author influenced by both sides and presents a bit of the Lanier/Weyl point of view for a lesswrong audience. Evaluates Eliezer and Scott as not very much the sort of Bad Person who L/W are against; Ray Kurzweil is a better example.
I think that's a reasonable conclusion, though as I already knew much of the background I kind of regret the time I spent reading it. I was motivated to because Weyl's work seems admirable and Lanier seems smart, so it kind of bugs me that I feel repelled by their writing: so a version of their worldview from someone more congenial was worth a try.
I think that listing a bunch of names as Good People one should be on the side of against all the Bad People who have the opposite ideology of the manifesto writer is a poor way to think and that there's already more than enough focus on "Which side are you on?".
Also, many of the Good People listed (basically names in the "tools for thought" tradition) don't rub me the wrong way as Weyl and Lanier do, making me skeptical of them representing the major dimension of difference.
What stabilizes the population of large carnivorous mammals? Pre-humans, how did the population of wolves or polar bears or killer whales change with time? Does it follow predator-prey population cycles? If so, how large are the oscillations? Are the population growing slowly most of the time, to decrease during rare die-offs? If so, how rare are die-offs usually? What causes the stabilization? Is it that predators die of starvation when they can't find food due to competition? Or do they kill each other for territory before food supply can become an issue?
How does this generalize to the ancestral environment of humans?
The mathematical descriptions for this are usually the Lotka-Volterra equations, which give you predator-prey cycles.
https://en.wikipedia.org/wiki/Lotka%E2%80%93Volterra_equations
They have been matched to concrete populations, so you may want to look into the citations if you want to know how large the cycles are.
My problem is that we don't have any data for large carnivorous mammals hat are older than say 100 years. And those population would not be in a pre-human state anyhow. So I guess I want something more speculative. Do we think that e.g. wolf population in North America in 30.000 BC followed the Lotka-Volterra equations? If so, was it lack of prey or competition for territory that drove population decline? Or did the wolf population in North America just grown steadily from the immigration of the first wolves until humans started to hunt them down?
I work in ecology, but not specifically on these types of problems, so I'm relatively qualified, but could potentially be missing nuances or "latest research". With those caveats out of the way, here is my understanding: This is one of those questions that we are unlikely to get very good answers to until human society progresses to the point where we can essentially remove our impact from the ecosystem (if you are a sci-fi fan, think the Nox from Stargate). The realization that we simply do not, and may never, know what ecosystems looked like pre-humans is a fact of the field. My ecosystem in particular is one that was completely re-engineered by humans probably close to a hundred years before _anyone_ was studying it in a more rigorous manner than "Here is what I saw there". We study how organisms interact in the ecosystem with barely an inkling of what the actual dynamics were in the evolutionary historical system.
So, all that being said, our best theoretical ideas are the above linked equations. Prey availability would be the largest determinant. Competition for prey resources could lead to territory disputes as you mention....but in the end it's all going to boil down to prey availability (competition for food is really the main reason why a concept of territory evolves in the first place). Barring some specific niche circumstances or else some new breakthrough in population dynamics theory (which is relatively well understood field)....the Lotka Volterra equations is what we've got. If for some reason you have extremely abundant prey that are able to reproduce faster than predators can eat them (which is only stable for the _very_ short term), then you may get to a high enough predator density that disease starts to strike. But that's sort of similar to a higher tier predator. Basically, long term trends in population dynamics entirely boil down to food availability from the bottom and predator constraints from the top.
Except of course in situations where climate or environmental changes cause issues for reproduction. But usually, on non-geological time scales, we can ignore those.
So it makes sense that prey is the limiting factor in the end. If prey is scarce, bigger terretories are needed. But is what does this "feel" like for a wolf (or an ancestral human) on the downward slope of the Lotka Volterra equations? Does the experience of the wolf become mostly "I can't find any rabbits no matter how hard I hunt" or does it become mostly "I must fight a lot to keep my hunting ground protected"? Is the extra mortality mostly from starvation or mosgly from conflict?
If you are willing to invest some time, perhaps you can dig into the example of wolves and moose on Isle Royal. This is one of the success stories of the Lotka Volterra equations. The data is for a national park in which hunting was forbidden for 50 years, so it's probably as close as you can come to "left alone by humans" as we can hope for. We have data for the number of kills per wolf pack and the size of the wolf packs. The number of kills per wolf does go down with scarcity and it can become really low (0.2 in the worst years, >1.5 in very good years), so I guess they are really hungry. But the authors also mention "aggressive interactions among groups". So probably it's a mix of both.
(But I don't think that wolves usually kill each other. I am no biologist, but I would assume that the bad thing that happens to a wolf is that you are driven away from prey by other wolves, and your offspring or even yourself starve. Might be different for other species.)
This is the source of the data
https://doi.org/10.1890/0012-9658(2002)083[3003:TEOPAP]2.0.CO;2
and this is the derivation of the equations
https://doi.org/10.1111/j.1365-2656.2005.00977.x
Thank you! I'll look into it! But I'm afraid 50 years is a bit short still...
A few years ago, I’ve read an article claiming that oscillations predicted by Lotka-Volterra equations are observed to be in phase even in geographically separated regions (islands vs mainland). This suggests that there are more factors at play besides predator-prey dynamics.
I made a Substack! https://denovo.substack.com/
My first post is about CRISPR base editing targeting PCSK9 to lower cholesterol levels and prevent heart disease.
https://denovo.substack.com/p/base-editing-coming-soon-to-a-liver
More posts are in the works.
You and Derek Lowe should link to each other: https://blogs.sciencemag.org/pipeline/archives/2021/05/20/crispr-editing-in-primates
(Off-topic bad chemistry joke about your name)
Have you ever met Orthocelsus?
No, but I'm a fan of Paracelsus
Freddie's comment got me thinking, I can't help but see the synergy between crypto mining, machine learning, and computing generally as putting us in some kind of a paperclip maximizer problem, but we're maximizing compute wafers and integrated circuits. There appears to be no end in sight to how many revenue-generating things we could do with more compute. BTC in particular seems to be particularly problematic because it's just an arbitrary Molochian arms race to dedicating as much electricity and compute wafers towards the task as possible. Would BTC, if actually widely adopted, just be a new form of economic rent? Everyone merely holding onto BTC because someone will buy it from them later at a high price, and you need it because it's the reserve currency?
I should clarify, because everyone needs it but people can profit by just hodling (like land, because it's a fixed supply), wouldn't the same criticisms of Georgism apply to a BTC-denominated monetary system? Where the value of BTC can arbitrarily rise to just below the point at which doing any economic activity would no longer be worth it, and therefore it would become another rent-hole just like land?
And, if this is true, is this actually a bullish take on BTC because civilization is likely to just fall for it and deliver unto ourselves even more lackluster growth BUT pretty decent rents for the BTC hodlers?
I think people would just end up switching to BTC-backed notes, just like people historically switched from using metals directly to using metal-backed notes. So financial institutions might handle their clearing in BTC, but everyone else would be using BTC derivatives.
As Matt Levine says, "a model that I often use for cryptocurrency is that it is rediscovering traditional finance"
The explicit limit on the amount of BTC that can exist is analogous to there being a surprisingly small finite amount of gold on Earth.
The economic logic behind the stylized sequence of "X *is* currency" -> "currency is backed by X" -> "currency isn't backed" (i.e., X falls out of the system doesn't require X == gold.
BTC could remain a valuable scarce asset, but its prospects as a medium of exchange are dim.
I think my question would apply to gold as well. Does the gold standard also introduce a monetary layer which is exploited by no-value-add rent seekers? By George, anyone that needs to acquire land upon which to do their business is exploited by a rent-seeking class of land speculators. At small levels of business, this might be negligible (I could run a business through my home, for example, and consider my living expenses separate from my business expenses, and then my business is not being exploited by any land speculators.) But any business that grows large enough needs to acquire commercial/industrial real estate near a labor force and then is exploited by land speculators.
Similarly for gold, perhaps small businesses this is negligible, but if you're a large business you, at some point, needed to start concerning yourself with the medium of value-store for your cash balance. If you're under a gold standard, that means either directly or indirectly (via bank notes) acquiring literal gold. Is gold not like land because it's fungible?
Gold is not like land because it's not an input to production*, just a representation of value.
>you ... needed to start concerning yourself with the medium of value-store for your cash balance
Cash itself is a medium of value store. So are Treasuries, corporate bonds, demand deposits, CDs, accounts receivable, unsold inventory, or IOUs from friends & family. There are legitimate reasons to care about the liquidity of how you store value, but there's no requirement that gold per se ever comes into play even under a fully-backed gold standard.
*Unless you're making jewelry or precision electronics.
I don't think this makes any sense at all. Land matters because it's an actual factor of production, you need a place to put your factory, you need a place near where people live to put your store front etc. A store of value doesn't work like that.
I think I agree with you that I'm probably wrong but would you help me understand why?
It seems like, in practical terms, in order to do any production you need to acquire land, raw materials, talent, and other things, and in order to acquire those things you require *some* store of value, be it food, gold, dollars, or bitcoin, in order to barter/trade/transact with. Why is *some* store of value not also an "actual factor of production" if you practically do require it in order to do any production?
And if you grant that, then by removing one level of abstraction, we can say that for global trade it's basically required that you have US Dollars, simply because the US Navy enforces that everyone does their global trade in US Dollars. Does that not mean that US Dollars are an input to production for anybody doing global trade? Is it wrong to consider trade a part of production, or am I making some other analytic error?
Remember at the beginning of the Georgian post where he talks about how money isn't wealth, it's more like a pointer to wealth. Bitcoin is like that too
The Georgist analogy would be that mining bitcoin is like homesteading. A homesteading regime causes utility loss from people occupying land 'too soon' for the sake of title to it, but eventually all the land is claimed, just as eventually all the bitcoins are mined -- so there's a bounded amount of deadweight loss in the end. Once the bitcoin mining rate is very low, with the network paid for by transaction fees, will there be some other aspect that fits Georgism? I'm not seeing it -- should I?
To be clear, I may be the idiot here, it may not tie into Georgism at all. I do really like Georgism, but one thing I didn't like about it (which may spout from my ignorance) is this treatment of land as this unique thing. What are the specific properties of land that give it this rent-seeking economic power, and do any other things have the same properties? I'm kind of going out of my way to see if any other things in the economy also act like land ownership rights and drag down human economic productivity. Candidates in my mind are cryptos that don't provide useful auxiliary infrastructure (so BTC but not necessarily ETH), domain names (which I think satisfy it exactly, but they're just not as important because search engines act like an intelligent teleporter for the internet), and intellectual property rights.
For example imagine that computer scientists were just generally a lot less smart and no one figured out any good search algorithms like PageRank, just create some convoluted scenario to imagine that there were no search engines, but the internet was still really useful. Domain names would presumably be much more important, because people would probably try to discover new websites by typing in their name. In this world, you would expect returns on an ecommerce business to be slightly better than half of what they are in our world, because roughly half of the profit is wisely eaten up by the hodler of the domain name, who knows he can shut you down at any moment by choosing to stop renting to you and to rent to someone else with a similar ecommerce business. The users just want to buy dog food at pets dot com, they don't exactly care who is selling to them. (This thought experiment is a bit of a stretch but I think it demonstrates well-enough that other things, in theory, could be like land in the Georgian sense.)
I usually just assume that the supply of BTC is already fixed because 18/21ths of it has been. Having thought it through, though, I don't think it applies to cryptocurrencies because of the fungibility.
However, I do think this somewhat applies for intellectual property, and I think we see many of the same pernicious rent-seeking as we do with land. Intellectual property is not fungible, two patents are not interchangeable. We do have entire industries built around hodling intellectual property and charging people rent to use ideas, or suing people to prevent their use at all. And while new ideas can always be created, sometimes you need a particular solution to a problem and that might just have already been patented, and the patent hodler is in a position of extracting maximal rent from you.
Domain names fit this well.
I think intellectual property when overextended is another homesteading-like setup with the same problem of a 'land rush'. Nobody's going to write the same novel as you, at the level of granularity where a reasonably faithful translation is 'the same'; but modern copyright law goes quite a way further, and independent discovery of patents is routine. So I'd agree rent-seeking seems like a major problem in IP, but I'd lean more towards dialing back the overbroadness than to taxing it.
Well right, interestingly, taxing IP on some inherent, independently-assessed value doesn't seem to make any sense. If you think about the goal of LVT, though, it's just to make sure that landowners do not have incentives to speculatively trade land. So it would seem you want to apply the same idea to IP. I would go further and argue that we should be incentivized to share IP as far and wide as it is useful, just like a Georgian landowner is incentivized to make his land as economically productive as possible. Not sure what that looks like though.
The special property of land is that we can't make more of it. If I own a factory that turns out to be very valuable because of the things it makes, other people will build similar factories and competition will drive the price down. But if I own a piece of land that turns out to be very valuable because of its location, the amount of land in that location is fixed. Competition is limited.
There are other assets that produce "rent" in the economic sense. In New York City, you need a "medallion" to operate a taxi, and there are a limited number of them. This limits competition, so the extra money you make from the limited competition is "rent".
Bitcoin is a weird example because there's a limited supply, but it doesn't produce any income. Rent is a type of income.
As everyone easily remembers to point out, land has a fixed supply curve, that's special property A. Equally important to it's unique status is special property B, land is a "positional good". That is, an equivalent amount of land with identical physical properties down to the atomic scale might be valued entirely differently due to their relative position in space. The positionality is what let's the land speculators win- because the value of their land can increase greatly based _entirely_ off the actions of uninvolved and unrelated economic agents, without anything ever being added to the land in question.
The fixed supply curve just takes the problem and makes it a disaster. And the fact that land is an input to production takes the disaster and makes it relevant to everyone.
So BTC (and gold) isn't a Georgist good because it isn't positional and doesn't _have_ to be an input to production. So you don't care which gold you've got, and you may be able to do work without it.
And I think you could make a strong argument that BTC isn't _really_ fixed supply. Sure this protocol mathematically is, but I don't see why a new fork or just a new Bitcoin would be limited.
Yeah I think you've set me straight. What then, do you think of intellectual property? It isn't exactly a positional good, at least not in a literalist/physicalist sense, but intellectual property is quasi-positional in the sense that certain pieces of intellectual property are truly positioned somewhere in a graph of possible knowledge. Sometimes you may need to make a discovery only to find out that someone else already made it and now seeks to extract rent. This has always seemed imperfect to me. It is near-costless to share information as widely as possible; it seems like a coordination problem where everyone would be better off if some scheme of information sharing can be arranged. At the same time, treating intellectual property just like physical property that someone has to pay you to use seems to create a working incentive structure. Is some element of the intellectual property regime purely rent-seeking and holding us back, in your opinion?
Yes IP is definitely positional (position in time). Its not enough to develop an idea, you have to develop it _first_. And there is no doubt an element of our (and most) IP regimes that is purely rent seeking.
The difference to land is that the supply of IP isn't fixed- in fact that the whole scheme was devised as an attempt to incentivize the production of new useful IP. The ability to temporarily collect rent is given in exchange for finding new IP. So a georgist analysis probably isn't that applicable.
I think our current system is undoubtedly suboptimal but not necessarily fundamentally inefficient, I'd just have to see the data. In fact our biggest current check on IP rent seeking is that you are only granted this right temporarily, eventually the patents expire. As opposed to land, which currently you can hold in perpetuity. George would probably say that only being given temporary rights to extract rent from your land would work pretty well to remove the DWL and long term ills, but temporary land rights are a bad idea socio-economically, youre better off confiscating the rent.
I can't speculate much on practical policy changes I would make to our current IP system because thats so data heavy and I dont have it. The only thing I can say with reasonable certainty is that weve got a complete and total mismatch concerning the protection of arts and media, these IP are extendable way way too far.
The enormous energy use from proof-of-work is basically due to sticking with one algorithm and refusing to innovate no matter the consequences. Bitcoin is still using the first algorithm that worked. It's as if people were still flying replicas of the Wright Brother's airplane. It's not inherent to cryptocurrency, though, and hopefully competing algorithms will win out eventually.
I don't think the analogy is quite apt. For Bitcoin, the energy consumption is the goal: it can't be done any other way. This isn't necessarily a suboptimal algorithm, it is what it is. I'm not anti-BTC for energy usage purposes, I think that's a Malthusian perspective that we should avoid.
Distributed consensus is the goal. A zero-sum energy use competition is an unfortunate consequence of the first implementation chosen. I don’t know if they will work out, but proof-of-stake and Chia’s proof-of-storage look promising.
Has the rise of SpaceX & other low-cost space carriers finally made Project Thor realistic? The proposed weapons system involves dropping bus-sized tungsten rods from space onto targets- it would have most of the power of a nuclear strike, but without the fallout, radioactivity, and presumably extreme escalation of actually using a nuclear weapon. It seems like it would be particularly useful for hardened bunkers (read- Iran's underground labs, or North Korea's). Unlike a cruise missile, I don't really see how you'd defend against it.
https://en.wikipedia.org/wiki/Kinetic_bombardment
The knock against Project Thor (first proposed I believe in the 50s) is that launching that much sheer weight into orbit was always far too expensive to make sense- especially with advances in cruise missiles and long-range bombers. But costs to orbit seem to have fallen a lot, not just due to SpaceX but other low-cost launchers like India. So- could the US, Russia or China just drop big rods of tungsten outta orbit now? I'm not really clear how 'steering' would work, or even if a guidance system could survive the orbital drop.
Another one thing I always wondered about Project Thor is- could they drop smaller chunks of tungsten, maybe to specifically target say warships? Like a scooter-sized or smaller one? This would require vastly more precision to hit a ship, and I'm not clear if a smaller rod would burn up in orbit. It would be a pretty extraordinary weapon, especially if a satellite launch system could drop say 20-50 precision smaller rods against a navy (like one blockading Taiwan, say)
> without [...] presumably extreme escalation of actually using a nuclear weapon.
I doubt that very much. If you dropped an RFG with nuclear-level yield, i.e. enough to create a mushrooom cloud, I expect that would be treated as an equal escalation to dropping another fission warhead.
And for various reasons, we have decided that the ability to drop that sort of munition from orbit is deeply destabilising. SALT II article IX 1(c) was put in place for a reason.
> it would have most of the power of a nuclear strike
No, it wouldn't. The kinetic energy of something in orbit is roughly comparable to the chemical energy of its weight in *conventional* explosives or rocket fuel: that's why we can get there in two stages (and one stage is within the theoretical feasibility). At that point, you can just drop a bomb on it. You can use a ballistic missile with a conventional warhead as well.
If you solve the targeting problem, RFGs are in fact a replacement for tactical nukes. They don't have the same overall energy output, but the energy density and penetration power is far higher than a conventional explosive. RFGs are bunker-busters well beyond normal warheads.
"At that point, you can just drop a bomb on it. You can use a ballistic missile with a conventional warhead as well" I can't speak as to the rods' kinetic potential, but it seems pretty clear that they'd be much faster to deploy than a cruise missile or bomber. It still takes longer than an hour for either of those to reach some targets on Earth
Note the words "bus-sized" and "tungsten". A bus-sized solid chunk of tungsten weighs 2 kilotons, so "packs its own weight in TNT" actually does put it on par with tactical nukes.
Indeed. I was merely pointing out the maths of it all.
The way I'd do this is to simply build a giant railgun or coilgun that can fire kiloton objects at the needed ICBM velocity, and hook it into the power grid. Blacking out a few million residences for a few hours during DEFCON 1 is not a huge deal, and this way you only pay for the KE when you need it rather than spending excess in a really inefficient way (rockets) and then needing to bleed off some to deorbit.
Unpleasant hypothesis, Possibly suitable for science fiction.
So it's really expensive to put bus-sized chunks of tungsten in orbit. What's the densest material that might be feasible to mine in space? Not that mining in space is cheap.
I mean, depends on what you mean by "feasible". The stuff that's cheapest to mine in space isn't the same as the stuff that'd become profitable first; the most profitable are probably gold, platinum-group metals (including osmium and iridium, the densest metals in existence) and tellurium, as these are depleted on Earth's surface compared to their abundance in the Solar System (they sank into the core during the formation of Earth).
