All the news that’s fit to print, but has nowhere to go.
This important rule is a special case of an even more important rule:
Dirty Hexas Hedge: One of the old unwritten WASP rules of civilization maintenance we’ve lost is: when someone behaves insincerely for the sake of maintaining proper decorum, you respond by respecting the commitment to decorum rather than calling out the insincerity.
The general rule is to maintain good incentives and follow good decision theory. If someone is being helpful, ensure they are better off for having been helpful, even if they have previously been unhelpful and this gives you an opportunity. Reward actions you want to happen more often. Punish actions you want to happen less often. In particular beware situations where you punish clarity and reward implicitness.
Another important rule would be that contra Elon Musk here you shouldn’t ‘sue into oblivion’ or ‘ostracize from society’ anyone or any organization who advocated for something you disagree with, even if it plausibly led to a bad thing happening.
Even more importantly: When someone disagrees with you, you don’t use the law to silence them and you most definitely don’t choose violence. Argument gets counterargument. Never bullet, no arrest, no fine, always counterargument. I don’t care what they are advocating for, up to and including things that could plausibly lead to everyone dying. It does not matter. No violence. No killing people. No tolerance of those who think they can have a little violence or killing people who disagree with them, as a treat, or because someone on the other side did it. No. Stop it.
This seems like one of those times where one has to, once again, say this.
This seems like a lot of percents?
Benjamin Domenech: New death of the West stat: 42 percent of people in line to meet Buzz Lightyear at Disney theme parks last year were childless adults.
Source: author A.J. Wolfe on Puck’s The Town podcast.
PoliMath: When I went to Disney in 2019, my kids were in line to meet Sleeping Beauty and the guy in front of us was a 30ish single dude who gave her a bouquet of roses and weirdly fawned over her. I admired the actress for not displaying her disgust.
If that is who gets the most value out of meet and greets, okay then. It also presumably isn’t as bad as it sounds since it has been a long time since Buzz Lightyear has been so hot right now, I presume characters in recent movies have a different balance. The price sounds sufficiently high that they should add more copies of such characters for meet and greets until the lines are a lot shorter? How could that not raise profits long term?
A-100 Gecs (1m views): the pearl-clutching about no young white men being published in The New Yorker is so funny like men, writ large, are basically a sub-literate population in the US. men do not read literary fiction. if you have even a passing interaction with publishing you realize this.
Kelsey Piper:
Open disdain for people on the basis of their sex is bigoted and bad.
Men obviously did read and write literary fiction for most of the history of literary fiction; so, if that has changed, I wonder why it has changed! Perhaps something to do with the open disdain!
Like in general, I try not to waste too much of my time on “this hobby has too few Xs” or “this hobby has too few Ys,” since that can happen totally organically, and pearl-clutching rarely helps.
However, if the hobbyists are saying “our hobby has no men because they are a ‘subliterate population,’” then I suddenly form a strong suspicion about why their hobby has no men, and it’s not that people innocently have different interests sometimes.
John Murdoch gives us many great charts in the FT, but often we lack key context and detail, because as John explains he only has very limited space and 700 words and everything needs to be parsable by a general audience, so spending space on methodology or a full y-axis is very expensive. We appreciate your service, sir. It would still be great to have the Professional Epistemically Ideal Edition available somewhere, any chance we can do that?
Reading books for pleasure continues to decline by roughly 3% per year. Alternatives are improving, while books are not improving, indeed the best books to read are mostly old books. So what else would you expect? Until recently I would say people are still reading more because a lot of screen use is reading, but now we have the rise of inane short form video.
Madeleine Aggeler figures out very basic reasons why you might want to not be constantly lying, and that she would be better off if she stopped lying constantly and that you really can tell people when you don’t want to do something, yet she fails to figure out that not lying does not require radical honesty. You can, and often should, provide only the information needed.
The IQ tests we have are drawn from a compact pool of question types and so can, unsurprisingly, be trained for and gamed. If you want to raise the result of your IQ test this way, you can totally do that. Goodhart’s Law strikes again. That doesn’t mean IQ is not a real or useful thing, or that these tests are not useful measures. It only means that if you want to make the (usually low-IQ) move of pretending to be higher IQ than you are by gaming the test, you can do that. So you need to not give people strong incentive to game the tests.
I often hear discussion of ‘masking’ where autistics learn how to fake not being autistic and seem like normies, or similarly where sociopaths learn not to act like sociopaths (in the clinical sense, not the Rao Gervais Principle sense) and seem like normies, because they realize that works out better for them. I mention this because I notice I rarely hear mention of the fact that (AIUI) the normies are mostly doing the same exact thing, except that they more completely Become The Mask and don’t see it as a strange or unfair or bad thing to do this kind of ubiquitous mimicry, and instead do it instinctively?
Eugyppius: when you’re with girl, do not quietly remove bugs. call her attention to bugs first, then heroically remove them for her. they love this.
Lindy Man: This is also good advice for the workplace. Never fix anything quietly.
Sean Kelly: When I discover a bug and figure out the solution, I don’t fix it.
I have an accomplice report it and play up how bad it is in the stand up.
Then I sagely chime in, “I bet I can figure that one out.”
If the bug looks tough, the accomplice suggests the H-1B with the shortest queue.