To elaborate - the stuff that's cheapest to mine in space is normal stuff like iron, carbon, water, etc. It'd be worth mining those (and everything else) in space for consumption in space (as anything exported from Earth has a huge delta-V tax), but gold/platinum-group/tellurium is the only stuff plausibly worth importing to Earth. Of course, given that osmium/iridium are in the platinum group, these obviously are the answer to your question.
I find if humorously ironic that Tellurium, which essentially means "Earth metal," is likely much more available in outer space.
Indeed, I have remarked on this myself at times.
A Falcon 9 costs about $60 million and can put 22,000 kg in orbit. If we loft a rocket and fuel, and assume the rocket is about 92% fuel by mass (the Saturn V ratio), then that rocket can give about 11km/s delta V to about 1800 kg, for a total cost to change velocity of about $3/(kg*m/s).
Most objects near the Earth will be traveling relative to the Earth at roughly the Earth's orbital speed, which is 30 km/s. So the cost to match velocities (i.e. bring the stuff stationary with respect to the Earth) is about $90/gram. (We're ignoring a lot of factors of 50% or so here, such as the delta-V required to get the rocket to the rock in the first place, as well as clever schemes like gravity assists.) The current retail price of pure gold is $60/gram.
Low Earth orbit has an orbital speed of about 8 km/s, which implies a cost to bring stuff to the ground of about $24/gram, but you can cheat because you can aerobrake and let the atmosphere do most of the work. The only tricky part is that if you're deorbiting something as soft and low-melting as gold you would probably lose a lot of it to ablation on the way down. On the other hand, you would spread a fine mist of pure gold over everyone under your re-entry trajectory, which seems nice.
This is mostly speculation, but here are my thoughts:
We definitely have the capability to launch that much weight into orbit. [1] For SpaceX, this seems to cost about $100 million. [2] I just typed "cost to build a nuclear weapon" into Google and the cheapest ones seem to be a few million dollars [3], so it's still not competitive, although they often put multiple weapons on a single missile / bomber / submarine that is substantially more expensive.
The energy density should be similar to conventional explosives. [4] So dropping a bus sized conventional bomb releases a similar amount of energy. Yes, bus sized conventional bombs exist. [5]
The blast wave of an explosion tends to reflect off the ground. This is why hydrogen bomb test sites don't look like volcanoes, even though they release similar amounts of energy. [6] It can be quite difficult to get an explosion to influence things far underground. This is a potential niche for Project Thor. They also strike with a lot of downward directed momentum, which should allow it hit things farther underground.
To defend against it, you need to deflect it. Hit it with something of similar size going a similar speed (or smaller & faster or bigger & slower). This wouldn't be easy, but building a missile that can deflect one would be much easier than building a missile that can launch one.
Guidance is always really hard during reentry. Even on manned spaceflights, the astronauts just sit tight and there is radio blackout from the ball of plasma surrounding the object. No guidance makes deflection much easier.
Smaller versions might burn up on reentry, but they probably won't entirely. Fairly small meteorites make it to the ground. Guidance becomes even harder because the wind can push around small things more easily. The bigger problems is that it wouldn't hit the ground that fast. The atmosphere would slow it down to its terminal velocity before hitting the ground. For humans, you hit the ground at the same speed regardless of whether you jump out of a 10 story building or if you jump from space. [7] Larger and denser things have higher terminal velocity, so it takes longer for them to reach their terminal velocity. But there is no benefit for dropping small things from space compared to dropping them from an airplane. Really small versions were used during WWI from airplanes. [8]
[1] https://en.wikipedia.org/wiki/Heavy-lift_launch_vehicle
[2] https://en.wikipedia.org/wiki/Falcon_Heavy#Launch_prices
[3] https://www.brookings.edu/what-nuclear-weapons-delivery-systems-really-cost/
[4] This 10240's argument, although they say "power" instead of "energy density".
[5] https://en.wikipedia.org/wiki/GBU-43/B_MOAB
[6] The largest hydrogen bomb tested, Tsar Bomba, released twice as much energy as Mt. St. Helens.
[7] Yes, space diving is a thing. https://en.wikipedia.org/wiki/Space_diving
[8] https://en.wikipedia.org/wiki/Flechette
" But there is no benefit for dropping small things from space compared to dropping them from an airplane. Really small versions were used during WWI from airplanes"
Yeah, that's probably fair. I was actually thinking about it more and the effective range of a 'plane' (that's a broad category) is pretty huge, I believe we could already enter low orbit just with manned planes by the 60s. The X37-B can seemingly go higher than that now, unmanned. So you could probably get most the benefit of Thor with smaller munitions & some type of plane, no need to actually use a rocket.
How do naval ships defend against munitions launched from very high up though? Just thinking out loud, like what's the standard defense for that now? Seems rather difficult to defend against, and a ship is not a fast moving target
You shoot it down. Weapons coming down from orbit are harder to hit, but not impossible. The US Navy's SM-3 is capable of hitting ballistic missiles.
Also, "don't get found" and "don't be there when the missile arrives" are valid defenses. While a ship isn't that fast compared to a plane or missile, it's pretty fast compared to *how long the missile takes to arrive.* A carrier sails at 30 knots. If your Thor weapon takes, say, 15 minutes to fall from orbit, then in the time between "carrier spotted!" and "impact," the target can move anywhere within about 8.6 miles. And you have to search that whole area of ocean with sensors small enough to fit on a missile.
Now, if you have someone constantly tracking the carrier and providing guidance, that's another story, but if you can get a plane or ship close enough to get a radar lock without getting shot, then they could just shoot a missile themselves instead of waiting for the rods from god.
You can get to space on some jets, but you can't get to orbit that way. Space =/= orbit; orbit involves not just the physical location of >100km, but titanic velocities of Mach ~25 (that's twenty-five, not two-point-five). A subsonic ramjet can't go past Mach ~5, and even a scramjet TTBOMK can't get to that speed; you need a rocket to get to orbital speed.
(You also need to circularise your orbit once you've thrust out of the atmosphere, as your initial "orbit" dips into the atmosphere where you exited. This requires thrusting in space, which by definition an air-breathing jet can't do. But this is moot because of the speed issue.)
You don't need to be in orbit to drop stuff from space. In fact, it helps not to be, since orbit is all about *not* falling to the ground.
Sure, but that's not what I was driving at (hash said "we could already enter low orbit just with manned planes by the 60s; that itself is bad terminology regardless of whether it relates to the original point).
(Also, a slight de-orbit will still have way, way more speed than something merely dropped from 150km; you don't need to lop off very much of the sideways speed to intersect the Earth.)
[4] is wrong. "Bus-sized" is volume, and while orbital speed is not immensely more than high-energy chemical reactions in terms of kJ/g (it's roughly 8x TNT; bipropellant rocket fuel tops out at 3-4x TNT), those high-energy chemicals have low densities (~1-2 g/cm^3) while tungsten has a density of 19.3 g/cm^3. You can fit a lot more grams of tungsten in a bus than you can grams of TNT.
What's the point of using on-orbit storage? If you want to drop a 500kg tungsten rod on someone, put it on an ICBM and fire it on a suborbital trajectory. It will get there in just about the same time, maybe less, since there's no guarantee your orbital depot is in the correct location in its orbit to hit your target -- you may have to wait 30-60 minutes for it to come around again.
I mean, there is the direction aspect, and if you have enough of them you can guarantee you'll always have some at the right point in orbit.
As noted above, though, I agree for pure kinetic impactors.
(On-orbit storage is militarily worthwhile for Casaba-Howitzers - they only need LoS and strike essentially instantly. Of course, that would contravene the Outer Space Treaty and be incredibly destabilising.)
Well, sure, you can put tons of satellites in orbit, but, again, what's the point? Your time to deorbit isn't likely to differ significantly from the time for a suborbital ballistic trajectory, which gets you all the advantages of a near vertical high-speed final impact, lets you launch at any time, from your own soil, and with as much oomph as you can pack into your rocket.
You probably get much better terminal trajectory characterisics, actually, since unless you're prepared to add a huge increment of delta-V to your orbital weapons -- massive fuel reserves required -- you're going to come in at a shallow angle, like all re-entering spacecraft, which gives you aiming problems as well as giving your opponent more warning and more opportunity to intercept your warhead.
I don't think a plasma weapon is very practical for space to ground use. Plasma is very high energy but low momentum, so the atmosphere should provide pretty effective defense -- the stuff will basically slow to stop as its modest momentum is absorbed, and then it will heat up a heckton of N2 to no purpose.
Atomic Rockets says it can punch through with significant (though reduced) force. I'm not well-versed enough to be able to confirm or deny.
Huh. Well, I'm a priori skeptical just because the plasma generated by an ordinary (spherically-symmetric) nuclear blast doesn't punch through much atmosphere, about 150m for a modest 50kt device, 500m for a 1Mt monster. Maybe focussing it really gets you a factor of 500 but I guess I'd have to think about that for a while. I'm also wondering how you keep your plasma collimated to such an exquisite degree it doesn't spread out to near harmlessness on its 400km journey.
Now...on the other hand, if you were to use a small nuke to propel a solid projectile downward, by ablating off the near side of it and using the rocket effect...*that* would generate some really impressive vertical velocities...if we compare to what is achieved in the second stage of a thermonuclear weapon 1000 km/s seems not implausible. That means from, say, the orbit of the ISS you can hit anything in a 1000 mile radius from the point below your orbital weapon with less than 2 seconds travel time, which I think makes defense out of the question.
But you need 1 nuke per shot, which seems very expensive.
I feel I should point out that 500m of atmosphere at 150km is not an equal shield to 500m of atmosphere at sea level; using mean sea-level pressure as an estimator of the total weight of air in a column, and crunching the numbers with sea-level density and Earth's gravity, you wind up with the atmosphere being about 9km worth of sea-level-density air.
As I said, though, I'm not the master of fluid dynamics I'd need to be to give a definite, personally-understood Yes or No answer on the overarching question here.
It doesn't solve the latency problem, where if you deploy a weapon in Low Earth Orbit and need to attack a particular target on the Earth you're going to have to wait an average of six hours for it to be in position to strike, whereas the same weapon in a suborbital rocket on Earth can hit pretty much anywhere in half an hour. You'd need on the order of twenty orbital weapons platforms to match the responsiveness of one suborbital ballistic missile, and SpaceX isn't nearly cheap enough for that to be the winning solution.
It doesn't solve the targeting problem, where we don't have any good way to hit point and/or moving targets with a projectile moving Mach 25+ through the lower atmosphere because the plasma sheath surrounding the projectile will block just about any sort of targeting sensor or command signal.
It doesn't solve the problem where people who don't understand the performance or utility of nuclear weapons will mistake orbital kinetic weapons as nuclear weapons substitutes and completely misunderstand what they are capable of. In particular, your "smaller chunks of tungsten to specifically target e.g. warships" are actually *larger* chunks of tungsten; the original was to target tanks. Which is extremely difficult but maybe worth doing if you can pull it off; trying to build nuclear-weapons substitutes out of tungsten was never sensible and never seriously proposed.
It doesn't solve the problem that you are putting major strategic assets in plain view of your enemies, in space that you don't control and that they can easily attack with 1960s technology and maybe subvert beyond your ability to regain control with more modern cleverness.
It doesn't solve the problem that if your talk of "power of a nuclear strike" isn't just plain hyperbole then you've just violated well-established international law and a treaty your own nation has almost certainly signed and ratified, so the people who notice that they can shoot these things out of the sky will be on pretty solid ground doing so preemptively and in peacetime.
If you live on Earth and you somehow do find a good reason to attack people with chunks of tungsten moving at Mach 25, you will find it in every way better to do that using suborbital rockets deployed on large cross-country trucks parked in tunnels or caves somewhere in your nation's heartland, surrounded by your own country's soldiers and commanded by buried fiber optics, than to put them in orbiting satellites. "Project Thor" was a clever idea whose cleverness vanishes if you look at it closely; it's natural habitat is science-fiction stories and pop-science articles for an audience that isn't going to think about it too much. And maybe as a weapons system for people who live in outer space, but really Rule Zero for space warfare is "don't attack Earth from space, ever".
OK thanks. I checked with a couple of people I know who run a satellite tech company, they agreed with your characterization of the number of satellites needed for full coverage. (The one guy thought, conservatively, hundreds of them). He was a bit more optimistic that SpaceX currently has the guidance system needed for re-entry.
"If you live on Earth and you somehow do find a good reason to attack people with chunks of tungsten moving at Mach 25, you will find it in every way better to do that using suborbital rockets" I wonder if an ICBM carrying a shotgun spread of smaller rods could work against, say, Chinese ships blockading Taiwan? The ICBM fires and bursts open high in the atmosphere above the target, a few dozen or more tungsten rods with crude guidance systems rain down on the ships. Tough to defend against dozens of them all falling at the same time
Wait, why is rule 0 "don't attack Earth from space"? Attacking the ground of Earth from LEO is stupid, certainly, and an orbital habitat is almost the definition of indefensible, but I don't see why a(n industrially-established) polity of space colonies (lunar, asteroidal, or on icy moons in the outer system) would be helpless against Earth. Even if they lost the bombardment match - in which they'd have a substantial efficiency advantage for kinetics, and no "strategic surface-mounted energy weapons self-destruct from bloom" problem - it's highly nontrivial to bust bunkers on low-gravity objects due to the extreme depths to which one can burrow (Ceres is probably a bit too big to dig literally all the way to the core, but 50km seems pretty easy).
Good catch on the text of the Outer Space Treaty, though; I was under the mistaken impression that it only covered NBC.
>Wait, why is rule 0 "don't attack Earth from space"?
For roughly the same reason that Rule 0 for 20th century Japan was "don't attack the United States". It's populated by a really huge number of very industrious people who would very much prefer to ignore your entire part of the universe and shouldn't be too much trouble to keep placid but they'll react very very badly and in unison to external attack and there's really an awful lot of them and they're industrious enough to have already built a planet's worth of industry. And all of your imagined martial advantages will not compensate for oh fuck where did all those missiles come from and why didn't I declare my missile-counting variable an unsigned long?
You could imagine a future where this isn't true and offworld humanity outnumbers and outguns Earth and yet somehow considers Earth to be an enemy worth attacking, but basically nobody ever does that in science fiction and it's far enough in the future of any plausible reality as to be beyond the scope of this discussion. If you have to ask, Earth is too powerful for you to attack and you're just going to die horribly if you try it. If you think you don't have to ask, then you probably should have asked. Don't do it.
Leaving aside the sf hypotheticals, my impression was that orbital nuke platforms were never developed (and it was generally assumed in the 50s and early 60s they would be) because by the early 70s it was clear that national technical means were highly stabilizing, hence there was great value to all sides in *not* forcing an ASAT arms race -- which the deployment of offensive platforms would have inevitably done. Since the military benefits of on-orbit storage are a bit marginal, the command and control issues are more painful, and the cost exorbitant, I think the general consensus was...meh, let's build the MX instead.
It did come up again in a complicated way with SDI, of course, but that initiative morphed so quickly, and the fateful year of 1991 arrived so soon after, that I don't think it materially affected the thinking.
But of course all the real discussion happened in secret, so this is just a guess.
What do you mean by "national technical means were highly stabilising"? I'm confused.
Basically means that spy satellites made it impossible to sneakily build up a powerful first-strike capability to defeat an unsuspecting enemy, so arms races were just a way to waste money and increase risk of accidents. Hence people started looking for negotiated solutions to the arms-race problem, and since orbital weapons platforms would open a whole new race track everybody agreed to not do that.
But I think the decision to not build orbital weapons platforms came well before that point. Really, I think there was never a decision *not* to build orbital weapons platforms because there was never really any serious consideration to building them in the first place. The idea was big among SF writers and futurists because drama, and the military studied it seriously because it was part of their domain and the contractors wrote up proposals in case we wanted to give them umpty-billion dollars to build the thing because that's one of the parts of greed. Then the generals came back from their studying and concluded that because of the latency and the vulnerability and the superiority of terrestrial alternatives this wasn't really a useful way of doing anything they wanted to do and it didn't look their adversaries were trying to do it either.
I'm not a UFO buff, but I was super-intrigued by the recent "60 Minutes" episode about mysterious flying objects. Afterwards, I started looking into it, and I discovered some fascinating analyses done by aviation expert Mick West. He argues that the videos probably have mundane explanations, like camera artefacts caused by triangular apertures (which makes out-of-focus objects look like pyramids) and illusions caused by parallax. For some of Mick West's debunkings, see this this video (there are also other videos like it):
https://www.youtube.com/watch?v=Le7Fqbsrrm8
Mick West's explanations seem reasonable to me, but I'm not really qualified to judge. I'm wondering if any rationalists here have looked into the UFO sightings and come to any conclusions. I also wonder if the US government might have an ulterior motive in encouraging people to (falsely?) believe in extraterrestrials visiting the earth.
I don't think it's aliens, but if it is I'd say their secrecy is good enough. We can't see them well enough to be sure what they are or figure out what they're doing.
The Navy and Pentagon people have figured it out and reached the same conclusions... The video Mick explains as a gimbal rotation is titled as "GIMBAL" by the Navy. The video he explains as a weather balloon is classified under "UAPs and Balloons" by the Navy.
As for why the government at the highest levels continues to let people believe they have confirmed these are aliens, when they have done nothing of the sort? I don't know. Why *would* they be okay with everyone suddenly more likely to think that strange things in the sky are aliens? Only the US government knows, I guess.
> If true, and they are so easy to photograph, why do all the photos look like smudges on a screen? In an era with so many high def cameras we should have good photos of them by now.
Smartphones can't actually take good photos of distant things. Try photographing a flying bird or a plane with one sometime.
> if secretly observing they are doing a really shitty job on the “secret” part.
Well, they could just be observing and not especially care if we see them or not. We don't care what animals think of us for example.
Tyler Rogoway over at The Drive has been arguing that a lot of them bear the signs of surveillance drones or balloons:
https://www.thedrive.com/the-war-zone/40054/adversary-drones-are-spying-on-the-u-s-and-the-pentagon-acts-like-theyre-ufos
That article's "ZOMG it's ignored!" contention seems dubious. As the article itself points out, the US military doesn't publicise *everything* it knows, and this seems like the sort of stuff that would be classified due to concerns of egg-on-face and don't-let-them-know-we-know, so official disinterest doesn't mean much.
IMO you need more than camera artifacts to dismiss David Fravor's account. Now he's just one story, and it's sociological data, but I would appeal to his expertise. Navy pilots have four-year degrees, usually STEM, with perfect or better vision, and are highly capable of corroborating instrument data with their eyes. If you don't believe this is a smart guy, you should really listen to his interview on Lex Fridman where he goes all over what flying one of these things is like, his time in the Navy. When you listen to him explain his whole perspective, he seems too knowledgeable, humble, and skeptical to be mistaken, and afaict he's not making any money doing this. Instead it's probably pretty reputationally costly to be doing what he's doing. I'm far more compelled by his story than Bob Lazar.
Something else that I feel like the Mick West viewpoint seems to miss is that this is not just based on the video footage that has been released.
In the Nimitz incident, for instance, not only did several pilots see the same thing the video recorded, but it was also detected by radar from both the planes and the ship. This seems to make it a lot harder to explain away as a camera artifact.
This seems to be consistent with several of the other accounts that have come out - it’s not just grainy video footage from targeting cameras, it’s video footage plus visual observation plus radar.
(I do think he is likely right about the one from a couple weeks ago of the ‘pyramids’ over a ship, though - that really does look like a video artifact.)
?? Mick West argues persuasively that the TicTac video is a weather balloon or something equally innocuous, not a "video artifact". (The on-screen data shows that it's a ~6ft object travelling at wind speed at several thousand feet -- NOT large, close to the water, and extremely fast. Fravor continues to insist the video shows something unusual, IIRC.)
You're arguing against a strawman to say that we skeptics think they're all video artifacts. I recommend actually hearing Mick West out; he makes good videos! And he's interviewed three of the Nimitz eyewitnesses, if you're interested in a more back-and-forth dialogue.
Fair enough, will check it out.
I do think that if his explanation doesn't account for the radar tracking said object moving at extremely high rates of speed, it's not a great explanation.
From the NY Times back in 2017:
https://www.nytimes.com/2017/12/16/us/politics/unidentified-flying-object-navy.html
Cmdr. David Fravor and Lt. Cmdr. Jim Slaight were on a routine training mission 100 miles out into the Pacific when the radio in each of their F/A-18F Super Hornets crackled: An operations officer aboard the U.S.S. Princeton, a Navy cruiser, wanted to know if they were carrying weapons.
“Two CATM-9s,” Commander Fravor replied, referring to dummy missiles that could not be fired. He had not been expecting any hostile exchanges off the coast of San Diego that November afternoon in 2004.
Commander Fravor, in a recent interview with The New York Times, recalled what happened next. Some of it is captured in a video made public by officials with a Pentagon program that investigated U.F.O.s.
“Well, we’ve got a real-world vector for you,” the radio operator said, according to Commander Fravor. For two weeks, the operator said, the Princeton had been tracking mysterious aircraft. The objects appeared suddenly at 80,000 feet, and then hurtled toward the sea, eventually stopping at 20,000 feet and hovering. Then they either dropped out of radar range or shot straight back up.
The radio operator instructed Commander Fravor and Commander Slaight, who has given a similar account, to investigate.
[...]
The two fighter jets then conferred with the operations officer on the Princeton and were told to head to a rendezvous point 60 miles away, called the cap point, in aviation parlance.
They were en route and closing in when the Princeton radioed again. Radar had again picked up the strange aircraft.
“Sir, you won’t believe it,” the radio operator said, “but that thing is at your cap point.”
“We were at least 40 miles away, and in less than a minute this thing was already at our cap point,” Commander Fravor, who has since retired from the Navy, said in the interview.
We don't have the radar data. I wish we did. We just have Commander Fravor's recollection of what it was.
Like I said elsewhere in this thread: the fact that the actual data we do have, the video of a different Tic-Tac incident during the same time, as unambiguously traveling at wind speed (as can be derived from the on-screen data), yet Commander Fravor is always in the media insisting that the *video* is *also* showing something unusual, is concerning to me! How am I supposed to trust his eyewitness recollections (and yes, that's all the radar data is right now, no one has the actual data) if he is constantly out in the media getting *that* part wrong?
Since it seems relevant, here's Mick West's response to David Fravor's Lex Friedman interview: https://youtu.be/fT1uRf5_dF4
Look, the Nimitz Tic-Tac *video* is unambiguously just a balloon on something else innocuous (the on-screen data shows that it's a ~6ft object travelling at wind speed at several thousand feet -- NOT large, close to the water, and extremely fast). So why do David Fravor and some of the others insist the video shows something mysterious and impossible? That calls into question their eyewitness testimony about the other TicTac sighting that occurred in the same place around the same time.
* ... balloon _or_ something else innocuous
Is substack ever gonna give us an edit button?
Most likely, never :-(
> Afterwards, I started looking into it, and I discovered some fascinating analyses done by aviation expert Mick West.
Sorry, but where did you get the idea that Mick West is an aviation expert? All I can see in his credentials is that he has his pilot's license.
My mistake. Maybe that should read, "Licensed pilot and long-time skeptic", or something like that.
No worries, I just wasn't sure if I missed something since I've been reading about this. Frankly, I'm mystified at people who dismiss actual aviation experts, like the Navy fighter pilots who dispute some of Mick's analyses. No doubt Mick is right about some things, but the personal accounts that accord with infrared and radar frankly can't be accounted for so easily.
I think that there are lots of instances where military pilots and commercial pilots misinterpreted what they were seeing. Philip J. Klass presents examples of this in his book, "UFOs Explained" (published back in 1975). It's been a while since I read the book, so I don't remember the details. More recently, a military flight crew in Chile were mystified by something that turned out to be a commercial plane (probably). https://skepticalinquirer.org/newsletter/curated-crowdsourcing-in-ufo-investigations/
I'm looking for an SSC post in which Scott mentioned how some of his clients told him horror stories from their childhood that didn't seem to affect them as much as Scott would have expected. Something how even though they had been beaten to a pulp as a kid, the reason they'd gone to the psychiatrist was, I don't know, that they had trouble sleeping due to a loud neighbor. The gist of it was that people were more resilient than we think, or that people's expectations color their trauma. "Being beaten is normal, so it was not a monumental event in my life, so I don't count it as traumatic." Am I making this up? I can't seem to find the post.
Oh man, I can't find this right now either. The closest thing I can find right now is this: https://slatestarcodex.com/2014/02/08/seeming-and-being-empathetic/ which I'm pretty sure is not what you're looking for.
> I remember, one of my first few months of internship, listening to a patient – not a PTSD patient or anything, just someone presenting with something totally different like bipolar disorder or drug addiction – explain the brutal abuse he suffered as a child. And the whole time, I was thinking “Oh god oh god this is the worst thing I’ve ever heard I want to go home and cry.”
> [...]
> Now if I get a patient like the one who told the child abuse story, I’ll be more likely to just put on my best Concerned Face and ask something like “How are you doing with that now?” And a surprising amount of the time, the patient will say “I’ve put that behind me, it’s not really an issue anymore.”
I think this was indeed what I was remembering, thank you! But, contrary to what I expected, the focus is not on how our priors influence whether or not something feels traumatic. If there is something like that, I'd like to read it.
Van Der Kolk, McFarlane and Weisaeth, Traumatic Stress:The effects of overwhelming experience on mind, body and society, Guilford press, 1996 might have information on what you’re looking for. Robert Scaer, The body bears the burden:trauma, dissociation and disease, also has interesting information.
Thank you! I will look into these.
You and your thirty friends have been imprisoned by a mad logician.
You are going to be placed in a room together, each wearing either a buttercup yellow hat or a hat or a cadmium yellow hat (I like yellow!)
Each of you will then be simultaneously and secretly given the option to guess which shade of hat you are wearing (although you are allowed to demur and not guess if you choose). If any of you guess correctly and no-one guesses incorrectly then you will be released, but if anyone guesses incorrectly, or if all thirty-one of you decline to guess, then you will be sentenced to spend the next twenty years either unable to lie or unable to tell the truth, guarding a door which may or may not have a goat behind it.
Once the trial has started you are forbidden from communicating, but you can agree a strategy in advance. What is the best chance of escape you can manage?
I'm probably missing something, but I don't see any logical way to deal with this. You simply don't have enough information. What could you do besides arbitrarily designating someone to make a wild guess?
Rot13 hint/answer: Sbe rirel pbeerpg thrff, gurer vf tbvat gb or na vapbeerpg thrff. Ohg lbh pna gel naq nterr n fgengrtl jurer ybgf bs pbzovangvbaf bs ung pbybhef erfhyg va bar pbeerpg thrff naq mreb vapbeerpg barf, naq n srj pbzovangvbaf bs ungf erfhyg va ybgf bs vapbeerpg thrffrf fvzhygnarbhfyl.
100%, because you didn't forbid us to look at our own hats. Except less because I can't distinguish those two shades of yellow reliably in practice.
Otherwise, 50%, by designating one guesser. The hats aren't determined randomly or by a process with known logic, so seeing everyone else's hat is 0 evidence as to the color of your own.
No, you can do a lot better than that. Trying to turn the "you can't do better than 50%" intuition into a formal proof, and seeing where it fails, might give you an insight into how to do so?
Whatever strategy you choose, the mad logician can listen to and choose the pessimal distribution of hats. There is no order to exploit, and if there was it could be turned against you. 50% is the best you can do.
How "simultaneous" does the guessing have to be?
That is, are we allowed to use information about whether some specific other individual has guessed when making our guess?
If so I think the strategy is something like: everyone lines up from tallest to shortest. If the tallest person sees no buttercup hats, he guesses that his hat is buttercup. If the second-tallest person sees exactly one buttercup hat and the tallest person has not guessed, then he guesses his hat is buttercup. And so on.
No, all the guesses must be made without any information about one another.
I'd be curious about the mechanics of our being able to see each others' hats but being unable to communicate. Can I not wiggle my head a bit iff the designated guesser has a butternut yellow hat? What is the room like?
Does the mad logician have access to our pre-game discussion? For instance, if we designate one person as guesser, does she know which person it is?
The mad logician has no information about your strategy; you can assume that she assigns hats uniformly at random.
Uniformly at random! Okay, that's tough - am I right to conclude that there's no relationship of any kind between what hat one person is wearing and what hat another is? Or might "at random" include cases where the logician assigns each person 75% cadmium, 25% buttercup? (If there's a possibility of the latter, then presumably one can do better than average by guessing the hat more commonly worn by others.)
Since it's random, the proportion of hats of any one kind tells you nothing about your own.
Is a random distribution random? If the 30 other hats you see are all the same shade of yellow can't you statistically infer that the distribution is not likely random?
To be "scientific", you could first look at only 15 hats, hypothesize from that what the distribution likely is, then test your hypothesis on the 15 other hats.
Not true. If the distribution was random (which the problem doesn't guarantee), you could exploit assymetries in who does the guessing. For example, if someone sees that every hat but theirs matches, they guess that their hat is the other color. 1/(2^31) of the time this would generate 31 simultaneous incorrect guesses, but 31/(2^31) of the time it would generate one solitary correct guess. Every individual will make one incorrect guess for every correct guess, but the group makes 31 correct attempts for every incorrect attempt. You can use Shannon codes to generalize that to doing substantially better, though I don't remember how far that can stretch - definitely above 75%. **IF** it's random. Which, as stated, it isn't.
Oooh that's clever.
Tatterdemalion said hats were assigned "uniformly at random," in which case the logicians assigns buttercup hats at known but constant and independent probability p. Here's what I'm thinking in that case:
If we don't know what p is, then it seems that you should suspect your hat is the same as the majority of other hats. (Population sample as evidence of the base rate.)
If you DO know what p is, then you should guess buttercup when p>.5 and cadmium when p<.5, obviously.
If you know that p=.5 and you can't do any other sneaky communication then I'm stumped.
Assume that the hats are IID Bernoulli(0.5), sorry.
I suppose one class of solutions would build on priors about what the mad logician's hat assignment algorithm is. If the logician likes to assign hats according to some secret but deterministic criteria, then the guesser can use information about others to guess how she would be assigned. (For instance, if the guesser is a man and all the other men have buttercup and all the women have cadmium, he should guess buttercup.)
This doesn't apply if the game is adversarial (logician strategically assigns hats with the goal of maximizing you losing) and the logician can know your discussion, though. I think if I were a mad logician I'd like to give my victims a chance, though, and to have deterministic criteria, so you might just want to designate the person who can hold the longest boolean formula in their head as the guesser, so to speak.
No probability distribution over hat configurations is specified, but let's assume that the logician flips a coin when deciding what colour hat to put on each person's head. By agreeing on a strategy, my friends and I are accepting that there will be some configurations of hats where we lose, and some where we win. Call the winning configurations blue and the losing configurations red.
Consider a blue configuration. Blue means that we won, which means that at least one person has guessed their hat colour correctly. If we invert that person's hat colour, then the resulting configuration is red. Therefore, for every configuration that we can colour blue, there must be a red configuration one colour change away. Let's call configurations that differ only in the colour of a single hat "neighbours".
In fact, this property is the only one necessary to devise a valid strategy: If someone gives you a colouring of configurations where each blue configuration has at least one red neighbour, you can devise a strategy that lets you win all blue configurations in the colouring. Each prisoner will look at the the colours of all the other prisoner's hats. This will allow any given prisoner to reduce the possibilities to just two neighbouring configurations: One where her hat is buttercup, and one where it's cadmium. If both configurations are coloured red, then she declines to guess. If both are coloured blue, then she likewise declines to guess. If one is coloured red, and one is coloured blue, however, then she guesses that the blue option is true. (So if her hat would be cadmium in the red configuration, and buttercup in the blue configuration, then she guesses buttercup.)
If the hat configuration chosen by the logician happens to be blue, then all prisoners will either decline to guess (if flipping their hat colour would result in another blue configuration), or will guess correctly (if flipping their hat colour would result in a red configuration). And we're guaranteed that at least one prisoner will make a guess, since every blue configuration has at least one red neighbour.
The trick now is to maximize the chances of winning by maximizing the number of configurations that we can colour blue while still ensuring that each one has at least one red neighbour.
One simple and stupid trick we can do is consider the number of cadmium hats in the configuration. If it's divisible by 3, then we colour the configuration red. If it's not, then we can colour the configuration blue. For any given blue configuration, we can make it red in one flip by either adding another cadmium hat, or removing one so that the total becomes divisible by 3. This scheme colours 715827883 out of 2147483648 possibilities red, giving a success probability extremely close to 2/3.
It's probably possible to do much better, but this at least proves that it's possible to beat a success rate of 50%.
After reading JSK's comment above, I remembered about Hamming codes [1]. Hamming codes have codewords consisting of 2^n-1 bits, and have the property that any 1 bit error in transmitting a code word is correctable. Also, all possible bitstrings of length 2^n-1 are either codewords, or can be corrected to a codeword by flipping 1 bit. That means that if we colour all the codewords red, and all the non-codewords blue, each blue configuration will be neighbours with exactly one red configuration, which is optimal. In this case, we have 31 = 2^5-1, so we want to use the n=5 Hamming code. Each red codeword has 31 blue neighbours, so the chances of success are 31/32, which is 96.875%.
[1] https://en.wikipedia.org/wiki/Hamming_code
Spot on.
This is very interesting. I'm still working through how Hamming codes solve the puzzle. Are those in the room allowed to communicate in some implicit but not explicit way? Can they que up according to what they see others wearing? Can A signal to B which hat C is wearing?
I'm having trouble thinking it through because the allowed communication rules aren't clear. Obviously one can't simply signal to another what hat they are wearing, because then it would be trivial. But I sense the answer involves being allowed to signal to others what others are wearing, if it can be answered correctly 97% of the time. Or is no communication at all involved once hats are on?
OK, without exactly working out the math, I think I get it from phi's description. Tell me if I am wrong. We all have fantastic memories, and we all gather in a circle after being madly hatted.
Based on the exact configuration of the other hats, we either answer or don't. Pretty sure this means if I'm going to answer, a necessary but not sufficient condition is that I see that more than 15 of the hats are a different shade of yellow than I think mine is, but it is more complicated than that because it depends on the specifics of what every other hat in the circle is.
No, more than 15 can't be right as a necessary condition. The configuration has to be more complicated than that or we would get too many wrong answers.
Okay, I've got a strategy that wins 75% of the time. Consider the three person case: (rot13 spoilers)
Gur fgengrtl vf vs lbh frr zber pnqzvhz ungf, thrff ohggrephc, vs lbh frr zber ohggrephc ungf thrff pnqzvhz, bgurejvfr qba'g thrff.
Guvf fgengrtl jvaf 75% bs gur gvzr, fvapr vg ybfrf va rvgure OOO be PPP, ohg jvaf va gur bgure 6 pnfrf juvpu nyy unir n "gjb naq bar" frghc: gur crefba jvgu gur havdhr pbybe jvyy thrff pbeerpgyl, juvyr gur bguref jba'g thrff.
It seems like the success of this strategy varies depending on the number of people, using that strategy with 4 people would, I think, only give a 50% success rate, so in the 4 person case, it's better to just elect 3 people to be the designated potential guessers, and so you can reduce all cases above 3 to the three person case. (Maybe 3 people and 75% isn't the maximum, though)
Solution candidate:
Perngr n fbegrq dhrhr ol univat rnpu crefba rkrphgr guvf nytbevguz:
(1) jnvg n enaqbzvmrq nzbhag bs gvzr (gb nibvq zbivat ng gur fnzr gvzr nf fbzrbar ryfr).
(2) vs lbh'er gur svefg crefba gb zbir, fgnaq va gur zvqqyr bs gur ebbz, ryfr: pynvz lbhe cynpr orsber nyy gur pnqzvhzvnaf naq oruvaq nyy gur ohggrephcf. Dhrhr nyjnlf erznvaf fbegrq guvf jnl.
(3) vs lbh pbhag gjraglavar crbcyr va sebag bs lbh, thrff pnqzvhz, ryfr thrff abguvat. Jvaf hayrff nyy ungf ner ohggrephcf.
Fvapr V syvccrq n pbva gb fryrpg gur pnqzvhz/ohggrephc beqre, ybfvat bqqf ner orgjrra svsgl creprag naq svsgl creprag gb gur cbjre bs guvegl, qrcraqvat ba ubj rivy gur ybtvpvna vf
I think I can get 75% chance. rot13:
Fgneg ol nffhzvat jr whfg unir guerr crbcyr. Ab znggre gur ehyr, rnpu crefba jvyy unir nf znal fgngrf jurer gurl thrff pbeerpgyl nf jurer gurl thrff vapbeerpgyl, ohg gurl nyfb unir gur bcgvba gb abg thrff ng nyy. Fb bhe tbny vf gb tvir qvssrerag vaqvivqhnyf qvssrerag ehyrf va n jnl fb gung gurve vapbeerpg thrffrf yvar hc ohg gurve pbeerpg thrffrf qba'g. Sbe guerr crbcyr V guvax gur bcgvzny ehyrf ner:
1) Svefg crefba thrffrf gur fnzr pbybe nf gur bgure gjb vss gur bgure gjb unir zngpuvat ungf.
2) Frpbaq crefba thrffrf gur bccbfvgr bs gur guveq crefba vs svefg naq guveq crefbaf unir qvssrerag ungf.
3) Guveq crefba thrffrf bccbfvgr bs gur frpbaq crefba vs svefg naq guveq crefbaf unir qvssrerag ungf.
Rnpu bs gurfr crbcyr jvyy thrff jebat vss jr ner va gur pnfr jurer svefg crefba'f ung qvssref sebz gur bgure gjb, naq bgurejvfr rknpgyl bar bs gur guerr jvyy thrff pbeerpgyl. Nyfb sbe gur shyy guvegl bar crbcyr, lbh pna gevivnyyl rkgraq guvf fbyhgvba gb jbex sbe gung ol whfg gryyvat rirelbar cnfg gur guveq crefba gb arire thrff naq whfg vtaber gurz. Znlor lbh pna qb orggre gbb.
Fun puzzler.
Does anyone know of a good cost benefit analysis of vaccinating teens/kids? Googling is only turning up stuff that seems a bit one-sided.
Are you meaning the covid vaccines or vaccines in general? For the latter there are decades of tests and very strict standards on what is allowed with medicine for children, even more so than for adults. So the risks are negligible.
For covid there is a very small chance that there is some kind of side effect as yet undetected that applies with kids but not adults. But I'm not aware of any particular theoretical reason to think there would be, other than the general extra precaution around medicine for children
Sorry, yeah, I meant the covid vaccines. I guess I'm interested in the question of how we know the chances of an as yet undetected side effect are very small. Perhaps relevant background: I'm not a vaccine skeptic in general, and I got the covid vaccine myself. But given that the chances of serious adverse outcomes of covid with kids are so low, it's not obvious to me that the vaccines are worth the risk. I will probably just go ahead and follow the CDC recommendation to get my kids vaccinated, but I'm just trying to do my due diligence, plus it would be good to have some arguments for convincing others.
The CDC gives you some crude data. For example here:
https://covid.cdc.gov/covid-data-tracker/#vaccinations
We find that total people fully vaccinated to date in the US 18 and over is 128 million and the number 12 and older is 130 million, from which we infer that about 2 million people aged 12-18 have been fully vaccinated.
From the VAERS system, which gives you access to (anonymized) data on adverse post-vaccine events reported to the CDC, which you can get into with a bunch of clicks starting here:
https://vaers.hhs.gov/
...if I've searched the database correctly, I find 85 adverse events of any type reported to the CDC for children aged 6-17 with the Pfizer/BionNTech vaccine, the only one authorized for teens. Bear in mind "adverse event" can mean anything from a nasty headache to death, and can be reported by anybody, not just physicians -- you can dig into the database to get descriptions of the events, sort them by severity, et cetera, if you have the time.