Caroline: People will fix things quietly and then complain they’re underappreciated. Does anyone even know what you did? lol
Kyle Junlong: ah yes, “half the work is showing your work.”
honestly this is so powerful. i’m realizing how valuable communication and visibility is, not just in work but in relationships and life.
i used to think managing other people’s perception of me was stupid and frivolous, but now i realize how *ijudge other people is solely based on my perception (eg., the convenient information) i have of them. so of course it makes sense to present myself well, because i like those who do present themselves well to me.
Over time, if you don’t take credit for things, people notice that you silently fix or accomplish or improve things without taking credit or bothering anyone about it, and you get triple credit, for fixing things, for doing it seamlessly and for not needing or requesting credit. The problem is, you need a sufficiently sustained and observed set of interactions, and people sufficiently aware of the incentive dynamics here, so that you can move the whole thing up a meta level.
There is also the reverse. If you know someone who will always loudly take credit, you know that at most they are doing the things they loudly take credit for. If that.
I am generally skeptical that we should worried about inequality, as opposed to trying to make people better off. One danger that I am convinced by is that extreme inequality that is directly in your face can damage your mental health, if you see yourself in competition with everyone on the spectrum rather than being a satisficer or looking at your absolute level of wealth and power.
Good Alexander: I think the main reason you find a lot of very unhappy tech people even at the highest levels
– when you’re a typical employee everyone around you is making .5-2x what you are
– when you start breaking out wealth goes on log scale. ppl with 10-1000x your net worth become common
– this is native to network effects, scale associated with AI training, and other winner take all dynamics in tech
– all of VC is structured this way as well — (1 unicorn returns entire fund rest of investments are zero) which psychologically reinforces all or nothing thinking
– this makes competitive people miserable
– this leads them to do hallucinogens or other psychoactive substances in order to accept their place in the universe
– the conclusions drawn from these psychoactive substances are typically at direct odds with how they got to where they are
– and after getting one shotted they’re still ultimately in a hard wired competition with people worth 10-1000x more than them
– due to the structure of technology it becomes more or less impossible to break out of your ‘bracket’ without engaging in increasingly dark things
– you realize that time is running out — and become aware of synthetic biology (peptides, genetic alteration of children)
– you end up getting involved in police state investments, gooning investments, or crypto — and view it as non optional to take the gloves off bc everyone around you is doing the same thing
– you’re on a permanent hedonic treadmill and you can’t ever get off or go back to where you were before bc after doing all of the things you’ve done you can’t possibly ever relate to normal humans
– you get involved with politics or Catholicism or other Lindy cults to try and get off the treadmill
– of course it won’t work and you bring all the weird baggage directly into politics or religion and poison those wells too
the current configuration of economics/ wealth distribution is pretty solidly optimized to drive the wealthiest people in society batshit insane, which – to some extent – explains a lot of things you see around you
w this framework you can understand:
Thiel Antichrist obsession
Kanye getting into Hitler and launching a coin
Trump memeing himself into becoming President then running again to escape imprisonment
Elon generating Ani goon slop on the TL
A16z wilding out
Eliezer Yudkowsky: – supposed “AI safety” guys (outside MIRI) founding AI companies, some of whom got billions for betraying Good and Law.
I have felt a little pressure to feel insane about that, but it is small compared to all the other antisanity pressures I’ve resisted routinely.
David Manheim: This is definitely not wrong, even though it’s incomplete:
“the current configuration of economics/ wealth distribution is pretty solidly optimized to drive the wealthiest people in society batshit insane, which – to some extent – explains a lot of things you see around you”
Speaking from experience, it is quite the trip to be in regular contact and debates with various billionaires. It can definitely make one feel like a failure or like it’s time to make more money, even though I have enough money to not worry about money, especially when you think you definitely could have joined them by making different life choices, and there’s a chance I still could. Whereas, when in prior phases of life I was not in such contact, it was easy not to care about any of that.
It helps to remind myself periodically that if I had a billion dollars, I could make the world a better place, but except insofar as I prevented us all from dying my own life would, I anticipate, not actually be better as a result. At that level, there isn’t that much more utility to buy, whereas more money, more problems.
It’s not easy buying art.
cold: Bro you make $500k at OpenAI you can go to the art fair and buy a little $10,000 painting to hang up in your SF apartment’s living room
You tell them this and then they’ll be like “I’m sorry 🥺 do you think a $15,000 desert meditation retreat will fix me so I’m not like this anymore??”
Daniel: The lack of personal art purchasing in SF is insane. A $3000 oil on canvas can change your whole living room and they won’t do it.
I know people who earn much more than $500k at openai and their living rooms are making them depressed.
Paul Graham: The main reason rich people in SV don’t buy art is that it does actually take some expertise to do it well. And since the kind of people who get rich in SV hate to do things badly, and don’t have time to learn about art now, they do nothing.
diffTTT: Rich SV people need an expert to tell them what kind of art they like?
Paul Graham: In a way. They need to learn how not to be fooled by meretricious art, how to avoid the immense influence of hype and fashion, etc. Most people have to figure this out for themselves or from books, but a truly competent expert could help.
If it takes some expertise to buy art well, that is a real problem with buying art. The thing is, if you do not buy art well, you will lose most of the money you spent on art, and also you will look like a fool, and also the art will not make you feel better or much improve your living room.
That leaves four options.
The one these people and I have taken, which is to not buy art.
Buy cheap art that you don’t mind looking at. Safe, but still annoying to do, and then you have to look at it, does it actually make you feel better?