From this one can reasonably surmise a full course of vaccination in teens results in an adverse event of any kind 1 out of every 24,000 times. The original Pfizer Phase III study results for adolescents were announced here:
https://www.pfizer.com/news/press-release/press-release-detail/pfizer-biontech-announce-positive-topline-results-pivotal
They had 2260 adolescents in thheir data and reported no serious side effects, so that establishes a priori odds of less than 1 in 1,100 for consequential side-effects.
Note that vaccination is not generally recommended for adolescents for their own sake, but to break transmission between them and someone much more vulnerable (grandmother, 60yo neighbor with cancer, et cetera).
> But given that the chances of serious adverse outcomes of covid with kids are so low
Consider that if the virus can't gain a purchase on older populations, then adaptations that target younger populations will start appearing due to selection pressures. Ultimately, everyone should probably be vaccinated.
Of course, the same argument applies to vaccine protections, ie. perhaps adaptations that circumvent the protections will start appearing. It's an arm's race for sure.
> Consider that if the virus can't gain a purchase on older populations, then adaptations that target younger populations will start appearing due to selection pressures.
I'm not sure that follows.
Say there are 100 copies of the virus in a population, and 2 can spread better among kids at the cost of being worse at spreading among adults. Those 2 would be just as likely to spread among kids whether or not the adults were vaccinated. The virus doesn't "know" to give up on the adult strategy.
The lack of adults to provide breeding ground for mutations reduces the chance of finding that rare mutation that could spread among kids.
There's been million of cases and kids have not been good carriers for whatever reason. There have been over 100 million infections around the world and they still haven't found that mutation that would let it become worse in kids.
Just because there is a plausible mutation that something could evolve to be better doesn't mean that random search plus selection pressure is sufficient to find it. There are bottlenecks that need to be passed through.
> Say there are 100 copies of the virus in a population, and 2 can spread better among kids at the cost of being worse at spreading among adults. Those 2 would be just as likely to spread among kids whether or not the adults were vaccinated. The virus doesn't "know" to give up on the adult strategy.
You're forgetting that vaccinations change the adults' behaviour. Prior to being vaccinated, adults will be more careful to take precautions to limit spread, which limits the kids' exposure to possible mutations that may be better at infecting them.
After vaccination, some adults will relax those precautions and so those 2 of 100 will have more exposure to the populations for which they're better suited.
> Just because there is a plausible mutation that something could evolve to be better doesn't mean that random search plus selection pressure is sufficient to find it.
Right, that's why I said it's an arm's race. Investing more in arms isn't a guarantee you'll have superior military power, but it improves your odds. Analogously, investing in rapid vaccination won't guarantee we won't see adaptations that circumvent their protections, but it improves our odds of avoiding that outcome.
With my made up numbers, 2% of cases mutate into something that would be worse for kids. But we had 100million+ cases, and if 2% of mutations really did end up going into kids, I think we would've gotten it by now.
I still want people to get vaccinated to just close this out. And India still has lots of cases that could mutate into something that bypasses our current vaccines.
> given that the chances of serious adverse outcomes of covid with kids are so low
One should look at the issue systemically. Covid almost never kills kids, but without vaccinations it can spread rather freely among kids (especially after restrictions are lifted). From those kids, it could
(1) spread to unvaccinated adults like my elderly parents, who aren't taking the vaccine because Fox News and other sources told him it's deadly - https://dpiepgrass.medium.com/vaccine-risks-a-letter-to-my-father-d1419486e4f1
(2) spread to the 5-10% of people in whom the vaccine was ineffective,
(3) mutate, possibly creating new strains that can evade vaccines. At the least, this would increase costs to society by creating a need for more vaccines & vaccine drives.
(here I've implicitly leaned in a way of thinking that is big/important to me: the "what if everybody did that?" way of thinking. Realistically lots of kids will be vaccinated so maybe we can get herd immunity without your kids' participation, yet I've imagined the scenario where all parents think "vaccine isn't worth the risk for kids" and all decide not to vaccinate.)
That's probably because the facts lead to a pretty one-sided decision. Vaccines are cheap, side effects are minor and/or rare, and benefits to both the child and society tend to be large. There is perhaps not a more effective, in terms of life-years saved, medical invention than vaccines. It's probably been nearly, if not more, effective than antibiotics and the ideas of sterilization for doctors. Maybe social hygiene (ie, clean water and dealing with waste) are the only things that have saved more lives.
if you are talking specifically about COVID-19, then I'll admit that the benefits are smaller (children are very unlikely to have very bad results from COVID), and, as a new vaccine technology, an argument could be made that the risks are less well defined than usual. However, from everything I've read on the topic, we understand the mechanisms behind the vaccine well enough (we got lucky that the pandemic hit when it did, being able to make the vaccines rests on 50 years of biomedical research....these are not new concepts) that, while not impossible, long term side effects are quite a bit more unlikely than the odds that your child gets severe COVID. Since the government is paying for it, this again seems like the pros outweigh the cons.
Yeah, sorry, I'm asking specifically about the covid vaccines. I would be interested in the background information/data/reasoning behind the thought that the risk of long term side effects is less than the risk of severe covid for a kid. That seems intuitively true to me, but I have no relevant training. My thought process is that the risk with the vaccines comes from the immune response to the spike proteins, but you also get an immune response to spike proteins from getting covid. So that's a tie. But when you get covid, well, you get covid: the infection itself might have or create some sort of adverse effects. So the vaccine is just a complete win-win. My problem is that I haven't seen anyone respectable give that argument, and if it was a good one then it would have entailed that there was no reason whatsoever to pause the J&J rollout (although maybe the mechanics of that vaccine are different--the "immune reaction to spike protein" thing applies to Pfizer, but I'm not sure what others).
"And considering the complete absence of long term understanding, combined with the very mild effects if at all on children and young people, I argue for not vaccinating them."
Where are you getting the long term understanding of children and young people getting covid? Children and young people getting covid could have devastating long term effects, we just don't know.
I think this is a little confused about how immune responses are generated in the first place. It's not that the immune system recognizes the intact virus while it's circulating in the blood or lymph or something. Immune cells need to be "activated" to respond to foreign particles. (It's extremely important to keep them in check so they don't destroy your own cells.)
The way that happens is that the virus infects a cell, the cell then "presents" pieces of the invader on its surface in various holder molecules which are more less painted bright red with yellow letters that say "WHAT I'M HOLDING IS EVIL." Immune cells detect these presented antigens because of the triggering holder molecules, and then generate a response against them wherever else they find them. So the antigens are naturally bits and pieces of the virus coat. Which bits and pieces? This we don't know, but the evidence suggests if they are bits and pieces of the spike protein, this works very well.
Also bear in mind that an mRNA vaccine is rather close in how it behaves to a natural infection. It subvers the cell's ribosomes to make a bunch of foreign proteins, which are (somehow) recognized as alien weirdos and presented on the surface to warn the immune system. The important difference is that the infected cell doesn't get destroyed, because the infection is inherently self-limiting -- as soon as the alien mRNA is degraded (a few minutes) it's over.
So a natural infection would probably generate a better immune response if there are some important immune-stimulating effects that happen late in the cycle of viral infection, when the cell is in much more distress. That's certainly possible, but it's also possible all the really useful stuff happens early, when the cell is still in good shape. I mean, that would be good engineering design, for what that's worth.
Well, we don't know, is the answer to that. We don't know enough about how cells choose antigens to display, or how the immune system is stimulated by them, to know whether having the full capsid to work with (but also laboring under the handicaps of a full infection) means the final result is better. Maybe yes, maybe no, or maybe it depends on the nature of the infection.
It's not an unreasonable instinct to think the answer is "yes" but it's just an instinct, which is a long way from empirical fact. You may be looking for more certainty than is possible with the current state of knowledge. My own feeling is that I'm agnostic on the point: I don't think we know enough to even make an educated guess. But one of the valuable side-effects of the whole pandemic panic is that people are now spending a lot of time and money looking into this, so maybe we *will* known by and by.
My understanding of why experts are not worried about long term side effects is that most of the technical difficulty in creating the vaccine was getting mRNA to last long enough in the body to elicit the desired response. Which means that the active ingredients break down _very_ quickly on their own. A long term impact would require some completely novel biological effects that we haven't observed so far. Possible, but very very unlikely.
I'm not sure I understand your reasoning behind the "tie". That immune response is the entire point of the vaccine: to get that response _without_ having to actually get covid. So I don't understand why that counts as a tie? Other than the sense that yes, getting COVID _also_ gives you that immunity, but that comes with the risk (albeit low for children) of severe COVID, which the vaccine does not.
I think we agree: I should have said "that's a tie so far". Both generate an immune response to spike proteins, with the good and (currently unknown but possible) bad effects that comes with. But there could also be currently unknown bad effects of the covid infection itself. So the overall risk of currently unknown bad effects is greater for getting covid than for getting vaccinated. That "seems right" to me, but I'd be interested in an expert opinion.
> getting COVID _also_ gives you that immunity
Empirically speaking, the numbers I've seen suggest that the adenovirus vaccines give about 70% immunity, actual infection by SARC-CoV-2 gives about 80% immunity, and mRNA vaccines give about 95% immunity.
I have no idea whether anyone knows *why* the different methods of immune training give different levels of immunity, but empirically it seems to be the case, and that difference does seem like it would matter.
None of these studies I'm talking about are measuring immune response. They are looking at how many people with one type of protection end up with new infections, compared to a control population without that type of protection.
It does look like the studies of the effectiveness of prior infection are more mixed than I had thought. This study cites the one I had heard of (which found 80% protection) as well as one that finds 50% protection: https://www.jwatch.org/na53276/2021/03/26/just-how-much-protection-against-sars-cov-2-reinfection
But this other one finds 95% protection: https://www.medrxiv.org/content/10.1101/2021.04.20.21255670v1
I think there are fewer of these, particularly with matched controls, than there are studies of the vaccines themselves, so we have less specific information about the protection provided by the virus itself. But what information we do have suggests that it is at least not stronger than the protection given by the two dose mRNA vaccine protocols.
> the risk with the vaccines comes from the immune response to the spike proteins, but you also get an immune response to spike proteins from getting covid. So that's a tie
This doesn't quite seem to be right. If you get vaccinated, then you get the spike proteins, whereas your alternative to getting vaccinated isn't a deterministic covid infection, but rather the chance of a future covid infection. Depending on the future of the pandemic (both this year, and in future years), one might have different estimates about how likely that is. But unless one thinks it is well above 80% likely, I could see someone legitimately thinking that vaccination is less bad than infection, and still preferring not to get vaccinated (particularly if they think they have good ability to continue other protective strategies for the rest of their life).
I think that's really overstating both cases. The lasting immunity is to *that particular strain*. We don't know to what extent it protects you against variants, or more precisely how "varied" they have to be to evade the original immunity. This is similar to the fact that new influenzas emerge every year and your immunity to last year's version doesn't do you much good with this year's.
Same thing with the mRNA vaccine. The only reason people think you might need booster shots is to protect you against variants that can evade the original immunity, because they are sufficiently different.
It *may* be the natural immunity gives you broader protection (against more variants), but this is something that is still not known and is being actively investigated.
There's a whole lot of theory going on in this set of inferences, and no empirics. We know that a lot of things that happen in the real world with infectious diseases defy any particular theory, and that's why empirics are so important. (Otherwise we would have just started vaccinating people in March of last year.)
So first off, let me say, it seems pretty safe. But I’m going to throw a few points that at least could be brought up in a serious argument.
First off, in adults at least there appears to be a risk of a bad immune response. Some of the vaccines appear to be linked to people going into anaphylactic shock or blood clotting (the shock is, from my understanding, why you are recommended to wait 15 minutes after the vaccine). From everything I’ve read, while both of these can be bad/fatal, if you monitor symptoms they’re easily treatable. Furthermore, people are screened for highest risk of a reaction, so it seems like pretty low odds of something actually bad happening if you wait the 15 minutes and overreact to any of the symptoms of blood clotting. To note (from my non-professional understanding), these severe reactions are not reactions to the “COVID” in the vaccine, rather they are reactions to the chemicals used in the fluids and stuff.
Now then, we can get a label for the Pfizer vaccine to see the odds of your kids being miserable:
In a clinical study, adverse reactions in adolescents 12 through 15 years of age included pain at the injection site (90.5%), fatigue (77.5%), headache (75.5%), chills (49.2%), muscle pain (42.2%), fever (24.3%), joint pain (20.2%), injection site swelling (9.2%), injection site redness (8.6%), lymphadenopathy (0.8%), and nausea (0.4%).
Obviously these don’t seem fun to deal with for your kids. I’d sort of see this first as like a flu shot; they have odds of being sick for a day or two after their vaccine, but also odds of being sick from covid. But the odds of them being sick from the vaccine are real, and while I don’t think it outweighs societal benefits, it’s still worth taking into your calculus.
Finally, I don’t have the numbers on this, but my understanding is that children have low odds of both contracting COVID seriously, and of passing on a COVID infection, so society benefits less from your kids getting vaccinated than from you getting vaccinated.
These are the only three arguments I can imagine against it. For the sake of posting things on the internet, a disclaimer: This is meant as info for someone who is well informed about the benefits and risks of the vaccine and doing due diligence, and not to discourage them or there children from getting vaccinated.
I think this would be exactly the type of website that you would want to use, it is dedicated to providing neutral information to help you make your own decision. Except that it is in German. If your browser has a translation tool, it might still be helpful.
https://coronaimpfung.share-to-care.de/coronaimpfung/coronavirus/coronavirus/
For the "average" person under the age of 18, the cost/benefit result is that it's *probably* not worth it, but it's more complicated than that.
In the aggregate, the fatality risks of Covid-19 decrease smoothly along with age. There's no clear-cut point at which mortality or morbidity substantially jump. So the direct benefits of the vaccine drop as the person gets younger. Likewise, it looks like children are also less likely to spread the disease. This means that the indirect benefits are also lower.
However, the harms of vaccination (injection-site soreness, feeling terrible for a day or so) are pretty much constant. At some point there has to be a cross-over point for the harm/benefits curves. My best guess is that for the typical person, it's somewhere between 15-18.
But individuals are not aggregates, and individual values matter. A person who's healthy, doesn't interact with people who are at-risk, and has a major fear of needles might be better-off waiting. Someone the exact same age who has a few health conditions, really wants to be able to visit their grandparents undergoing chemotherapy, and has continual nightmares about dying from Covid might be better off getting vaccinated.
The best thing to do are to evaluate the life circumstances of your child, their particular interest and values, and contact your child's pediatrician for professional advice.
I'm curious to know what others think about the reporting on Bill Gates following his divorce. Here's one from Vox: "Bill Gates Will Never Be the Same" - https://www.vox.com/recode/22441627/bill-gates-scandal-divorce-epstein and an earlier article from the New York Times: "Long Before Divorce, Bill Gates Had a Reputation for Questionable Behavior" - https://www.nytimes.com/2021/05/16/business/bill-melinda-gates-divorce-epstein.html. That loaded word "questionable" implies bad judgement on his part and a call for a larger social condemnation by these journalists, who insinuate that there's more to the story. I say it seems gossipy, prudish and opportunistic.
If somebody hangs out with Epstein that seems newsworthy (if not particularly surprising from someone of Gates' echelon) and if he does I'm not sure of a more polite word for it than "questionable."
If the mayor has dinners with the local mob boss, I would be in favor of newspapers opportunistically publishing the story - that seems to me the absolute baseline of the kind of thing you'd want newspapers to do. If the paper didn't have anything on the mayor's relationship *other* than those dinners, or if the dinners themselves were from disputed sources, or whatever, expect a lot of cagey language about "questionable." But the implication isn't just an artifact of the language.
Depends on whether you see Epstein is "rich pedo guy" or "honeypot CIA plant designed to entrap wealthy people with blackmail material" or some combination of both.
These sorts of matters fall into my “None of My Business” bin. I guess I’m a bit of a Boy Scout when it comes to celebrity news.
I suspect they're about as reliable as the NYT piece on Scott - Bill Gates is exactly the sort of person nyt/vox would have it in for and go out of their way to paint in an unflattering light. Whatever your priors were on Bill Gates, you should probably just stick to them (or even slightly improve your image of him, since if there were any real smoking guns they probably would have reported those instead).
> since if there were any real smoking guns they probably would have reported those instead
Alternatively, given that he's one of the richest men in the world and therefore has access to a bear infinite supply of lawyers a reluctance to say anything concrete may be more out of fear of being sued, or otherwise facing negative consequences. There is a big gap from things that journalists are confident of on the balance of probabilities, and things that they can prove in court, particularly in cases where witnesses would be reluctant to talk.
Right. It's weak evidence - there's a lot of potential reasons the media wouldn't report a smoking gun and only one of them is "it doesn't exist" - but this non-reporting is still more likely in a world where he did nothing wrong than in a world where he did, so I'd say overall the media reports should be (weak) arguments in his favour (but mostly just stick to your priors).
I think nobody will really care in about six months, especially if he makes a pretty aggressive PR push and effort to get some new venture out there. Parts of the media might, but that's because a lot of online media has a leftist bent and dislikes billionaires on principle.
I read those articles, and I didn't think either one was a "call for a larger social condemnation". They seemed a lot like the article about Scott - it uses words like "questionable" but is overall not really negative.
I wonder if that's another case of the editor choosing the juiciest possible headline, whether it fits the article or not
Scott, any chance of something like the old classified threads making a comeback? I have a niche job posting I'd like to list, but I don't know if that would be allowed in these regular open threads.
Yeah, I should bring those back, but feel free to post it here for now. Also, there's always https://www.datasecretslox.com/index.php?board=10.0
since the main optimizer behind human achievement isn't individual competence but the evolution of behaviors we call culture, the ability to discern and harvest the most useful memes of today's multibillion megaculture is in itself invaluable.
rationality is mostly useful not for creating novel solutions (leave that to evolutions, idiot), but because it offers some good heuristics for selecting existing ones.
I gotta think about this. I’ve had the same first thought, I tend more toward: “Meme conversations are a very dense ways of communicating and hence valuable. There’s important meta information in memes and people who understand may have an advantage.” I think that signs with what you state.
The next part I don’t follow so well: Rationality doesn’t create solutions it just best aggregates and selects them?
I don’t see the connecting tissue maybe.
the connection may be that I think knowing whether a meme is useful is *hard*.
There's so many supposedly useful memes - more than you can ever absorb - and a huge portion of LessWrong style memes is basically tools for identifying BS, or for deciding when versus when not to rely on established solutions / social consensus:
Chesterton's Fence, Dissolving Questions, the Algernon Argument, Absurdity Bias ( / most of the biases), Mysterious Answers, The Toxoplasma Of Rage, Change Is Bad, Simulacrum Levels, etc
I'm curious about your model, what do you mean by memetic meta information?
We might be talking about different things. You're talking about more formal memetic ideas distilled into shorthand, I think. I’m talking about the shorthand.
I really like internet memes, not because the existing set are intelligent or deep but because they are dense packets of information. Take “Motte and Bailey” for instance. Many people have a tough time recalling which is which, yet I could show you a picture of a stone tower in a field with the word Bailey on the field and motte on the tower and you’d never again forget which is which. Then I can show you a second picture with some arguments in place of the names and the concept is solidified. Two pictures, maybe 20 words instead of hundreds of words to communicate a complex idea. Additionally, I can reach people with varying reading skills and possibly across language barriers. It’s an amazing brain hack but I think we’re only in the early days, still developing the language.
As the language and use grow, savvy memers will be able to add complexity to their memes adding symbols, links, slogans, branding, etc. to add further context. The best will be able to supercharge them. Maybe this is all obvious stuff, but I find it amazing.
One thing I’ve kind felt was actually missing from much of the Rationalist, IDW, online-intelligentsia is an ability to condense information in ways that make an impact on people in such a way that they can easily recall and communicate the concept. Scott is actually pretty good at this, but things like slogans and phrases that act as placeholders for deep concepts are important but perhaps not sufficiently dense.
Some ideas are really easy to get and have taken off, for instance, “steel man,” which I see everywhere now. People get it because it carries the phrase itself is a meta contextual reference to “straw man,” which is a known concept. Others, like, say, “Dunning-Kruger” can often be misused or misunderstood because the concept is sufficiently deep and the term not informationally dense enough to convey the nuance. Someone, somewhere could likely create a single panel meme to convey this concept better than the millions of words already written about it.
I think this is important for very smart people to figure out because if done well it can act as a bit of a corrective for the abbreviated thinking that will inevitably happen when dealing with deep concepts and can aid in the spread of important ideas. Historically, I think this was done with songs, poems, jokes and powerful anecdotes. internet memes are just the newest (also oldest?) tool.
As for your original question I think I understand better what you were asking, which is are the memes (commonly used ideas) we discuss here specifically geared toward sense and decision making and I’d say yes. New ideas are still going to come from original and creative thinkers (who can also be rationalists) but over all, rationalism and it’s memeplex is going to be about justifying decisions and orienting toward a better way of acting in the world.
/end ramble
I've seen any number of explanations for which is the motte and which is the bailey, and it doesn't stick. Maybe repetition will help, or maybe I can come up with a mnemonic that works for me.
Meanwhile, I think there's a lot to be said for tower and field. Maybe fort and field.
The unfortunate thing about motte-and-bailey in particular is that using archaic words as a metaphor to descibe a concept is obviously counter-productive, and doesn't do any favors regarding the stereotype of rationalists being pretentious nerds.
Ah! Yeah I was thinking about the more general term.
I like the thought of internet memes as dense communication. Highlights a leverage of theirs I had missed.
I started a blog a few months ago at thechaostician.com .
It contains posts on a wide variety of topics. The ones I think this community would find most interesting are the ones on book reviews ( http://thechaostician.com/category/books/ ) and on the philosophy of science ( http://thechaostician.com/tag/philosophy-of-science/ ). I also write descriptions of science for a popular audience, posts about Mormonism, and occasionally posts on history.
I am currently updating every Monday and Thursday.
I hope you can find something interesting there !
Anyone else surprised there's still IRC drama going on in the year of our Lord two thousand and twenty-one?
(moreover, is anybody *not* surprised that the ruling dynasty of Korea is still extant?)
International Rescue Committee?
International Residential Code?
Internet Relay Chat?
Internal Revenue Code?
Internet Relay Chat. Freenode is dead, long live Libera.chat.
(There was a power struggle for control of Freenode, the vast majority of the staff resigned and most of them moved over to their new Libera.chat network. Freenode is now controlled by a guy who sold his VPN company to an Israeli malware purveyor: https://restoreprivacy.com/private-internet-access-kape-crossrider/ , and bought himself into the defunct throne of Korea for some reason: https://en.wikipedia.org/wiki/Andrew_Lee_(entrepreneur) .)
https://www.metafilter.com/191507/To-the-over-30-crowd-Whats-the-saddest-way-youve-injured-yourself
There are some injuries which seem like extreme bad luck (slipped on pet poo and hit one's head on the bathtub, or stepped in a deep narrow unmarked hole and broke bones), but a lot of them are more like "sneezed wrong and screwed up my back". "Something mysterious happened while I slept and now my shoulder is fucked up."
They do tend to happen more often to people over 30, though not always.
While this is a good place to share more stories, I'm also interested in what might be going on. Coordination becomes less reliable with age? Connective tissue is weaker? Those little muscles which specialize in support and stability are weaker?
Is this sort of thing less likely to happen to athletic people? Do sports injuries make up the difference? Is anyone even keeping track of weird injuries that happen for no apparent reason in normally risk-free activities?
There's a lot of talk about life extension. Should we assume that life/health extension will simply lead to at least some degree of rejuvenation?
I agree that an uploaded human brain might well need human-style sensory input. Or maybe it could adapt to some other sort of input that's at least of comparable density.
Some of those things may be simply damage accumulated through decades.
Over 40 I have a problem with my spinal disc. Almost certainly a consequence of the sedentary lifestyle (software developer). But it doesn't necassarily imply than these days I sit *more* that I used a decade ago. It might mean that some small damages in the spinal disc tissue were added perhaps since I was a teenager, and at some moment it was too much and something broke. But the thing that "made it break" could be the last 1% of the damage, with the previous 99% caused by other things.
In my case, one day I woke up, stretched my legs... and started screaming with pain. Of course it would be silly to say that stretching my legs *caused* the problem.
> There's a lot of talk about life extension. Should we assume that life/health extension will simply lead to at least some degree of rejuvenation?
You can extend the *average* life span a little by fixing some of the causes, while not fixing others. Though extending to 200 years would probably require some rejuvenation, because many parts of the human body seems to have a shorter warranty period.
There are people who live into their nineties in good health and die after deteriorating for a little while. It's genetic, and I'm hoping there will turn out to be a way to bottle it.
So it should be possible to extend the average life span and health span something like 10 or 20 years-- that's a lot from one angle and very little from another.
Accumulated damage is worth adding to the theories about those weird accidents, though it might also be plausible that if people's aging were stopped, damage would accumulate more slowly.
Stress. As you age, your body learns to store stress in obscure parts of the body where it can't interfere with everyday life. But one day you stretch, take an odd step, or turn at an odd angle, and suddenly, that obscure part of your body is called upon to bear an unaccustomed weight or motion. Stiff and unprepared, it does not respond flexibly by transferring the weight or motion to the parts of the body that can handle it. Instead, it takes the whole load upon itself, disrupting the body's fine balance of forces and causing injury.
So for instance, 99 percent of the time, you can stretch your arms and enjoy a huge, gaping yawn. But one day, your body may be storing stress in some in a tiny pair of muscles in your jaw, and when you open wide to yawn, your jaw painfully locks.
Most commonly you are lifting furniture, a box or some other weight that your body is perfectly capable of lifting without injury, even if you lift it "wrong", but on this particular day, you resent having to move furniture or lift boxes, or your boss or spouse is criticizing your work, or you are tense about something else and your body happens to be storing stress in those back muscles you don't ordinarily use while standing or sitting, and those muscles are suddenly called upon to bear the weight of whatever you're lifting - you will soon be lying on the floor howling.
I've been wondering how much of this is actual new weakness/damage to older people, and how much is just new awareness of existing weakness/damage that younger people just ignore. Surely some of the changes are real. But some of it is probably also just that after decades of your body being largely a fixed shape and size, you notice little changes more than you do just two years after you stopped growing.
The accidents I started this thread with are presumably the revelation of weaknesses which have been building up slowly, but it's not that the weakness is directly perceptible.
I think this is from general decay of health of muscles and connective tissue. With connective tissues, you often won't notice anything is wrong until an acute injury.
While you're young and growing, tissues tend to recover fast and stay healthy and flexible. As you age, the tissues tend to stiffen and degenerate, and accumulate wear-and-tear.
Being active helps: it can help stimulate recovery processes in the tissues, and having a surplus of strength can be helpful for preventing injuries (e.g. being able to catch yourself from a fall without getting hurt). However, even being active, wear-and-tear accumulates in a lot of tissues: cartilage, tendons, and ligaments heal slowly and incompletely. Weightlifting is a somewhat serious hobby of mine, and even though I've done my best to train safely and have almost no acute injuries in 15 years, my body doesn't recover the same at 40 as it did at 25.
I think life extension is a multifaceted problem, and rejuvenating connective tissues is one of those facets. It's quite possible we'll figure out how to regrow cartilage before we can extend lifespan, and vice versa.
Carrying on from the Eurovision thread, I have to link to the Icelandic jury results from Saturday night because this is life imitating art https://www.youtube.com/watch?v=aocIlMp-WWY
Italy were the eventual winners, if anyone is interested 😀
I'm interested. It was the first time I ever saw it.
Was very surprised Italy won. I now know they were the favorites going into it, but I didn't know that while watching and didn't think they had a chance. I thought it would be between Iceland and Russia. Was very surprised Russia didn't score higher. All I can figure is that the contest is about producing maximal irony and a 1980s style glam rock band winning in 2021 was maximally ironic.
Or was it simply a good strategy to put a rock band against all those dance numbers? The dance number fans are going to divide their vote, whereas the rockers only had to choose between Italy and Finland.
Or am I trying to make to much sense of it? The jury voting seemed random. Was that mainly politics? I was totally confused by the other votes showing up on the board while the juries were voting. What were those?
Why did the juries like Switzerland so much? My girlfriend and I guessed that was one of the acts that would get zero votes because it sounded horrible to our ears, yet obviously the juries thought differently. Many of the others made sense. England and Germany were genuinely horrible, although at least Germany was a bit funny. France was fine, but that performance sure didn't seem to be in the spirit of the contest. No originality, no choreography, no fun.
I suppose Iceland was too much in the spirit of it all to win it all, too on the nose, particularly after the movie. I suppose I can see why all the 13-year-olds voting wouldn't vote for what all us Olds liked best. Yet Iceland did get a lot of votes, just not compared to the Motley Crue of Italy.
Is the British jury still upset over the cod wars with Iceland? I read what you wrote in the last open comments about Eurovision and how political the voting is, and noticed how Cypress and Greece voted for each other. Were all those votes for Switzerland due to politics or did people actually like that?
The jury vote isn't only the 12-point awarded verbally. The juries also award 1-8 and 10 points, but instead of being announced those just show up on the screen. Could that be what you mean by other votes during jury announcement?
I'll just be thankful that they've not taken that opportunity to make the already excruciating jury vote announcements even longer! Especially when they could at least be pre-recorded (the juries vote the night before) so as to remove delays andess-ups. Though Iceland jury announcement was worth the waits!
Thanks for the explanation about the jury votes. I thought that may be the case but wasn't sure after reading about the scoring system and was confused while watching whether those were jury votes or phone votes.
Agree about Iceland. That guy was hilarious!
To bring this back to a Scott theme. I honestly and naively thought the feminist power of the Russia song would go over well. It was a good song. I'm no fan of Russia politically but this song was no pro-Putin thing (unless you take a very Cowenian (Straussian)) look at it. Yet it had no resonance, despite being a great performance.
That leads me to believe the feminist thing is over, for now, internationally. Perhaps Putin even allowed it because he realized it didn't matter.
More countries with low numbers of black people had black performers than I expected. Sweden, The Czech Republic, Netherlands... but nobody seemed to care. The Swedish song wasn't horrible but it wasn't anything. The individual countries seemed to be engaging in affirmative action, in terms of the acts they put forward, but the juries didn't give a shit. Whatever was driving the politics of the jury votes, it wasn't wokism.
Israel had a black performer who I first thought was a ringer because she was so good, but it turns out she is not ringer. I was surprised Israel didn't score higher, because I thought that song and performance was really good.
San Marino, wtf? I didn't realize that was a country, the song was horrible, and when I looked it up, it's an American singer??? They should not have made the finals. But neither should have England or Germany.
Not that it really matters, but it's the UK entry, not the England entry. It was bloody awful though.
My sympathies to all the UK countries.
I have read that black, in the American sense, isn't really a thing in most other counties. Other minorities fill that role. In France in the Algerians, in Germany the Turks . . . those are the only two I have actually. I know a famous black writer, James Baldwin I think, talked about how freeing to be able to put down the burden of being BLACK for a while and just live.
In The Netherlands there is a huge American influence and the Dutch woke are definitely trying to make it all about black people. We do have a decent number of migrants from the Carribean part of the Kingdom, but there are far fewer of them than black Americans, so the claim that they should be way more prominent is really only popular where they 'pool,' which is mainly in Amsterdam.
The Carribean migrants have relatively high criminality and poverty, but not really that poor of a reputation. The ethnic group with the worst reputation are the Moroccans.
People liked the Russian song, but not enough. And I saw some commentary while liveblogging the final about "Good song, but fuck Russia" so there is definitely a political angle going.
"What kind of song will win Eurovision?" is a perennial guessing-game. There tend to be a lot of ballads anyway, for some reason, but if a ballad wins this year, next year most of the entries will be ballads.
Since a (glam)rock band won this year, I expect to see a lot more rock-type acts next year. Nobody knows what the magic formula is, which makes the Swedish interval act by the hosts in 2016, about what the perfect winning song would be, so funny: https://www.youtube.com/watch?v=aMgW54HBOS0
"More countries with low numbers of black people had black performers than I expected."
Austria - singer is Filipino-Austrian (born in Vienna to Filipino parents)
Czech Republic - singer is Angolan-Czech (born in Czechoslovakia to Czech mother and Angolan father)
Israel - singer is Ethiopian-Jewish (born in Jerusalem to Ethiopian mother and Israeli father)
Malta - singer is Nigerian-Maltese (born in Malta to Maltese mother and Nigerian father)
The Netherlands - singer is Surinamese, a former Dutch colony
Sweden - singer is Congolese refugee living in Sweden since a child
San Marino - singer is Eritrean-Italian (born in Italy to Eritrean parents); Flo Rida (the American rapper) got involved with the production and because San Marino is one of the smaller countries, they're allowed bring in outside singers as there are no rules, technically, about where the singer has to come from (they don't have to be from the country they're representing).
They have links to the countries they're singing for, I don't see why this is a huge surprise (given the amount of African-American singers and musicians, acting as role models for black youth in other countries, going into pop music doesn't seem that unlikely).
People mainly votes based on historical ties/friendships/feuds between countries.
Okay, now we delve into the history of Eurovision! (The following is a personal take, don't treat it as Gospel).
There are two really big divisions: Eurovision up to when all the Eastern European countries entered, and afterwards.
(1) The History
Eurovision started out as a relatively high-minded cultural exchange notion, as well as a test of the live broadcast technology of the day - members of the European Broadcasting Union https://en.wikipedia.org/wiki/European_Broadcasting_Union transmitted the first contest, where seven nations took part, in 1956. The ideas of the time were still working on post-war hopes of preventing yet another conflict by increasing mutual understanding and appealing to a common identity, and this was one way of doing it - through music!
The winner of one year is the host country for the next year. There's been a little bit of controversy over languages - there was a rule introduced that countries should send an entry in their native language, but then in the 70s this was relaxed and many countries choose to enter songs in English. Numbers of countries competing increased to around 20 every year, with some countries dropping out and new countries entering. This is how we get Israel, Turkey and Australia competing, even if technically they're not in Europe as such. Each competing nation holds a public song contest to select their Eurovision entry which is voted on by the national public (or however the country decides to select) and that act is then sent on to Eurovision.
The 70s were the hey-day of the first version of Eurovision, with ABBA the biggest success story, but not the only one to later have successful music careers. However, by the end of the 80s/start of the 90s, it started to feel stale. The divergence between commercial pop/rock music and the kind of songs that went to Eurovision was very great at this stage, and it didn't seem relevant any more.
Enter the second stage! After the end of the Cold War, break-up of the Soviet Union, etc. suddenly there were a lot of Eastern European countries that were anxious to be part of the whole pan-European experience, and Eurovision was one way of doing this. They started entering, and they took it very seriously. So this acted as a rejuvenation in one way, but on the other hand, a lot of the new acts were stuck in musical genres of 20 years or more before. https://eurovision.tv/countries
Eventually it settled down that everyone adjusted their expectations, it was taken semi-seriously (but neither too seriously nor not seriously enough) and it has transmogrified into a continent-wide night of extravaganza, rather than a song contest to find the best song/most representative European culture/new pop star/whatever.
So many countries were now entering that there had to be a selection process. It has now settled on two semi-finals, where the public vote for who goes forward to the grand final, and the grand final thereafter. However, there are countries that automatically qualify for the final, without having to go through the selection process:
(a) the host country
(b) the Big Five (because they pay the bulk of the funding to host Eurovision) - France, Germany, Italy, Spain and the UK
And that brings us on to the second part:
(2) The Voting
This is commonly taken to be the best part of the entire show, and again there is the old version and the new version.
You asked about "the other votes showing up on the board while the juries were voting". Traditionally, especially when the numbers competing were lower, the jury would announce every single vote cast from "one vote for X" up to the highest "12 votes for Y". (The famous "nul points" or "zero vote" was never officially announced, but you can always see by the scoreboard who gets no votes from anyone https://eurosong-contest.fandom.com/wiki/Nul_Points).
As more countries entered, this became much too long and drawn-out. So they tweaked the voting system, as well as experimenting with televoting. Right now it's a mix of "first the professional/national jury give their vote, then the public vote by phone or text or app is given".
Now the only result formally announced is "the Polish (or wherever) jury gives 12 points to North Macedonia (or wherever)". The other votes (10, 8, 7-1) are put up on the scoreboard at the same time.
You can see this year the difference between the professional (music industry, broadcasters, journalists and so forth) jury votes - they tended to go for Switzerland - and the public votes which resulted in Italy winning. There's often a lot of controversy about this, and it's often political voting.
Cyprus will *always* vote for Greece, and Greece will *always* vote for Cyprus. This goes back to the 1974 coup d'etat carried out by the Greek military to make Cyprus part of Greece, and the Turkish invasion in response, and even before (there was a UN intervention in 1963 to keep the peace between the Turkish Cypriots and the Greek Cypriots) and the island is still divided between North (Turkish) Cyprus and the Republic (Greek) of Cyprus.
Malta used to reliably vote for the UK, as it is a former British colony, but not even they gave them a vote this year.
Scandinavian countries tend to vote for each other, but not exclusively. Ditto Central and Eastern European countries. Neighbours vote for neighbours and against old enemies.
UK got nothing this year; not alone was the song not really a Eurovision winner, but you know, Brexit means Brexit 😀
The French entry was *incredibly* French; an impassioned ballad sung by a gamine in the vein of Edith Piaf/Jacques Brel, in black and white (as near as dammit)? https://www.youtube.com/watch?v=Unj9WbeLzRU
And no sooner had the Italian win been announced, than the jokes about the Italian Finance Minister going "Oh hell..." started.
I don't know if the Italian economy is in bad shape (when is it not?) but I imagine they'll somehow manage to scrape together a few euro to host the shindig next year even if they have to fudge something together https://thehill.com/opinion/finance/533974-italys-challenge-to-the-global-economy
I just want to recommend the work of comic artist Anders Nilsen to everyone. I think that this is probably a pretty weird environment to recommend his work, but what the heck? This thread is about everything right?
Here's a link:
https://www.andersbrekhusnilsen.com/
Anders draws comics but to say that they are nothing like what you might think of when you think of comic books is a wild understatement. He's an artist who never ceases to shock me with his humor and unutterable artistry.
I got the fourth comic in his book TONGUES, his current ongoing project, and just read it today. It's just next level work.
The last really big work he completed is a tome of a book called BIG QUESTIONS. Like it sounds, it evokes really big topics.
He takes big archetypal imagery and weaves them together into dreamlike stories that can be unpacked for years.
He's always been a brilliant artist but his current work feels like he's deconstructing the whole form. Like you're seeing the future.
I don't have much agenda here. I've been sitting here all afternoon kind of in the hum of how good his most recent entry in TONGUES is and just wanted to say it somewhere. I saw this thread open and thought: why not.
This is the one I just read, but you'd want to start at the beginning:
https://www.andersbrekhusnilsen.com/books-comics/tongues-4
Here's that:
https://www.andersbrekhusnilsen.com/books-comics/tongues-001
Or just go all in and order a copy of BIQ QUESTIONS. Your libraries might have it. It's pretty famous in the comics world.
Huge comic nerd here, though almost no Marvel or DC stuff. I haven’t heard of this guy before, so I’m excited to check him out. Thanks for the recommendation!
Other question: has anyone ever written about how the current use of the term "rationalist" doesn't really square with like
How they talked about it when it was a question fo Descartes (a rationalist) v. Hume (an empiricist)
As I see it (and look I could be wrong)
I see today's rationalists as more like empiricists, but not quit so pure about it as Hume. Like they see reason itself as a decent little backup to observation, but they would probably put observation above reason.
Descartes tried to shake everything out JUST from reason and Hume basically said you can't get anywhere with reason (right?)
So it seems like the current conception in the SSC/Galef/LW end of today's culture is sort of a marriage of the two?
has that been dealt with substantively anywhere?