Spend a lot of time figuring out how to buy expensive art properly. Yeah, no. I understand that Paul Graham can be in renaissance man mode, but if you are coding at OpenAI at $500k+ per year the cost of this is very, very high, and also you probably don’t expect the skill to stay relevant for long.
Find someone you trust to do it for you? Not cheap, not all that easy or quick to do either, and you are still the one who has to look at the damn thing.
Besides, who is to say that a constant piece of artwork actually helps, especially if it doesn’t hold particular meaning to you? I mean, yeah, in theory yeah we should get some artwork here, I suppose, but no one wants to do the work involved, also it should definitely be cheap art. At one point I bought some Magic: The Gathering prints for this but we never got around to hanging them.
Also at one point I tried to buy the original art for Horn of Greed, which at the time would have cost like $3k. I say tried because my wife wouldn’t let me, but if anyone wants to buy me a gift at some point, that or another original Magic art I’d look back on fondly seems great.
If there is one thing to learn from rationality: Peter Wildeford is 100% right here.
Wikipedia (Wet Bias): Wet bias is the phenomenon whereby some weather forecasters report an overestimated and exaggerated probability of precipitation to increase the usefulness and actionability of their forecast.
The Weather Channel has been empirically shown, and has also admitted, to having a wet bias in the case of low probability of precipitation (for instance, a 5% probability may be reported as a 20% probability) but not at high probabilities of precipitation (so a 60% probability will be reported as a 60% probability).
Some local television stations have been shown as having significantly greater wet bias, often reporting a 100% probability of precipitation in cases where it rains only 70% of the time.
Colin Fraser: If you believe it will rain with probability P, and getting caught in the rain without an umbrella is X times worse than getting caught in the sun with an umbrella, then it’s optimal to predict rain whenever P ≥ 1/(1+X). So e.g. for X=2 you should predict rain at P ≥ 1/3.
Peter Wildeford: I think you should only predict rain according to the correct p(rain), but you can change your behavior around umbrella carrying at lower values of p(rain).
Colin Fraser is right if you effectively can only predict 0% or 100% rain, and the only purpose of predicting rain is that you take an umbrella if and only if you predict rain.
Peter Wildeford is right that you can say ‘it will rain 40% of the time, therefore I should take an umbrella, even though more than half the time I will look foolish.’
The weather reports are assuming that people have a typical bias, that people respect 40% chance or more, but not 30%. Thus, there is a huge jump (AIUI, and Claude confirms). If the Google weather app says 30%, treat that as at most 10%, but the app doesn’t want to get blamed so it hedges. Whereas if it says 40%? That’s pretty much 40%, act accordingly.
If you don’t know the way the conversion works, and you don’t have the typical biases, you’ll respond in crazy wrong fashion. The bias and nonlinearity become self-perpetuating.
The right rule in practice really is to take the umbrella at 40% and not at 30%, almost no matter what the cost-benefit tradeoff is for the umbrella, since it is obviously wise at 40% and obviously unwise at 10%.
The ‘predict 100% instead of 70%’ thing that other sources do is especially maddening. This means both that you can’t tell the difference between 70% and 100%, and that on the regular you notice things that were predicted as Can’t Happen actually happening. The weather forecaster is consigned to Bayes Hell and you can’t trust them at all.
As Bryan Caplan notes, this is frequently a better question than ‘if you’re so smart, why aren’t you rich?’ It is a very good question.
His rejections of various responses are less convincing, with many being highly oversimplifying and dismissive of many things people often care about deeply. He presents answers as if they were easy and obvious when they are neither of those things.
I endorse rejection of the objection ‘the world is so awful that you have to be stupid to be happy.’ I don’t endorse his reasoning for doing so. I do agree with him that the world is in a great spot right now (aside from existential risks), but I don’t think that’s the point. The point is that it isn’t improving your life or anyone else’s to let your view of the world’s overall state make you indefinitely unhappy. If you think you ‘have’ to be stupid not to be perpetually unhappy for whatever external reason, you’re wrong.
I also agree with him that one good reason to not be happy is that you are prioritizing something else. He retorts that few people are extreme Effective Altruists, but this is not required. You don’t have to be some fanatic, to use his example, to miserably stay together for the kids. What you care about can be anything, including personal achievements other than happiness. Who says you have to care about happy? Indeed, I see a lot of people not prioritizing happiness enough, and I see a lot of other people prioritizing it far too much.
There’s also another answer, which is that some people have low happiness set points or chemical imbalances or other forms of mental problems that make it extremely difficult for them to be happy. That’s one of the ways in which one can have what Bryan calls ‘extraordinary bad luck’ that you can’t overcome, but there are other obvious ways as well.
From a public choice standpoint I suppose this was inevitable. However, this phases out at higher income levels, which means that none of the prominent people you are thinking of likely can benefit from this. As in, Republicans proudly embraced progressive taxation to partially offset regressive tariffs? So yes, I do accept tips, and very much appreciate them along with subscriptions, but alas after consulting with my tax lawyer (GPT-5 Pro) I have concluded that I cannot benefit from this policy.
I think I agree that in a purely aesthetic sense people did dress ‘better,’ but that is because people in the past put massive investment into this. They spent a huge percentage of their income on clothes, they spent a large percentage of their time and attention on understanding, creating and maintaining those clothes, and they were willing to suffer a lot of discomfort. And they faced huge social and status pressures to devote such efforts, with large punishments for not measuring up to norms.
Derek notes our reduced tolerance for discomfort and lack of effort, but skips over all the extra money and time and cognitive investments, seems to lack the ‘and this is good, actually’ that I would add. I think it’s pretty great that we have largely escaped from these obligations. The juice is not worth the squeeze.
Santi Ruiz interviews Dr. Rob Johnston on the intelligence community and how to get good intelligence and make good use of it. There’s lots of signs of a lot of deep competence and dedication, but clearly the part where they deliver the information and then people use it is not going great. Also not going great is getting ready for AI.
Texas becomes the seventh state to ban lab grown meat. Tim Carney becomes the latest person to be unable to understand there could be any reason other than cronyist protectionism to want this banned. Once again I reiterate that I don’t support these bans, but it seems disingenuous to prevent not to understand the reasons they are happening. James Miller offers a refresher of the explanation, if you need one, except that the demands wouldn’t stop with what Miller wants.
No, the Black Death was not good for the economy, things were improving steadily for centuries for other reasons. As opposed to every other pass famine or plague ever, where no one looks back and says ‘oh this was excellent for economic conditions.’
It is relatively easy to stay rich once already rich. It is not easy to get rich, or to be ‘good at’ being rich. It is also hard to be rich effectively, including in terms of turning that extra money into better lived experiences, and ‘use the money to change the world’ is even harder.
Roon: its amazing how little the post-economic people i know spend. many people are bad at being rich. you should teach them how to do it.
i think this is often why the children of the mega-rich are the ones who even get close to squandering their parents’ fortunes. when you get rich later into life you often don’t think with enough 0s in terms of personal consumption, donations, having a lavish household staff etc.
Eliezer Yudkowsky: In their defense, once you’ve already got your personal chef, volcano lair with trampoline, and a harem that covers all your kinks, there’s just not much else Earth offers for converting money to hedons.
Zvi Mowshowitz: It is remarkably difficult (and time consuming!) to spend large amounts of money in ways that actually make your life better, if you’re not into related status games.
I got into a number of arguments in the comments, including with people who thought a remarkably small amount of money was a ‘large amount.’ Certainly you can usefully spend substantially more than most people are able to spend and still see gains.
Reasonably quickly you hit a wall, and the Andy Warhol ‘everyone gets the same Diet Coke’ problem, and to do better you have to spend obscene amounts for not that much improvement. Are you actually going to be that much happier with a giant yacht or a private jet? What good is that private chef compared to Caviar or going to restaurants? Do you actually want to live farther away in some mansion? Does expensive art do anything for you cheap art doesn’t? And so on.
Even in the ways that would be good to spend more, you still need to know how to spend more and get value from what you paid, and how to do it without it taking up a ton of your time, attention or stress. Again, this is harder than it sounds.
We talk constantly about ‘losing to China’ whereas in China there are reasons to worry that China is losing, not to an outside force but rather in general, and this is on top of a fertility rate that is 1.1 or below and an already declining population:
Mike Bird: Useful chart from @andrewbatson here covering one of the most under-discussed and useful macro metrics around. China’s capital productivity has been in consistent, marked decline even as panic over Chinese industrial prowess has reached fever pitch
Indeed as of 2019-2023 China’s marginal product of capital (basically how much output you’re getting from another unit of capital) was only very slightly higher than that of the US, though the US is at the frontier of GDP per capita and China nowhere near.
Clearly both things can be true – that some of China’s leading industrial firms are incredibly impressive, world-leading, and that in the aggregate they’re not enough to offset the misallocation and inefficiency elsewhere.
He quotes that the manufacturing share of GDP in China, for all our worries about Chinese manufacturing, declined 2-3 percent between 2021 and 2025, with the sector now having narrower margins, lower profits and more losses.
All of this is more reason not to give them opportunity to invest more in AI, and also reason not to catastrophize.
Cate Hall thanks you for coming to her TED talk, ‘A Practical Guide To Taking Control of Your Life.’ Which is indeed the Cate Hall TED Talk you would expect, focusing on cultivating personal agency.
In my startup roundups I muse about why don’t startups offered TMM (too much money, here presumably also at too high a valuation) take the money and then set expectations accordingly? A commenter pointed out that Stripe did do a version of this, although it is not a perfect fit.
Motivation and overcoming fear are tricky. You can get people comfortable with public speaking with complements. You can also do it by having people starting with you come up and give intentionally terrible speeches while they get crumpled papers thrown at them, to show that nothing actually bad happens.
When people rate their happiness, are they rating on an absolute scale that reflects a real treadmill effect, or are people simply asking if they are happy compared to what they know?
It seems obviously possible to raise national happiness. One existence proof is that there are very clearly regimes, policies and circumstances that make people very unhappy, and you can do the opposite of those things, at least dodging them.
Also there are things that consistently raise happiness and that vary in frequency greatly over time, for example being married and having grandchildren.
Abstract: We revisit the famous Easterlin paradox by considering that life evaluation scales refer to a changing context, hence they are regularly reinterpreted.
We propose a simple model of rescaling based on both retrospective and current life evaluations, and apply it to unexploited archival data from the USA.
When correcting for rescaling, we find that the well-being of Americans has substantially increased, on par with GDP, health, education, and liberal democracy, from the 1950s to the early 2000s.
Using several datasets, we shed light on other happiness puzzles, including the apparent stability of life evaluations during COVID-19, why Ukrainians report similar levels of life satisfaction today as before the war, and the absence of parental happiness.