Scott wrote about that a while ago at https://slatestarcodex.com/2014/11/27/why-i-am-not-rene-descartes/
I agree scot is *not* rene decartes. which is why i wish "rationalists" would actually rebrand as "empiricists"
Popular use of the words "rational" and "empirical" are much closer to synonyms than antonyms.
Since Kant (who is probably the origin of this potted history of the 17th and 18th century), nearly all philosophers have wanted to retain elements of both the rationalist and empiricist worldviews, while rejecting elements of each as well.
Statements like "China is keeping the Yuan undervalued" confuse me. What does that even mean?
Sounds like it should mean that you can buy more stuff with Yuan than with the exchange-rate equivalent of Dollars. But if that's the case, can't I just arbitrage that away by buying stuff with Yuan, selling it for Dollars and exchanging back?
You could, but that would mean buying it in China and selling it elsewhere, which is exactly what the Chinese government wants - they're basically subsidizing the exchange rate in order to promote exports.
are they planning a 'flip the switch' moment?
I don't think so - they probably value being a net exporter more than they value access to consumer goods.
So they're growing their industries at a cost in the present to be more competitive in the future?
Somewhat. Although this can also have a market distortionary effect that can hurt them (subsidizing exports costs them access to both consumer goods and intermediate industry inputs), so it's also not a riskless strategy even if you don't mind making consumer goods more expensive.
Assuming you’re starting with USD, how do you buy the “fairly priced”CNY?
There’s a version to this trade happening with physical goods between Shenzhen and HK.
I dunno if you can arbitrage it away, because your source of yuan is the Chinese government, and they can print and sell to you infinity yuan. So it seems to me you will run out of dollars before you can successfully bid up the price of yuan.
start with X dollars, convert to yuan, buy goods, sell goods for >X dollars.
Right. But if you (or a lot of you) want to bid the price of the yuan up to its "real" value you need to keep repeating this cycle, over and over again, and trust that your increased demand for yuan at step #2 eventually results in a higher price for yuan. Normally it would, but when the "seller" (the Chinese government) of yuan has essentially a zero marginal cost of production, they can sell you infinity yuan (as far as you're concerned) without raising the price. So long before you succeed in raising the price of the yuan, you run out of dollars. Well, more precisely, the price of your goods in dollars falls as you increase supply until step #4 no longer gets you >X dollars.
your last statement proves the point. in the above, i sell goods for >X dollars, so i will never run out. if the price of goods in dollars falls such that the cost is equivelant in dollars and yuan, then i have succesfully equalized the value.
I don’t know if this has come up here in the past. For people with strong feelings that they know the ‘best’ political/economic system to live under, they should consider Warren Buffet’s ‘Ovarian Lottery’ thought experiment before they go all in on it.
https://www.google.com/amp/s/www.businessinsider.com/warren-buffett-on-the-ovarian-lottery-2013-12%3famp
If the idea has already been discussed to death, well then, never mind.
Isn't this just the veil of ignorance? Which i mean, good on him for inventing, but it seems the same.
Yes i just read a quick summary of VOI and they do seem to be getting at the same idea. I think the ovarian lottery take might be a more viscerally experienced and easier to understand way to drive the idea home with someone not open to or experienced with a more academic take.
This has the same fundamental problem as Rawls' Veil of Ignorance: it depends on metaphysical dualism and falls apart entirely if humans are biological organisms instead of souls trapped within flesh.
There is no lottery involved in being your parents' child as that is the only possible option. Your laudable traits are inherited from them, including "mental" traits like intelligence or conscientiousness which depend on a physical brain, so anything they can be said to have earned through their laudable traits likewise should apply just as much to you. And vice versa for their condemnable traits.
If there is luck involved, it is at a different stage from heredity which is not a tremendously random process. Recombination smooths things out and of course there's always the chance of de novo mutations, aneuploidy, etc. but overall you're much more similar to your parents than our culture is willing to admit.
The Ovarian Lottery isn't supposed to be a real thing that exists. It's an idea to help people think altruistically.
As humans, we should care about the well-being of all humans. But it's easy to fall into only caring ourselves and those like ourselves. The point of the Ovarian Lottery idea is that it helps you consider the plight of everyone in the world. If there's a class of people that make you think "Oh, I'd hate to be one of those", then you should consider how to fix that.
Hi all,
I’m working at a job that is really dragging me down emotionally. The pay is not bad but the work is stressful and not at all stimulating or meaningful (I’d like to have at least one of those two). I’m stuck when it comes to deciding what to do with my life and I don’t think I have the financial luxury of hopping from job to job until I find something satisfying. I think I’m pretty intelligent and willing to work hard but my educational background is narrow enough that it probably won’t help me do anything I would actually want to do. Anyone have any ideas/suggestions?
You could educate yourself in something you want to do. E.g. I know someone who wanted to be a medical doctor but had an undergrad degree in english so they did a post-baccalaureate in pre-med for a year and then applied to med school. I know someone else with an undergrad degree in music doing a 2 year post-baccalaureate in CS.
I'd make general job satisfaction a relatively high priority. It's where you spend ~half your waking hours and I know several people who got crushed mentally by enduring a job they hate. That said, there may be many ways to solve this without quitting your job, including
- talking to your boss or co-workers (generally speaking, employees who enjoy their work tend to perform better and people have different things they enjoy / hate)
- reducing your financial needs and your work hours (or split work between two jobs)
- finding something worthwhile in your free time (if you fell in love tomorrow, would the job still suck so much?)
If none of this is possible and you find yourself in a downward spiral, please. Quit. Before. It's. Too. Late!!! It you're reorienting, the 80.000 hours guide on finding fullfilling work might help.
Heres the link: https://80000hours.org/career-guide/job-satisfaction/
Look for a job that gives you the time to think and plan about what you ultimately want.
(Hope this isn't too political for an odd-numbered thread.)
Nate Silver tweeted out that he puts the likelihood of the COVID lab-leak hypothesis being true at 60%.
https://twitter.com/natesilver538/status/1396550770368172033
Dr Fauci also said today that he considers it an open question that should be investigated.
What do people think about this? Are there any virologists in this community who can give an informed opinion on it? 60% sounds pretty high to me, although I don't view the lab-leak hypothesis as being a fringe conspiracy theory anymore.
I was a lab leak skeptic before, and I no longer am. I think 60% sounds reasonable. I'd maybe put it at 80%.
Is there any new evidence that is influencing people, or is this another case of a respectability cascade?
A bit of both. I've changed my mind on the lab leak hypothesis (I now think it is likely true), mainly due to evidence that I happened to see recently (but may have been available some time ago). The fact that we can't find a wild host population is the kind of evidence that only become stronger with time...
Yea, that was the case for me as well. Technically none of the evidence have been news, but they haven't really made it into the mainstream until quite recently. Kudos to the people who led the charge when nobody wanted to listen.
I thought the tweet quoted in the link above was quite significant new evidence ("if it spread from the lab, why are there no cases of employees there having it early on" previously seemed like a pretty compelling argument against the leak hypothesis but is no long).
Nate's tweet implied that he went from 40% before to 60% after, implying that he saw the new evidence (3 WIV employees visiting hospital in November) as actually somewhat weak (only 2.25 times more likely with lab leak vs. no-lab-leak). That's kind of where I am on this as well: I saw the lab leak as highly plausible before the new report, given:
1. the location (near a lab doing research using the most-similar known virus but 1000 miles from a SARS spillover zone)
2. new viruses that have just jumped from one species to another usually undergo rapid evolution as they adapt to their new host; SARS-CoV-2 did not, suggesting it was already well-adapted to infecting humans.
3. the lack of concrete evidence for natural origins, which was identified much faster in the case of SARS-1
4. lack of transparency from WIV (hiding databases, hiding the origin of RaTG13, obfuscating its connection to pneumonia cases in mojiang miners) that suggests they have something to hide
The new report is additional circumstantial evidence, but somewhat weak given that this could have been "common seasonal illness", and given that it was already reported that WIV workers were ill, just not that they visited a hospital.
60% has been a reasonable estimate from the start; I'd consider anything in the 40-90% range to be reasonable and I wouldn't consider it reasonable to spend too much time trying to pin it down. Not reasonable to spend too much effort to pin it down because A: it's unprovable and unfalsifiable barring a highly implausible outbreak of transparency in China and B: it's not actionable; everything we'd do if we knew it were true is something we should be doing just on the basis of it's being plausible.
What's interesting is that two months ago it was politically maximally incorrect to suggest this possibility, and now it's somehow OK and everyone's doing it. But that probably does get us into culture war politics in some way I haven't figured out. It's weird.
But, a priori, there's two ways for a zoonotic virus from a distant source population to show up in Wuhan. One is for a random walks of natural infection and bushmeat trade to proceed unnoticed across ~500 km until it shows up in that one wet market and no other, having left no trace of prior infections that anyone can find. The other is that people we *know* were in the business of going forth from Wuhan to collect strange new pathogenic life and bring it home to their lab in Wuhan, did exactly that and screwed up on the "don't let it out of the lab" part. This was clear from the start.
There's also the question of what people in the lab might have done to make COVID-19 more dangerous than it naturally would have been, but since the plausible answers mostly involve attempts to tease spoiler alerts out of Mother Nature as to what the virus would have evolved into on its own in another ten years. that too falls mostly into the unprovable-and-unactionable category.
Why not actionable? It seems that if we better understand the risks associated with certain form of research, that may inform us on how to better protect ourselves, and even if the benefits of such research outweigh the costs. I heard Josh Rogin saying that they're working on sextupling the funding for that kind of research, presumably as a measure to improve pandemic response. But if it turns out that kind of research actively caused the pandemic, that would probably be good to know.
I think the idea is that if the research changed the chance of this happening from 0.1% per year to 1% per year, that's quite significant whether or not this incident actually occurred that way, and the evidence about this one incident isn't likely to actually tell us much about the change in probability.
If we find out with 100% confidence that *this* time it was a wholly natural event, are we going to rationally respond with "OK, gain-of-function research is totally awesome and we don't have to be picky about whether it's done at BSL-4 or BSL-2 or BSL-I'm-too-lazy-to-follow-the-rules"? There's a set of behaviors, involving both laboratories and wet markets, that is rational based on what we already know and isn't going to change if we add detail or confidence to what we already know. So the details aren't actionable.
Why was it maximally politically incorrect two months ago but not today? I think this article has a good explanation of why the media initially treated it as a debunked conspiracy theory, even though it had some plausibility: https://www.washingtonpost.com/politics/2021/05/24/fix-china-lab-leak-0524/ (warning, politics plays a role). Once the hypothesis was placed in the land of "debunked conspiracy theory", it was hard for respectable people to be publicly seen to take it seriously, but this also meant the idea had much higher private support than it did public support. This created the perfect environment for a respectability cascade, which was further accelerated by increasing circumstantial evidence pointing in that direction, and a complete lack of any new evidence for purely zoonotic origins in over a year.
> Not reasonable to spend too much effort to pin it down because A: it's unprovable and unfalsifiable barring a highly implausible outbreak of transparency in China and B: it's not actionable; everything we'd do if we knew it were true is something we should be doing just on the basis of it's being plausible.
I don't think either of these are true in practice. China could be pressured by every other nation that suffered from their possible negligeance, so we could get more insight, and protocols for gain of function research should be significantly tightened.
I suggest reading the best article that reviews the available evidence to see why your B is likely false:
https://www.usatoday.com/in-depth/opinion/2021/03/22/why-covid-lab-leak-theory-wuhan-shouldnt-dismissed-column/4765985001/
So, if we find the smoking bat corpse that conclusively proves that Covid-19's origin was wholly natural, would you and Young then reason, "...and so my priors are now shifted towards US labs doing their work safely and we don't have to worry so much about lab safety"?
In order for a data point to be actionable, there has to be something you'd do *differently* if that data point were different. Preferably something more substantial that "use/not use this incident as a talking point when debating an issue that I'm not really going to change my mind on".
Rootclaim has a site on this - https://www.rootclaim.com/analysis/What-is-the-source-of-COVID-19-SARS-CoV-2 - with full quantitative reasoning and Bayesian likelihood calculations. Their reasoning may or may not be correct, but it crucially is *transparent* - you can see where the hypothesis starts, how and why various pieces of evidence affect their estimate, and what likelihoods they end up with.
Thanks, this is super interesting.
Thanks for turning me on to Rootclaim! Finally someone is doing actual Bayes math!
Next step: an app to help anybody do their own math? A collaborative app for crowdsourcing pieces of evidence? Both of these in one app?
There's 2 different lab-leak hypotheses you ought to keep track of:
1) the viral samples arrived at WIV with the same genome that leaked. SARS-CoV-2 is a natural virus that was brought from the wild to a city.
2) covid-19 is the result of GoF research.
I'd expect 2 to leave more evidence (of the impossible to cover up variety) than 1.
In terms of geopolitical implications, we're lucky that Wuhan doesn't rhyme with as many things as either Maine or Spain.
> I'd expect 2 to leave more evidence (of the impossible to cover up variety) than 1.
Like what?
Precisely. GoF is basically trying to speed up the natural process of jumping host, so I do not see how it could leave specific clues in the genome, contrary to editing (which specialist find highly unlikely, because engineering a efficient viral genome is much harder than let natural selection do its stuff (try a lot of variants and keep what work better -rinse a repeat). Some genetic histories may provide clues, especially the viral evolution speed after first observation, but if anything this plead for significant evolution before being observed in the wild, i.e. GoF. For example, many siginificant new variants may be linked to a passage through mink farms....Which is basically, from what I understand, an out-of-lab GoF...
My opinion about the likelyhood of an accidental lab release jumped to "damn likely" as soon as I learned that Wuhan lab did GoF studies and had gathered (so had access) to a genetically-close-to-covid bat coronavirus able to infect humans (the closest found yet). If the lab escape was wrong, those two facts would be a crazy coincidence...
There is well known, impossible to cover up albeit indirect evidence that SARS-COV-2 leaked from the lab: the fact that it was immediately so well adapted to infecting humans. Virologists have conjectured that it must have circulated among a sub-population of humans first to gain this ability, but no evidence of this population has been found.
I was hoping the the other poster had some other, more direct indicators.
That fact is also consistent with a natural origin if we also hypothesize that viruses which are mediocre at infecting humans in rural Guangdong are never noticed because nobody cares if a few Guangdongees farmers or guano-miners die and with only mediocre infectivity the occasional infected farmer visiting the Big City isn't likely to trigger a visible outbreak. So the virus gets to evolve in rural Guangdong for years until it *is* well adapted to infecting humans, and *then* the first carrier who visits the Big City starts a significant outbreak.
The bit where the Big City in question is Wuhan rather than Guangzhou is still quite suggestive, of course.
Posted this a bit late in the last OT and got no replies. While it contains the word "politics" I don't think it's actually political, so reposting:
If you think AI takeoff is likely to happen in the next 20-30 years but are useless at coding/politics/business management/finance, how do you do something useful with your life?
I mean, aside from literally stopping people from dying before singularity (which could actually be bad in case of Roko-style or MMAcevedo-style Virtual Hell), there's not much you can do for people that won't be rendered irrelevant by a bad *or* good AI takeoff, and without access to one of those you have ~0 control over the time and goodness of takeoff.
Twenty or thirty years is long enough for quality of life to matter quite a bit.
Also, likely to happen isn't the same as inevitable.
Barring anti-aging, everyone alive now is going to die within ~100 years anyway. If everyone is going to die (or ascend) within 30 years, I'm not sure that changes the calculus too much. If you were going to try to save/improve someone's life despite them dying in a century anyway, it's probably worth saving/improving their life when they'll die in 30 years.
Or if you really want to focus on the AI thing you can work a normal job and donate some percent of your income to charities trying to improve AI safety.
My passion is teaching. One could hope under normal circumstances that this would do some lasting good (i.e. helps people help other people etc. and at future time X for X < some large number the world is distinguishably better). However, if we all get eaten by Universal Paperclips in the year 2040, then the state of humanity is effectively erased at that time; I could be the best teacher in the world or the worst, and my students would still all be dead in 2041. And if we all get uplifted into a proto-super-Culture in the year 2040, then all my students could download all the knowledge of humanity into their brains instantly and would have essentially-identical living conditions regardless of my input (the only exception being if they're dead by then). Again, the information of what I've done is effectively chucked into a black hole.
If I thought AI were 100 years off, then I wouldn't be so worried, as I'd be more confident that I'd have some butterfly effect on the nature of what happens. But 20-30 years is rubbing up against "too short a timescale for this" (I think you might have said something along these lines at some point, or linked to such?).
And I can route around little problems like nuclear war; I can go into regional education, and I'm more than capable of teaching people to Duck and Cover. But a bad AI is under the category of "storms we cannot weather", along with alien invasion or divine/simulator intervention.
As far as charity goes... well, that has to do with why I mentioned finance. I'm not especially likely to get rich enough to be more than a drop in the bucket there, and of course this assumes charities are of positive use here (it's never 10^-67, but it's sometimes 0 or -0.001 - for instance, if an AI-safety charity has differential ability to affect Doctor Good and Doctor Amoral, they might end up giving the keys to the kingdom from what would have been a Doctor Mostly-Good 30%-shot to a Doctor Amoral 5%-shot). I mean, sure, that's something I'll consider, but it feels pretty limp.
The problem is a familiar one:
“The doctrine of the Second Coming teaches us that we do not and cannot know when the world drama will end. The curtain may be rung down at any moment: say, before you have finished reading this paragraph. This seems to some people intolerably frustrating. So many things would be interrupted. [...]
“A man of seventy need not be always feeling (much less talking) about his approaching death: but a wise man of seventy should always take it into account. He would be foolish to embark on schemes which presuppose twenty more years of life: he would be criminally foolish not to make, indeed, not to have made long since his will. Now, what death is to each man, the Second Coming is to the whole human race. We all believe, I suppose, that a man should ‘sit loose’ to his own individual life, should remember how short, precarious, temporary, and provisional a thing it is; should never give all his heart to anything which will end when his life ends. What modern [people] find it harder to remember is that the whole life of humanity in this world is also precarious, temporary, provisional. [...]
“Frantic administration of panaceas to the world is certainly discouraged by the reflection that ‘this present’ might be ‘the world's last night’; sober work for the future, within the limits of ordinary morality and prudence, is not.” (C. S. Lewis, “What If This Present Were the World’s Last Night?”)
“It is like a man going on a journey, when he leaves home and puts his servants in charge, each with his work, and commands the doorkeeper to stay awake. Therefore stay awake—for you do not know when the master of the house will come, in the evening, or at midnight, or when the rooster crows, or in the morning—lest he come suddenly and find you asleep.” (Mark 13:34-6)
I think the principles are very similar in the case of the singularity, although without the expectation of Judgement that gives a particularly sharp moral character to the whole question for theists. You cannot know when the paperclip maximizers will dissolve the earth, but then again you cannot know when you will be diagnosed with cancer or hit by a passing car—we all invest our lives in projects we may not see to fruition. All we can do is the work in front of us, the tasks that our social roles, obligations, talents, and resources, as well as the needs around us, fit us to pursue. I think (though I don’t know the right philosophical term) that this can even be made utilitarian: if you had perfect knowledge you might calculate that doing something very unusual would maximize utility, or that action on your part is useless and you might as well give up or do as you please, but human knowledge is never perfect and most people who believe themselves to be in these exceptional cases are wrong, so it is very likely best for everyone to stick with “sober work” and “ordinary morality and prudence” even when they believe their circumstances are exceptional. I would say, teach and serve in the time given to you; if the singularity comes, let it find you at your post, and you may after all be pleasantly surprised with more time than you’d planned.
If you're a genre fiction author, do you think of your novels also as story treatments for potential movies or series (in other words, are you actively hoping to get your books adapted)? If so, how does this affect your writing?
You'll probably get more relevant responses than mine, but I was once a wannabee fiction author and wannabee screenplay writer.
If you want your novel to one day be a movie, you should think cinematically, which mainly means thinking visually. Exposition and interior monologue are not cinematic. (Voice overs should be a last resort in movies.) All visual detail is good. The more granular the better.