Tyler Cowen: To give some intuition, the authors provide evidence that people are more likely engaging in rescaling than being stuck on a hedonic treadmill. I think they are mostly right.
This makes tons of sense to me. You get revolutions of rising expectations. There are definitely positional effects and treadmill effects and baseline happiness set points and all that to deal with, but the Easterlin Paradox is a paradox for a reason and things other than income vary as well.
That doesn’t mean life is getting better or people are getting happier. It can also go the other way, and I am very open to the idea that happiness could be declining (or not) in the smartphone era with kids not allowed to breathe outside, and everything else that causes people to feel bad these days both for good and bad reasons. But yeah, from the 1950s to the 1990s things seem like they very clearly got better (you could also say from the 1500s to 1990s, with notably brief exceptions, or earlier, and I’d still agree).
Camp Social is part of a category of offerings where adults go to sleepaway camp with a focus on making friends, complete with bunk beds and color wars and in one case a claimed 75% return rate, although also with staying up until 1: 30 getting drunk. The camp counselors are concierges and facilitators. Cost is $884 for two nights and three days, which seems rather quick for what you want to accomplish?
I do buy that this is a good idea.
Radiation is dangerous, but it is a lot less dangerous than people make it out to be, and we treat this risk with orders of magnitude more paranoia than things like ordinary air pollution that are far more deadly.
Ben Southwood: The life expectancy of someone hit with 2,250 millisieverts of radiation in Hiroshima or Nagasaki was longer than the average Briton or American born in the same year. Today in Britain we spend billions controlling radiation levels more than 100,000 times smaller than this.
2,250 millisieverts is a lot of radiation, like getting 225 full-body CT scans in one go. I don’t think anyone would recommend it. But it shows how ridiculous it is that we spend so much time, effort, and money on radiation levels of 1msv or 0.1 msv per year.
Andrew Hammel reports that the Germans are finally on the verge of losing their War on Air Conditioning, as in allowing ordinary people to buy one, because normies actually experienced air conditioning and are not idiots. The standard ‘urban haute bourgeoisie’ are holding out on principle, because they think life is about atoning for our sins and because they associate things like air conditioning with wasteful Americans. As you would expect, the alternative ‘solutions’ to heat wind up being exponentially more expensive than using AC.
I do note that they have a point on this one:
Andrew Hammel: First of all, *every oneof these people has a story about visiting the USA and nearly freezing to death in an over air-conditioned store or office. Every. Damn. One. I can predict exactly when they will wheel out this traumatic tale, I just let it unfold naturally.
I mean, I have that too, to the point that it is a serious problem. This happens constantly in Florida. Even in New York’s hotter summer days, I have the problem that there is nothing I can wear outside while walking to the restaurant, that I also want to be wearing once I sit down at the restaurant. It is crazy how often Americans will use the AC to make places actively too cold. We could stand to turn it down a notch.
Or rather, ‘the’ good news, as Elizabeth Van Nostrand lays out how Church Planting works and finds it very similar to Silicon Valley startups.
A counterargument to last month’s claim about rapidly declining conscientiousness. Conscientiousness has declined far more modestly, the decline here is still seems meaningful but is very is not be a crisis. What John did to create the original graph turns out to have been pretty weird, which was show a decline in relative percentile terms that came out looking like a Really Big Deal.
Cartoons Hate Her!is on point that germs are very obviously real and cause disease but quite a lot of people’s specific worries about vectors for being exposed germs and the associated rituals are deeply silly if you stop to think about physics, especially compared to other things the same people disregard.
Indeed, we should be putting all the old PBS kids shows on YouTube, and everything else that it would be good for kids to be watching on the margin. The cost is low, the benefits are high. There are low quality versions of the shows of my extreme youth available (such as Letter People and Square One TV) but ancient-VHS quality is a dealbreaker for actually getting kids to watch.
What TV show had the worst ending? There are lots of great answers but the consensus is (in my opinion correctly) Game of Thrones at #1 and Lost at #2.
After that it gets more fractured, and the other frequent picks here I am in position to evaluate were mostly bad endings (HIMYM, Killing Eve, Enterprise, Battlestar Galactica) but not competitive for the top spot. Dexter came up a lot but I never watched. Supernatural came up a bunch, and I’m currently early in its final and 15th season and is it weird this makes me want to get to the end more not less? Better a truly awful end than a whimper?
To be the true worst ending, it has to not only be awful but take what could have been true greatness and actively ruin the previous experience. You need to be in the running for Tier 1 and then blow it so badly you have to think about whether it even stays in Tier 2 because they poisoned everything. That’s why Game of Thrones and Lost have to be so high.
Indeed those two are so bad that they substantially hurt our willingness to invest in similar other shows, especially Lost-likes, which is enforcing the good discipline of forcing for example Severance to assure us they have everything mapped out.
(Briefly on the others: While at the time I thought HIMYM’s ending was as bad as everyone thinks, on reflection it has grown on me and I think it is actually fine, maybe even correct. Killing Eve’s ending wasn’t good exactly, but I didn’t feel it ruined anything, it was more that all of season 4 was a substantial decline in quality. Battlestar Galactica was rage inducing but I understand why they did what they did and that mostly made it okay, again mostly the show started fantastic and was dropping off in quality generally. Enterprise ended bad, but again not historically bad, whereas the show wasn’t getting bad, and mostly the frustration was we weren’t done.
I heard the claim recently that Lost’s ending is aging well, as it suffered from the writers assuring us that they wouldn’t do the thing they did, whereas now looking back no one much cares. There’s that, but I still find it unsatisfying, they said they wouldn’t do it that way for a reason, and the worse offense was the total failure to tie up loose ends and answer questions.
Scott Sumner: The public prefers 1980-2015, as you say. The movie experts say the 1920s-1970s were the best.
This highlighted the ways in which our preferences strongly diverge.
Another big hint is that Sumner and the experts claim an extremely high correlation of director with quality of movie. Great directors are great, but so many other things matter too.
As an example, recently I watched Mulholland Drive for the first time, which Sumner says might be his favorite film. I appreciated many aspects of it, and ended up giving it 4/5 stars because it was in many senses ‘objectively’ excellent, but I did not actually enjoy the experience, and had to read an explainer afterwards from Film Colossus to make sense of a lot of it, and even after understanding it a lot of what it was trying to do and say left me cold, so I didn’t feel I could say I ‘liked’ it.
From what I can tell, the public is right and the ‘experts’ are wrong. Also I strongly suspect that We’re So Back after a pause of timidity and sequilitius and superheroes.
Scott Sumner: There are two films that should never, ever be watched on TV. One is 2001 and the other is Lawrence of Arabia. If you saw them on anything other than a very big movie theatre screen, then you’ve never actually seen them.
I haven’t seen 2001 regardless, but on Lawrence of Arabia I can’t argue, because I attempted to watch it on a TV, and this indeed did not result in me seeing Lawrence of Arabia, because after half an hour I was absolutely bored to tears and could not make myself continue. There was a scene in which they literally just stood there in the sand for about a minute with actual nothing happening and I get what they were trying to do with that but it was one thing after another and I couldn’t even, I was out.
What I am confused by is how it would have improved things to make the screen bigger, unless it would be so one would feel forced to continue watching?
Here are his 13 suggestions for films to watch, although I have no idea how one would watch Lawrence of Arabia given it has to be on a big screen?
Vertigo, The Man Who Shot Liberty Valance, Touch of Evil, Some Like It Hot, Breathless, Jules and Jim, Last Year in Marienbad, High and Low, The End of Summer, 8 1/2, L’Avventura, The Music Room, Lawrence of Arabia.
I tried to watch High and Low, and got an hour in but increasingly had the same sense I got from The Seven Samurai, which is ‘this is in some objective senses a great movie and I get that but I have to force myself to keep watching it as outside of moment-to-moment it is not holding my interest’ except with more idiot plot – and yes I realize some of that is cultural differences and noticing them is the most interesting thing so far but I’m going to stick with idiot plot anyway. In addition to the idiot aspects, it really bothers me that ‘pay or pretend to pay the ransom’ is considered the obviously moral action. It isn’t, that is terrible decision theory. The moral action is to say no, yet there is not even a moment’s consideration of this question by anyone.
If the above paragraph is still there when you read this, it means I was unable to motivate myself to keep watching.
Jeff Yang explains some of the reasons Chinese movies tend to bomb in America, in particular the global hit Ne Zha 2. Big Chinese movies tend to be based on super complex Chinese traditional epic stories that Chinese audiences already know whereas Americans haven’t even seen Ne Zha 1. American stories have clear structure, understandable plots, payoffs for their events, central characters, and a moral vision that believes in progress or that things can be better. And they try to be comprehensible and to maintain a tonal theme and target market. Chinese movies, Yang reports, don’t do any of that. Effectively, they assume the audience already knows the story, which is the only way they could possibly follow it.
It’s as if Marvel movies were the big hits, and they didn’t try to be comprehensible to anyone who didn’t already know the characters and comics extensively? Certainly there are some advantages. It might be cool to see the ‘advanced’ directors cuts where it was assumed everyone had already either read the comics extensively or watched the normal version of the film?
As Jeff says, if they can make money in China, then sure, why not do all this stuff that the Chinese audiences like even if it alienates us Westerners. There are enough movies for everyone. It does still feel like we’re mostly right about this?
Like everyone else I think Hollywood movies are too formulaic and similar, and too constrained by various rules, and thus too predictable, but those rules exist for a reason. When older movies or foreign movies break those rules, or decide they are not in any kind of hurry whatsoever, it comes at a cost. I don’t think critics respect those costs enough.
I strongly agree with Alea here and I am one of the ones who want to stay away:
Alea: Novels with an empty mystery box should be explicitly tagged so I can avoid them. 110% of the joy of reading comes from uncovering all the deep lore and tying up every loose end. Some people get off on vague worlds and unfinished plots, and they should stay the fuck away.
I don’t especially want to go into deep lore in my spare time, but if you are going to convince me to read a novel then even more than with television you absolutely owe it to me to deliver the goods, in a way (with notably rare exceptions) that I actually understand when reading it.
As in: I know it’s a great book but if as is famously said, ‘you don’t read Ulysses, you reread Ulysses’ then you had me at ‘you don’t read Ulysses.’
And you definitely don’t read Game of Thrones until I see A Dream of Spring.
True facts about the whole ‘fleeing Earth’ style of story:
Ben Dreyfuss: The stupidest part of INTERSTELLAR is that the blight starts killing all the crops and after just a few decades they go “ah well, guess it won! Better leave earth. Hope we solve this magic gravity equation with the help of 5 dimensional beings and wormholes.”