Dialogue should be tight and minimalistic. Hemingway basically invented movie dialogue. His exact words didn't make it to the movies because the studios back then had a philosophy that everything needed to be rewritten, but all the good movie writers stole his style. Hollywood today would still want to rewrite your dialogue, but if you want it to have a chance keep it minimalistic.
The main difference between a movie and a novel is dramatic structure. A novel can be any length and follow whatever rhythms it wants. A screenplay generally needs a 3-Act structure. It's probably easier to turn a short story into a screenplay than a novel, but we no longer live in the era of the popular short story. You should probably think of what the short story version of your novel would be like and actually write one, if you want to imagine it as a movie.
Ultimately, though, the more commercially successful your novel the more likely you are to sell the movie rights to it for a lot of money.
I realize that I didn't answer your question, but I answered the question I was able to. For all I know you are trying to avoid the influence of the movies on your novel writing. If so, more power to you!
I value being an independent thinker -- it is one of my highest values -- yet I realize I go too far with it sometimes. I hate to go with the crowd. I was raised with the idea that the crowd is mad and hate the stink of the herd, yet these days the educated wisdom is that the crowd is wise not mad.
Emerson says "One recognizes in the genius of others one's own aborted thoughts." I believe in the wisdom that statement, so I try to see my own thoughts through, even if they seem anti-social.
Perhaps the Internet has made a major difference in how one should weigh their own thoughts vs the crowd. As a pre-Internet person, "the crowd" was the high-school halls. You were a fool to follow that crowd. Or very clever.
Now you can choose your own crowd. You can follow a crowd that you know is smarter than you. That was rarely the case for most people for most of human history.
Perhaps independent thought has less value these days. Perhaps there are fewer $20 bills on the sidewalk simply because information travels so much faster.
My question is -- I usually have a question -- are there good signposts for determining when your independent thinking has gone astray?
You could ask the general question of whether the crowd could be right. Or at least partly right.
Depending on who you're hanging out with, saying that the public is right about something is the contrarian thing.
Yudkowsky spends a lot of words trying to address that question, I think. In his Inadequate Equilibria book he talks about what basically is holes in the common wisdom, areas where you can hope to beat it.
Other than that, I have a feeling this boils down to 'assessing the quality of your thinking', which is basically the same as (the bottleneck in) improving it.
When evaluating my own thinking in comparison to others', I've found a helpful test to be "assuming other people are reasonable, can I explain how they arrived at their beliefs?" This is similar, but not the same, as steelmanning the other side--sometimes the other side is simply wrong, but there are social or cultural reasons why their beliefs persist in reasonable people. In particular, "other people are dumb/evil" is not an acceptable explanation. If you can't explain the reason for why the consensus views are such, then you're probably missing important information.
> yet these days the educated wisdom is that the crowd is wise not mad.
A position that's been perverted by poor science reporting. The crowd is wise when ts members are not all exposed to the same information and opinions, and thus must make up their own minds from a plurality of non-ovrelapping data.
Any bets on whether that describes the current media landscape?
Since my earlier post was linked in this open thread as “Russia strong”, I should clarify that “Russia mediocre” would be more accurate and less interesting.
I liked gizmondo's Guriev quote (though I traced to to Enikolopov?): “If you are in a swamp, you don’t expect to fall off a cliff.”
One time a boss told me something that has stayed with me. He said "It's easy to say 'This is fucked. What's hard is contributing something better.'"
There's truth in that, yet I think more and more people should be saying "This is fucked." at their job. I read endlessly on the internet about people complaining about how woke or whatever their workplace is. What happened to saying "This is fucked."? Granted, I'm a Gen-Xer and that's all we said, so it's easier for me to say. I just don't get why people can't tell their corporate bosses these days how they are fucked.
Corporate culture has so much power today because people worship at its altar. We used to want to die before we turned 30. If you work at Google and complain about it's culture, well, you had other options than selling out. Your politics on the internet don't matter: You are the problem. You support our woke power structures if you work for corporate America.
Well, some argue that the main reason for the wokeness push is exactly so that it would sideline economic leftism. Incidentally, one of its main features are labor unions, the very institution supposed to compensate for the powerlessness of a single employee in the face of general fuckedness of the workplace.
Wokism is right wing any way you slice it.
"freedom"
Right wing "freedom" as far as I can tell is freedom for me but not for thee.
“ Any freedom that right-wingers claim for themselves they will happily grant every american has.”
That’s very true of the libertarian wing of the right. It’s very much not the case for the (much larger) non-libertarian leaning majority of the right.
This is interesting - I thought the entire idea of woke-ism was people starting to say "this is fucked" without any idea of what to do better, and finally getting around to saying it, instead of just ignoring it like we did for the previous several decades.
Woke-ism is kind of like saying "this is fucked", and then contributing something worse.
Only half kidding.
Yeah, I think that's the usual failure mode of saying "this is fucked" without having a concrete better proposal.
I can't believe Kentaro Miura is actually dead.
Yep, that was very sudden, but then again the manga industry is both very secretive and extremely exploitative, so maybe it shouldn't be too surprising. Still, as far as "dark fantasy epics to remain unfinished by their author" go, I certainly expected ASOIAF to be confirmed in this status first.
I never even realized Miura was suffering from stress, since I didn't look much at his work aside from Berserk (I was on the boat for 4 years since I started reading late). It's sad. I seriously hope Martin takes care of himself.
Ugh. I crunched some Metaculus predictions against actuarial estimates yesterday, and my ballpark estimate is that we have a ~22% of getting A Dream of Spring before Martin's death, and that number roughly halves every year.
*every additional year from 2021 that passes without publication of The Winds of Winter, that is. Still a guesstimate, but that's what it's conditioned on.
Agreed. He was pretty young too as these things go -- 54. I recently saw an interesting take on the timing of his death with the plot of Berserk, which is that Miura's death came at the one point in the story where most of the characters have found some kind of peace. (Obviously a lot of bad stuff still going on, but relatively speaking, I think the main cast is at a pretty high point compared to usual.) Even though it's kind of silly, this makes me hope that Miura himself -- clearly being a guy with many personal demons -- found some kind of peace near the end. Maybe wishful thinking, but I've had a bad week, so I'll take it.
Sorry for posting so much. I'll go to sleep soon and won't post again for a few weeks. Just getting all my ideas out now.
Why do people in this crowd care so much about future humans? I can understand caring about your grandchildren, but your grandchildren's grandchildren? Really? I don't care about them.
The future of the human race? Is that even going to be a thing? We're not crocodiles. Our environment is changing and we are changing with it. H.G. Wells probably had the right idea in The Time Machine: the future of humanity will not be humanity, it will be some bifurcation of troglodytes and geeks. Why care about the future of your genes? They will be a different species than you. We are Cro-Magnons and Neanderthals, to our descendants, or just dump apes. Or maybe our descendants will be more like rats. Intelligence isn't everything. Why do you care about our great grandchildren rats so much? You want to fill the universe with them? We'll be the species that fills the universe with rats. I know I will feel proud about that.
The Sex Pistols had it right, there's no future for humanity. Our genes may have a future, but it won't be human.
Future humans? Grandchildren? I just want to live forever, man.
Why? Don't you disappear at every moment? Tomorrow you isn't you. That's a different being with a different brain and body. You die every second.
Well why want anything? Preferences are ultimately arbitrary.
What I do find remarkable is how volatile human terminal goals are. Or at least they feel like terminal goals, until some supporting assumption breaks.
Why should an agent change their fundamental decision theory just because of some philosophical insight? Humans seem to do that. Nihilism as a class of glitches.
Makes me wonder how our goals are really organized. What is the meta-utility-function? What do we truly want?
That's because terminal goals are unimportant from evo-psych perspective, they matter only insofar as they influence behavior. And it takes a very special kind of human to substantially change their day to day life after experiencing a philosophical insight, so rare that it's statistically insignificant.
Do you expect some complicated argument involving acausal game theory and simulation capture or something? Too bad, I just want to continue experiencing hedonic pleasures. I don't care whether the information-theoretic distance between "me" right now and "me" two weeks ago is within some epsilon-level of tolerance, "my" ability to experience things get taken along on the ride anyway.
The correct answer to "Why?" is "I just do. Wants don't require reasons."
Sounds like a question of "what kind of agents do I consider my ingroup / kingroup?"
We're just more-or-less hardwired to care about some varying subset of conceivable intelligences in a tribal way, no deeper meaning in that.
I'm not sure I care tremendously about future humans, but I at least find the idea of caring about them attractive.
My assumption is that genetic engineering will lead to an explosion of different sorts of humans. Some of them will be somewhat like us, some of them will be very different.
I've wondered about heritage from a different angle. There's a big deal made about ancient art being kept in the possession of the modern country which is on the territory where it was made, but I can't imagine the ancient Greeks caring about whether modern Greece has the Elgin marbles. Not that hypothetical mental states of ancient people should get a vote, but possibly modern nations project themselves into the past too much.
As for The Time Machine future being "troglodytes and geeks", it's funny but perhaps not entirely wrong to think of geeks as Eloi, or at least the production-consumption-and-discussion of art as Eloi behavior.
I don't think it's that people are thinking in terms of past mental states. The connection shared by members of a tribe, transcending both space and time, is generally understood to be spiritual at least as much as psychological. As most people perceive things, if I'm defending my nation's past, I'm also defending its present -- if nothing else, its present reputation and dignity.
They aren't thinking in terms of past mental states, but I've made some art and I thought it might be worth thinking about a maker point of view.
What I've done is unlikely to survive or be remembered for 50 years let alone millennia, but in my imagination, I wouldn't care about it being in the hands of whatever unimaginable culture holds what is now the middle Atlantic states. It would be nice if humans cared about it.
Because we're wired to. A billion years of ruthless natural selection have ensured that PROPAGATE THE GENOME is graven into every organism above a virus.
I don't think PROPAGATE THE GENOME is the right level of abstraction. For humans, it seems to be HAVE SEX and RAISE CHILDREN and possibly HAVE GRANDCHILDREN, with the strength of the first two varying a lot among individuals. My impression is that the desire for grandchildren is connected to the desire for grandchildren.
Anyway, we've got the tech for sex without fertility, and it turns out that a lot of people prefer that. Evolution doesn't care.
If it really was PROPAGATE THE GENOME, there'd be a lot of guys sneaking their sperm into sperm banks. Instead, while it happens, it's pretty rare.
I think you're thinking too narrowly. The same basic motivation underlies people who want to "make a difference to the future" as well as people who contribute to child-rearing in a more general way than having their own children, e.g. people who go into K12 teaching, child-care, child psychology, pediatrics. It's rare that jobs that relate to improving the lot of children -- any children, not just your own -- need to pay the kind of wage premium needed for less desirable jobs, like collecting the trash or negotiating divorce settlements. These jobs are "underpaid" (relative to others of similar qualification and responsibiity) precisely because part of the "wages" are the satisfaction of internal drives (and external validation of same).
None of these things can be reduced to "have sex" and "rear children," they are all more abstract urges to forward the success of genomes related to, but not necessarily the same as, your own. For that matter, evolutionary biologists have speculated that one reason whatever genetic predispositions to homosexuality exist haven't been suppressed long ago by natural selection is that the homosexual member of a family or tribe can provide a significant *advantage* to genomes related to his own, relative to another parent. A gay uncle, say, who makes a fortune and/or becomes socially powerful (because he's undistracted by child-rearing) and who then provides wealth and social standing to his nephews is doing a big service to the genome, much more than if he simply supplied another few copies of it.
The same kind of thinking underlies the "grandmother hypothesis" for the existence of menopause: at some point, the contributions of a woman to the prospects of her genome at a more abstract level outweight the simple creation of more copies of it (i.e. having more children of her own). She can provide wealth, guidance, wisdom, and these things enhance survivability more than a few extra copies.
The original question was along the lines of "Why do we give a damn about what happens to humanity after we are dead?" My answer is "because we're wired to, because most genes that coded for purely YOLO self-centeredness got edited out umpty millions years ago." I'm aware that most people are uncomfortable with thinking their noble aspirations can be reduced to biochemistry, but that's probably just our egoism. If you asked "Why do horses give a damn about the welfare of foals?" we would not indulge in similarly elegant flights of fancy about the philosophy of horses. For that matter, the fact that we *almost all* come up with very similar motivations (and even rationalizations for them) is rather suspicious, if the origin is presumed to be original philosophical introspection. That's like imagining we (almost) all like ice cream because we've all independently analyzed its nutritional value and logically concluded it's good food -- rather than that we all have taste buds which the DNA have caused to like being bathed in sucrose.
My point is that the drives aren't PROPAGATE THE GENOME, they are DO THINGS WHICH HAVE BEEN PLAUSIBLY RELATED TO PROPAGATING THE GENOME.
Well, OK. I'm not entirely sure I understand the difference, honestly. (Nor do I think "plausibly" belongs in your rephrase, as Mother Nature and Father Evolution never act on merely plausible -- it either does or it doesn't, and it only gets encoded in the DNA if it actually does.)
Why do people use birth control? Because instinct says "have sex" rather than "have children".
My theory explains why sperm fraud isn't extremely common.
I admit I don't understand why it happens at all.
Perhaps we should be thinking about the pleasures of trickery. For some people, it seems like a goal in itself, and for others it might be entangled with other goals, but still a strong drive.
Things to understand:
1. What works tends to be kept, what doesn't tends to disappear.
2. Evolution is stupid - it's just mutations. Variations on whatever currently exists. (and btw most variations are detrimental)
So the phrase "do things plausibly related" means "do things that (1) were correlated with gene propagation in the past, provided that (2) those things were discovered via mutation in the first place (e.g. a perfect memory, or a built-in biological compass, or an instinct for complex mental arithmetic, or extremely accurate toxin detection in the tongue, or the ability to hibernate, or a cryptographically secure immune system, might all be helpful for human gene propagation, but we don't have these features because mutations didn't achieve those things, and for some of those things mutation seems very unlikely to ever be able to achieve them.)
Any thoughts on the role of fundamental energy balance in mental health?
Mental illnesses can often lead to poor sleep, which in turn leads to both general misery and an energy debt that prevents the tackling of the issue in the first place.
Similarly, declining physical fitness can reduce a person's ability to handle stress and the energy they have to resolve the stressors.
Basically I'm wondering if concepts from sports science have relevance to psychology.
This might be interesting (if you buy into the predictive processing framework). https://www.reddit.com/r/slatestarcodex/comments/nj9sfi/annotated_quotes_from_immunoceptive_inference_why/
Perfect, cheers.
In Time Enough for Love, Heinlein has it that rejuvenation takes highly skilled professionals. I suspect this was at least partly because he wanted to have something for some of his characters to do, but does it actually make sense?
Also, though it was the story Heinlein wanted to tell, you might get a fairly dystopian society where whether you age and die depends on what you can afford.
I can believe rejuvenation is harder than extending life and health starting with a young person, but how hard would it be?
I talked with someone decades ago who was keeping records of his bloodwork so that when rejuvenation was possible, the doctors would know what his baseline is. He said that blood factors can vary a lot from one healthy person to another.
"There is no simple effective against death", or something like that. People grow old and decrepit for two dozen reasons simultaneously, and most of those reasons are side effects of biological processes that have good reason for existing. Most any possible countermeasure will have its own side effects, mostly ones we don't want. Trying to balance all of that even in the not-growing-old sense, so that we get a centenarian who is youthful and healthy and not e.g. tumor-ridden, is going to involve a great deal of applied medical expertise that probably isn't going to result in the same one-size-fits-all answer for everyone and probably isn't going to be automated this side of the strong-AI singularity. Adding in the requirement that we reverse aging that has already occurred, I suspect won't make it *lots* harder, but it will be hard enough to begin with. So, yeah, "anagathic internist" will be a recognized medical specialty if we get that far before the Singularity, and rejuvinative gerontology will be a more exotic specialty.
I think your answer is very solid, but just for fun let me throw out an alternate perspective, not because I think it has a greater chance of being true, but just because I think it has a nonzero chance of being true.
What if there *is* a fairly simple reason for the ravages of age? For example, it might be located in a programmed (one assumes) degradation of the immune system. We know senescent cells that escape recycling by the immune system seem to have some outsize influence on aging effects. We suspect age-related failures of the immune system are one reason cancer risk rises dramatically with the onset of late middle age. There is the Longo-group hypothesis about the connection between calorie restriction and immune system regeneration that might start to provide a mechanism for why calorie restriction extends life (at least in lower mammals).
In short, what if the only thing you really need for a substantial rejuvenation effect is to jazz up the immune system so it starts aggressively pruning senescent and abnormal cells, so you rebuild those tissues from stem cells? That might not be exceedingly hard to implement, and maybe stuff like CAR-T therapy is pointing the way. Still needs about 100 years of basic research, but it could conceivably end up being some fairly standardized therapy -- we take out a sample of your blood, do X and Y to assorted immunological cells, reinject them, and hey presto in a few months you look and act 40 years younger.
That doesn't mean you have an anti-agathic, of course. You might not even extend life by all that much -- there is plain evidence that cells have some ultimate reproductive limit when functioning normally (although the existence of immortal hybridoma lines might suggest that limit can be subverted relatively easily). But we might want to separate "rejuvenation" = "feeling 35 when you're 75" from "life extension" = "living to age 200 instead of 100." And there's definitely value in the former, even if the latter remains difficult or out of reach.
Maybe. If there are several biochemical trainwrecks happening at roughly the same time, the effect size will be unimpressive even if you manage to solve one.
The Hayflick limit is mostly irrelevant to aging, cells can produce telomerase in a variety of circumstances. AFAIR by Hayflick limit we'd live around 250 years, and by that point we'd figure out how to do away with it for good.
If there were a fairly simple reason for the ravages of age, there would fairly quickly (on evolutionary timescales) be many reasons for the ravages of age. Designing something to last centuries is *hard*. So if there's one simple thing that causes everyone to die at ~100 years, then the evolutionary process for every *other* thing is going to optimize at "designing this particular thing to last more than a hundred years is too hard and offers no payoff whatsoever, so we're not going to do that".
You're going to get e.g. joints that form with a hundred years' worth of cartilage and no good way to add more, and your One Weird Trick of aggressively pruning senescent cells isn't going to change that, and your rejuvinative gerontologists are going to need to learn a separate weird trick for adding new cartilage to joints. And many more separate weird tricks for other things that break down around the same time.
Progeria suggest to me that there's a mechanism which causes aging. I want know what happens if that mechanism is shut down, though I grant that there will still be a lot which doesn't last indefinitely.
Assuming it'll be some kind of complex gene therapy (which I'm giving 90% odds), probably a medical professional to administer it and monitor that nothing goes south will do. Can be downgraded to an intern who pretends to know what they're doing, if you relax the safety requirements.
I have started a Weekly Digest on tech, deeptech and futurism: https://www.jack-chong.com/2-jacks-weekly-digest/
https://thetyee.ca/Analysis/2021/05/21/The-Pandemic-Speaks/
From the point of view of plagues-- perhaps something Taleb could have written if he had more dignity.
https://www.youtube.com/watch?v=QScaLhDVacg
A little introduction about hyperloops.
I'm wondering whether congestion at terminals would be so bad that a lot of the speed advantage would be lost. Is this reasonable?
The biggest criticism of hyperloop all along (apart from unrealistic construction costs and technologies) has been its low capacity.
Unimpressed on comment #1. Of course Tim Cook, CEO of a company that tries to ensure its products are perceived as luxury (i.e. not cheap), under tremendous US domestic pressure to onshore some of their production given their absurd profit margins, would try to deflect "cheap" accusations and instead suggest American incompetence. It's hard to defend massive profit margins in an interview if you're just "cheaping out on labor".
The funny thing about tooling is you don't need to use it where you build it - the overwhelming majority of tooling globally is used, manufactured, and designed at different sites, often continents apart. Tooling is just as global a market as any other product. Makes you wonder, if China is such a high cost market, why those assembly lines don't make their way somewhere cheaper?
How quick we are to accept excuses when cost optimization pressures will suffice!