“We can’t make okra anymore. Better go explore this all water planet where one hour is 7 years of time and this ice planet where water is alkaline and the air is full of ammonia.”
Pretty sure you can’t make okra there either, buddy.
Kelsey Piper: every single movie about people fleeing Earth involves displaying a mastery of technology which would obviously be more than sufficient to solve the problem they are fleeing Earth about
climate change is not going to make Earth less habitable than Mars so you can’t have people fleeing to Mars because of climate change, you just can’t.
‘there’s a supervolcano/asteroid induced ice age’ oh boy I have some news for you about Mars.
Daniel Eth: Just once I want a movie about people fleeing Earth to have the premise “there are trillions of people, and we have a per capita energy consumption of 100,000 kWh/yr, which is straining Earth’s ability to radiate the waste heat. We must now go to space to expand capacity”
Movie could have a real frontier vibe (space cowboys?) – “of course back in the old world (Earth), population and energy per capita are subject to bureaucratic regulations to prevent total ecosystem collapse; but in new worlds we can freely expand anew”
A recent different case of ‘I can’t help but notice this makes no sense’ was Weapons. The link goes to a review from Matthew Yglesias that I agree with, it does cool things with nonlinearity and the performances and cinematography are good except when you put it together in the second half the resulting actual plot, while consistent and straightforward, makes no sense.
Zvi Mowshowitz reviews Weapons while avoiding spoilers: When you’re in, writing or deciding to go to a horror movie, you make dumb decisions. It’s what you do.
The difference is to him this adds up to 3.5 stars, and to me it means 2.5 stars, once the holes and idiot balls became too glaring, I stopped being able to enjoy the film.
My other problem with Weapons was that the first two acts made me care about various characters and relationships that were rich and detailed and well-executed and acted, and then the third act didn’t care at all about those things, only about the main plot that did not make any sense. There might actually be a pretty great movie here in which the missing kids are a tragedy that never gets explained or solved because what matters is a different third act that focuses on how people react to it.
Erik Gibbs: The bill’s language defines a micro bet as any live proposition bet placed during an event that pertains specifically to the outcome of the next discrete play or action.
This restriction seems clearly good. I don’t know where the line should be drawn, but I am confident that ‘ball or strike’ bets are over the line.
It is a very light restriction – you can’t bet on a ball or strike or pass or run under this rule, but you can still bet on the outcome of an inning or drive. Bets on the next play have all the worst gambling attributes. They cost a lot individually, they resolve and compound super quickly, they are especially prone to addictive behavior.
Clair Obscur Expedition 33 finishes at Tier 2. It does a lot of things very right and I am very happy to have played it, despite some obvious issues, including some serious balance problems in Act 3.
If someone suddenly buys up the contract on Taylor Swift and Travis Kelce getting engaged from 20% to 40%, and you’re selling into it, yeah, good chance they know. Also this means yes, someone knew and traded on the information in advance. Cool. Oh, and congratulations to both of them, of course.
I don’t understand why Wizards of the Coast continues to be so slow on the banhammer in situations like Cauldron. We saw repeatedly exactly the broken format pattern, such as here where Cauldron starts out at 30% and then goes to 56% after six rounds, then a much larger majority of the top 8s. This continued long past the point where there was reasonable hope it would be fixed by innovation.
Mason Iange: Doesn’t it make sense that in a rotating format like standard, wotc wants people to have confidence in buying product and building decks? Literally no one is going to play standard if the best decks just get banned every 2 month.
Saffron Olive: The way I see is it there are two paths: you design cards conservatively and don’t need to ban anything, or you design cards aggressively and need to ban cards fairly often. Wizards is trying to design cards aggressively and never ban anything, which I don’t think is actually possible.
Patrick Sullivan: What you’re saying is true/relevant, but there are other considerations; the current state of affairs would clearly not be tolerated absent the stuff you’re mentioning. That’s why I think they should allow themselves to be as agile as possible regardless of what they decide to do.
Brian Kowal: The opposite of how I feel about rotating formats. I want it dynamic. I’d rather they make a balanced format. With a rotating format I want to feel like I can innovate. If it is solved I quickly lose interest. There are many formats that never rotate to protect investment.
I think ‘mix it up every time it is solved’ is going too far given how quickly we now solve formats, but the solution has to not be ‘play this deck or else.’ Yes, banning the best deck every two months would make you reluctant to invest in Standard, but effectively banning all but the best deck for months on end, or having to face an endless stream of the same overpowered nonsense even if you’re willing to sacrifice win rate to go rogue, is even worse.
They came out with an explanation and update on the 9th. A big part of this is that they screwed up timing the ban windows, and have a crazy high bar for doing “emergency” bans versus bans on announcement days. They are mitigating this going forward by adding more windows next year, one every major set release.
That points out how crazy the situation was. You’re going to release a set, and then not have a ban opportunity until after releasing the next set? That’s crazy.
Based on past experiences, I believe Brian Kowal is correct that an extended period of a miserable format, with bans that everybody knows have to happen but are extensively delayed, creates a point of no return, where permanent damage to the format and the game begins to accumulate.
Brian Kowal: There should be room in the ban policy for emergency bans. Perception has hit the point of no return. A significant portion of players do not want to touch Standard now. Rotation should be when we are creating players for the next year and this rotation lasts until January 2027! (I’m 80% on this. Somebody let me know if I’m wrong) Players are quitting Standard again to look for other games and formats. New players are choosing not to invest in it.