I'm sorry, what pressure? Tim Cook has $200 billion cash on hand, and his products sell like ice cream on a hot day. If you can think of an effective way of pressuring a person in such a situation, I'd love to hear it. Maybe if 250 US Representatives and 60 US Senators *and* the President were all pissed at him *and* couldn't be bought off with a few $hundred million in campaign cash, there'd be some pressure that he'd actually feel.
Scott, I just watched the latest Robert Miles video: https://youtu.be/IeWljQw3UgQ?t=485
He mentions the possibility of an AI system working great, until someone factors RSA-2048 (https://en.wikipedia.org/wiki/RSA_numbers#RSA-2048), at which point the AI realises it's in the real world and not a simulation and begins acting in line with its *real* goals.
Made me think of some of the fiction you've written, and seems like something you'd like in any case.
Yeah, that scenario definitely had the potential to be a terrifying premise for a story.
Police anecdote time:
A small group of anecdotes, shared by my friend group across the nation in various urban/sub-urban areas.
I got robbed, door kicked in, various items taken, including credit cards and devices with logged in accounts.
Police called. Officer arrives 10 hours later. Looks around. Says "We'll get right on it". Leaves.
The same day, I locate logged in devices at a residential area near me, turned on, with ip address. Cops aren't interested, tell me to check in pawn shops for my stuff.
The next day, credit card is used at gas station. Gas station personnel say that the police can access security footage, and they have the plates of the car the stolen card was used on. Police categorically refuse to access the footage, require that I do it. I am told it is illegal for me to access the footage, I either need to get a lawyer or have the cops do it.
I find some stuff at pawn shop, owner will release copies of the probably fake id to the cops when they call. I tell the cops. Nothing. I call the pawn shop 2 weeks later; he hasn't heard anything from the cops. They didn't bother to call.
I cut my loses and just have my homeowners insurance comp me for everything, and anchor a jobsite box to my slab to lock laptops and guns and such in while I'm out and rout my NAS cables through the punch outs.
This is every story, with the details switched.
East coast friend was mugged at gunpoint; managed to get security camera footage of the dude using his card at a 7/11, cops weren't interested.
NW friend got his fancy bike stolen, got the security cam footage from his building, and the cops didn't even want it from him.
The only time I've ever seen cops do anything was being available for a guy that got fired at a jobsite, and clearing people who were too un-aesthetic from in front of big box stores.
While anecdotes are anecdotes, this has given me the feeling: The cops won't help. Calling the cops for anything other than a guy with a gun RIGHT FUCKING NOW is a waste of time.
I don't understand how 570 million for a medium metro area gets you about 20% above nothing. I'd expect to at least see cops out and about wasting my money for that amount.
I've seen versions of this comment from the Defund the Police maximalists, who actually want to defund the police. I think the counterargument/explanation would be that the cops figure your insurance will make you whole, so they're just there to provide documentation. But yeah, it's pretty weak tea.
I had a cop repeatedly gaslight me that I wasn't harmed when something was stolen from me. The thieves stole something from my bank card, and since the bank has refunded me the money the cop was "well it sounds like the bank was the only victim." The thieves charged something to a credit-card vendor (in an easily traceable way) and the cop was "well it sounds like that vendor was the only victim." I had to keep on reminding him that cash was stolen as well.
I don't like the "Defund the Policy" narrative but damned if the cops don't make it appealing. I just bitched about the susceptibility of schools being captured but at least we can opt out there. What's the escape for a do-nothing police system?
Same here (EU). Police is largely there to scare the population into obeying laws by existing. Their actual ability to enforce is nonexistent, unless the perpetrator is really careless.
It mostly works because those most likely to realize this are, in fact, really careless, and the rest is a rounding error in the grand scheme of things.
Reading Scott’s anger at people who want to deny people a chance to do better for their kids, I wanted to take some time to address why some of us who’ve spent a great deal of time in education oppose not just charter schools, but alternative forms of education outside of government run programs generally.
The basic idea is that these alternative (and sometimes improved) forms of education allow the upper class to “opt-out” of the general public system. This removes a strong direct incentive on those with the most political power to actually address educational problems and indeed you’ll often see people who use these alternative, and gated, forms of education work to reduce funding to public schools making things worse for those less well off.
To make an example of this dynamic, let’s visit a location near Scott and myself - the cities of Sausalito and Marin City inside of Marin County (near the Golden Gate bridge in California). In 2019, the city got hit with what LA Times reported was the first desegregation order in 50 years:
https://www.latimes.com/california/story/2019-09-21/desegregation-order-sausalito-marin-county
My family has been directly involved in the school system since the 90s so I’m aware of many of the public and not-so-public facts. The basic story was as follows:
1. The school district covered Sausalito (primarily white) and Marin City (primarily black and brown). The class makeup of Sausalito has continued to shift up over the years, from middle towards upper, while Marin City has not changed significantly from being lower class.
2. A charter school was set up that functionally only took students from Sausalito and took basically all of them, while children from Marin City continued to go to the public school.
3. Having a greater section of the population in the school district, Sausalito residents voted for officials who would allocate more resources to the charter school than the public school. The charter school would not only receive more money per student but many additional services both recorded and not (volunteer drives, waiving of fees, various services etc. can be granted without a paper trail).
I’ve watched this struggle one step removed for my entire life. Much has been made of the actual racism involved, and I can confirm it was there and it was ugly, but many of the parents who supported these policies were really just trying to get the best education possible for their own child damn the consequences. So if they could take a dollar per student from the poorer kids in the public schools and get away with giving it to their own kids, they did.
I’ve spent enough time in schools and education at this point to see that similar scenarios play out through a lot of the country, though generally the class lines are a little more pronounced and the racial ones a little less. Those with time and money available free themselves from the system, pool their substantial resources to provide an improved experience, then try to free themselves of obligations to those with less. We all end up significantly poorer for it. (I recognize some of this is institutionalized in property tax funding of schools, good essays have been written on that and I won’t repeat them here).
Rich people care deeply about their children’s education, allowing them to not encounter the same problems as everyday citizens allows them to not have to care nor expend power to improve the system for themselves in a way that also helps others. I think that’s a substantial loss for the society as a whole.
I’m sure I’d enjoy a lively debate with Scott on some of his other views on education, but that stood out to me as unaddressed.
How does the charter school filter it's applicants to mostly take the ones from Sausalito? If this part were fixed would you still see a problem? Do you think, besides demographics, the charter school is doing something special which is of benefit to its students?
Thanks for the post.
Probably a different school, but Steve Sailer shared this anecdote in the Cult of Smart review comments. (If you don't know who he is, google him).
"When I had my son apply to a charter high school founded by teachers at his middle school, I was told that admission was by lottery only, but the parents had to show up in person to find out if their child was picked at random.
When I showed up to find out if my son's name was on the list, the person holding the list was his old middle school math teacher. I nervously asked if he'd been picked in the lottery? His teacher said, without looking at the list: 'Of course he did.' I asked if he could check the list just to make sure. The charter school teacher said: 'Don't worry about it. He's in.'
"
I've heard about a charter school which required the parents to show up in person to apply. The school wasn't accessible by mass transit.
> The basic idea is that these alternative (and sometimes improved) forms of education allow the upper class to “opt-out” of the general public system
Damn right.
I've heard your complaint many times: people should stay in the broken system and try to fix it. Usually told to me by the people that have captured the broken system and will decide how to portion out the gains of any effort I put into fixing it.
Lots of systems are subject to capture by insiders. Escape is the safety valve.
If you spend your time worrying about people escaping East Berlin, your solution should not be adding more barbed wire to the wall. No wonder people feel anger.
But you haven't addressed his point about the non-rich getting shafted in the process.
Which of these seem right to you?
1. People should not be allowed to live where they want. They should be assigned to class-mixed housing projects - without even separate apartments, maybe you can have your own bedroom, but must share kitchens/bathrooms/living rooms etc - so that the rich people have an incentive to make sure the poor people have good living conditions.
2. People should not be allowed to eat where they want. They should have to go to class-mixed cafeterias, so that the rich people have an incentive to make sure the poor people have good food.
3. People should not be allowed to work where they want. The government should assign them jobs, with some poor people and rich people on each team, to make sure that the rich people can fight for good working conditions.
4. People should not be allowed to raise their own children. The government should raise all children communally, so that rich people have an incentive to make sure the communal child-rearing conditions are good since their own children are in them.
Like, I see where you're coming from - what you say makes some sense as a mechanism of ensuring quality for poor people - but I don't understand why the poverty concern outweighs the autonomy concern for schooling but not for anything else.
I'll have a longer post on this later.
In 2019, an interesting study came out about school district secession, a somewhat more drastic version of what wpow describes in Sausalito. Because Southern states have tended to organize their public school districts on a county basis due to being more thinly populated and therefore have tended to have more integrated schools post-Brown than other parts of the country, the effects of these secessions are easier to see because they're less bound up with other social phenomena like suburbanization.
https://www.vox.com/2019/9/6/20853091/school-secession-racial-segregation-louisiana-alabama
One of the more toxic inheritances that we American millennials are receiving from the Boomers is the _de facto_ racial segregation that is inscribed in our urban geography. As you say, trying to control where people live is no solution to this problem. You could try to create broader school districts that cover entire metropolitan areas but that brings with it a whole host of problems from greater scale and increased bureaucracy.
This is the main point that Ta-Nehisi Coates was making in "The Case For Reparations". You don't need to go back to 1865. The harms done by postwar redlining to black Americans, which stemmed in part from the federal policy that created the 30-year fixed-rate mortgage, have continuing effects to the present day. The racial segregation that postwar suburbanization created is only getting worse because individual families' responses to these perverse incentives exacerbate segregation in the aggregate.
I agree with you that we're not going to make any reforms that constrict families' autonomy to choose where and how they want to live, nor should we. However, poor people have less autonomy. They can't move into a better school district if they can't make rent. If we can find effective means by which to close the racial wealth gap and make it easier for _any_ family to vote with their feet and move from dysfunctional jurisdictions to functional jurisdictions, then the aggregate effect of individual families' choices about where and how to live could potentially tip from exacerbating inequality to reducing inequality.
I've seen suggestions that schools should be funded by larger regions than school districts-- you move the money instead of the people.
Certainly here in Illinois, the school districts' heavy reliance on property taxes tends to create positive feedback loops where losing a part of the tax base through a plant closure causes the schools to lose funding, which makes the district less attractive to young families, which suppresses residential real estate prices, which causes the schools to lose funding to inflation, u.s.w. That's true in both the suburbs and here downstate. Of course, nobody should be modeling public finance on anything we're doing here in Illinois.
Scott, this is really strawman-y. Housing, food, work and children are all private. And no one (or close enough to no one to round off) is proposing anything like ending the parallel private system. No matter what the government says, if you don't want your kid to go to school with poors, you can send them to private school.
But the charters are part of the public system, which is supposedly established by us all and for us all. So when the charters are being used as a publicly funded way to keep well-resourced upper middle class kids away from the poors, well, that's different. If you want exclusive private schools you should just send your kids to those. But I don't see any reason to pay for it with public money, and I think that we have the right to ask the public system to do things differently.
Or, to answer your points:
1. People should be allowed to live where they want, but if they want to live in public housing, they can be asked to live in certain places and even assigned neighbors according to public policy goals.
2. People should be allowed to eat where they want, but if they want to eat for free the gov't could provide cafeterias. If you show up to one of these, there are no private dining rooms if you make over a certain amount per year or can show up at an inconvenient time to make a reservation. There's just the common room that everyone shares.
3. People should be allowed to work where they want, but if you can't find work you can register at the re-established WPA, where you'll be assigned a job and a team. But you may not get to pick your job or teammates.
4. People should be allowed to raise their own children, but if they'd like to avail themselves of free daycare and afterschool programs, they can register at a central office and be assigned to providers.
While you could certainly implement any of the suggested programs differently, each of those seems reasonable to me, and each seems a better analogy than the ones you're using.
> 1. People should be allowed to live where they want, but if they want to live in public housing, they can be asked to live in certain places and even assigned neighbors according to public policy goals.
Does the principle change if my town government pursues a zoning strategy that just so happens to drive 90% of the population out of the private housing market, and into public housing?
The justification of using public services to pursue policy goals via the existence of private alternatives develops problems when the government that provides those public services is *also* the one who gets to dictate terms to its competitors. It creates a direct path to increased power by creating costly obstacles, and generally making the world worse in ways it can weather or ignore entirely.
If you force all kids to go to the same terrible schools, everyone will suffer, and the kids who actually want to learn, regardless of the family income, will suffer most. But the well-off parents will buy extra education outside of school, whereas the poor parents will just have to let their kids suffer and not learn. So the only thing you'd be achieving would be denying poor kids, no matter how talented and willing to learn, the chance to get an education.
There are way not enough charter schools, so this is already happening over most of the country anyway.
>Rich people care deeply about their children’s education, allowing them to not encounter the same problems as everyday citizens allows them to not have to care nor expend power to improve the system for themselves in a way that also helps others.
Yes, we get that you want to hold rich people's children hostage to their parents using their wealth in the manner you deem most socially beneficial. We've always understood this; you don't need to explain it.
It doesn't work. "Rich" people, and even more so the upper-middle-class people you're really going after, don't each individually have the power to substantially improve educational outcomes across the nation, or even a modest city. If you hold their children's education hostage to general educational improvements nationwide, or citywide or whatever, they will each and all of them hope that all the other people will act in a manner to generally improve educational outcomes.
And then each of them will privately defect and devote their own personal efforts to whatever it is they can do to improve the education of their own particular children. Having the wife stay home and do after-school tutoring, moving to a small town with good schools and no practical way to bus the students to the big city, paying a doctor to diagnose their kid with some bogus disability that only UMC children seem to get so that they can go to the "special" school that teaches only not-really-disabled UMC children, whatever.
Parental love is a powerful resource that can easily be channeled to improve educational outcomes in *many* schools, leaving you with the hard problem of fixing *some* broken schools. If you try to make the love of UMC parents fix all the schools, you break the connection and the parental love is refocused elsewhere and now you've got to fix *all* of the schools with the weaker tools remaining to you.
Also, it's morally wrong and abhorrent of you to decide that this is something you get to "allow" parents to do, or not. It's their choice, not yours.
Also, there's a matter of time. Children are going *through* the school system, so the the problem parents see in first grade isn't going to be solved while the kid is in first grade.
I'll make an exception for the condition of school buildings, but even that is unlikely to be fixed while the kid is still in the same school.
This comment is so uncharitable as to be non-productive. I think it fails to be either necessary or kind, particularly the first and last paragraphs.
On the substantive side, I am unclear about your claims. Are you claiming that well-off parents would help their children only in ways that did not help other students or in addition to doing so in ways that would help other students? If it is the latter, how does that run counter to wpow's argument? If the former, wouldn't that also suggest that such parents would currently not contribute to the school at large? If not, what am I missing? If so, would the theory then be falsifiable if we could show that those parents do contribute to the school at large (I'm thinking about things such as being more likely to volunteer, paying higher rates of school taxes, and contributing in forms like bake sales)? I do not have data on hand to show that this is true, but I strongly suspect that it is.
Hey Scott, when are you (will you ever) get around to putting out Unsong in print form?
The best recent popular-science book out right now on modern biology of ageing research is Andrew Steele's (2020) "Ageless: The New Science of Getting Older Without Getting Old." I'm only a few chapters in but already confident saying that. (I study the evolutionary-theory side of ageing as a master's student.)
It's accessible, leaving unnecessary technical details unspecified (but also not giving faux-precise "lies-to-children" substitutes for them*), but covering all the right stuff at the right level for pop science: e.g. the first chapter introduces long-lived species and talks about the demography of aging (increase, roughly exponential, in the risk of death over time) and how life expectancy and the age distribution has changed over the 20th century. The second chapter talks about the evolutionary theory (Medawar's "mutation accumulation" theory, Williams' "antagonistic pleiotropy" theory, the "disposable soma" version of the antagonistic pleiotropy theory), and also corrects the fallacy in thinking aging is "just" entropy. The third chapter tells about how modern biogerontology was born out of studying the mechanisms by which life was extended in model organisms by the Calorie Restriction diet (first observed in rats in the 30s) and singe-gene longevity mutations (first seen in C. elegans nematodes in the late 80s / early 90s). Etc.
Its account of the mechanistic side of ageing (introduced in chap. 4) is based on the mainstream "hallmarks of ageing" framework (cf. https://dx.doi.org/10.1016%2Fj.cell.2013.05.039), also giving a shoutout to its dad, Aubrey de Grey's SENS approach: this categorizes those age-related changes, that we have good reason to suspect are causal, into a modest number of categories based on what kind of process they are and how they might be treated. This gives a semblance of order to the fact salad that was the biology of aging, at least useful for learning, and also likely useful if we do ever manage to treat aging in humans. Steele uses a slightly different list of categories than the usual hallmarks list (merging two and introducing two, for a total of ten hallmarks). Steele's list is:
1. Trouble in the double helix: DNA damage and mutations;
2. Trimmed telomeres;
3. Protein problems: autophagy, amyloids and adducts;
4. Epigenetic alterations;
5. Accumulation of senescent cells;
6. Power struggle: malfunctioning mitochondria
7. Signal failure;
8. Gut reaction: changes in the microbiome
9. Cellular exhaustion;
10. Defective defences: malfunction of the immune system.
I'm partway through chapter 4 now, and excited to read the rest.
---
*e.g. he talks about ageing as a roughly exponential increase in the risk of death w.r.t. age, over much of the adult lifespan. The precise version of this would mention the "Gompertz equation," and that "risk of death" in the Gompertz equation is defined as the "hazard rate" or "force of mortality" (instantaneous relative rate of decrease of the survival function; i.e. on a graph of a newborn's probability of still being alive at each age x, the hazard rate is the slope, divided by the height, times minus one); the hazard rate is not a probability, and in principle has no upper bound, though its growth in fact may or may not turn out to asymptote in some or many species. The hazard rate is sometimes conflated, either accidentally or as a lie-to-children, with the per-interval probability of death, which is a probability and has an upper bound of one. Even some textbooks accidentally say that the Gompertz equation says the probability of death grows exponentially (it couldn't possibly, at least not forever)—but Steele avoids that mistake.
I noticed a while ago that the people who got eulogies in Tai Chi Magazine tended to live into their late 80s or early 90s. Not bad, but not impressive for a philosophy which (at least in some versions) aimed at greatly extended lifespan.
In a recent Bruce Frantzis workshop, he mentioned that people who study bagua and hsing-i tend to live about 10 years longer than people who study. He attributed it to them working with their tendons as well as their ligaments.
Anyone have a sample of people who've done serious study of bagua and hsing-i?
Thanks for the recommendation, I downloaded the book on Audible
https://nationalinterest.org/feature/why-are-people-who-live-mountainous-regions-almost-impossible-conquer-30597
So, Fermi paradox time. Could an inhabitable planet have a much higher proportion of mountainous terrain than earth? Or a much lower proportion of mountainous terrain?
Would that affect the odds of a technological civilization which could get out into space?
If there are lots of aliens, it's quite possible that lots of them are trying to get into space from planets they can't launch from, for various reasons. Bad geography, bad gravity, bad atmosphere, whatever.
But for it to explain the Fermi Paradox, it has to apply to all civs.
However, enough medium-sized filters could add up to nothing getting through.
Is there any way to indicate blockquotes in a comment? Markdown's `>` prefix doesn't work, and googling variations of "substack comment syntax" is failing me.
I think Markdown's prefix is still the best option, as it's clean and used often and easy to interpret.
Who are the really fascinating people you follow, who are not related to each other? To keep an open-mind and a broad perspective on things? I'm looking for real quality material.
I follow:
Scott
Derek Sivers
Taleb
But I'm looking for more widely-different people
During WWII, the U.S. shifted almost all of its clothing production towards making military uniforms, creating shortages of new civilian clothing. At the same time, worn-out or damaged uniforms at the front lines were thrown away. Why didn't the U.S. ship those garments back to the homeland again to clothe civilians? Old pants, shirts and jackets could be cleaned, patched up, and dyed to look almost new. There was no shortage of available shipping for this since the merchant ships went west across the Atlantic practically empty.
Can we add in the nutritional value of either choice?