When format perception hits this state everybody knows something is getting a ban. So a lot of die hard competitives are even taking a break rather than buying 4 copies of the two most expensive cards in Standard.
The best way to go imo is to just suck it up and ban Cauldron immediately. Again, we all know it is happening anyway. Not taking action over and over again and just letting everybody suffer months of a bad format makes WOTC look like they don’t care.
Jenny: WotC took HOW long to decide to do nothing and ruin another Spotlight Series and RCQ season? Using the Arena ladder meta to judge the health of the format is *insane*
Pactdoll Terror: My 2-slot RCQ this Saturday in NYC sold 8 spots. I usually do 50. Someone who built Vivi to grind RCQs would be annoyed that it got banned, but Standard is DEAD locally. Weeklies aren’t launching, RCQs struggle to make money. Holding bans is bad for everyone except Vivi players.
Instead they’re going to do a strange compromise, and move up their next announcement date from November 24 to November 10, which still leaves two full months of this.
We should never have more than a month ‘in limbo’ where things are miserable and we know what is coming. Even if you decide to keep playing you are in an impossible position.
They say ‘Standard has not yet reached its final form’ but they are grasping at straws.
They say the Arena ladder is looking less awful. The Arena ladder is not real life, not only because the play level is low but also there’s nothing forcing the players to play the best deck. I learned that the hard way during the Oko era.
I get Carmen’s argument here that we ran the experiment and when you don’t have ban windows, you get constant speculation about potential bans and a lot of uncertainty, And That’s Terrible. You can’t fully embrace The Unexpected Banning. There needs to be a substantially higher bar outside of a fixed set of days.
The current situation was still ludicrous. While insufficiently competitive play is not as lopsided, that’s largely about card access and players wanting to have fun and of course not wanting to do this into a future ban. This ban would not be ‘from under players in a surprise move’ even if no formal warning was given. The idea that ‘we won’t make a move based on competitive play, only on non-competitive play, you competitors don’t much matter’ is definitely giving me even less desire to come back.
Which of course I get. Magic is not made for me. I’m just deeply sad about it.
I see the argument this isn’t a pure ‘do it today or else’ situation but it is an emergency. If I was Wizards, the moment it was clear we probably had a problem I would have created a new announcement date much closer in the future than two months, with the clear statement that at that time they would choose whether to ban Vivi, Caldron, both or neither. And then done it by now.
Sam Black: The house games are more fun when we don’t play for draws. Similarly games in top 16 are more fun.
Ultimately, I don’t think any solution would satisfy me, since it is going to come down to pure politics and kingmaker decisions. One potential approach is to say that wins are 10, draws are 1, and we pair people accordingly, so taking the draw is not obviously good for you, it might be wiser to lose and get paired against others who aren’t playing for draws. In the 0-0-4 bracket I don’t like your winning chances, and you have to win at some point to make the cut.
The exception (he seems to assumes in the modern era this won’t ever happen, which seems wrong to me) is if you are in danger of not having a deck, because you lack either enough cards or a key component, such that taking a usually bad card actually does provide substantial improvement.
Some cards that look bad, and have bad win rates, are instead good in the sense that they have high upside, but are being used badly by people who use them without the upside case. Sam’s example is a card that defaults to being a bad Divination but enables never running out of cards, so you can build your entire strategy around this, but if you put in your deck as a bad Divination then it will be bad.
As of the start of September, there are still only 2,000 Waymos: 800 in the San Francisco Bay Area, 500 in Los Angeles, 400 in Phoenix, 100 in Austin and ‘dozens’ in Atlanta.
As a point of comparison, San Francisco has ~1,800 taxi medallions, and an estimated 45,000 registered rideshare drivers, with Claude estimating there are typically 5,000 available rideshares at any given time, peaking in prime hours around 10,000.
This continues the recent trend of noticing that holding back self-driving means tens of thousands of people a year will die that didn’t have to.
Ethan Mollick: It seems like there is not enough of a policy response to the fact that, with 57M miles of data, Waymo’s autonomous vehicles experience 85% less serious injuries & 79% less injuries overall than cars with human drivers.
Think of EV policy and do long-term support: subsidies for R&D to bring down costs, incentives for including self-driving features, regulatory changes to make it easier to deploy, building infrastructure for autonomous-only vehicles (eg HOV lanes), independent testing.
Takes time.
There are many problems with this approach, including that it causes fixation on the lives saved versus cost and similar calculations, and also you sound like you are coming for people’s ability to drive. Whereas if you sell this purely as ‘Waymos are awesome and convenient and efficient and improve life greatly and also happen to be actually safe on top it’ then I think that’s way better than ‘you are killing people by getting in the way of this.’
Alice From Queens: Self-driving cars are like the new weight loss drugs.
Their value is so large, so obvious, and so scalable, that we can confidently predict their triumph regardless of knee-jerk cultural resistance and their wildly exaggerated downsides.
Yes, I’ve been saying the same thing for years. Because it still needs saying!
I mean, they totally are killing people by getting in the way, but you don’t need that.
Mostly you need to make people believe that self-driving is real and is spectacular.
Matthew Yglesias: I keep meeting people who are skeptical self-driving cars will ever happen.
I tell them I took one to the airport in Phoenix several months ago, did a test ride in DC, they’re currently all over San Francisco, etc and it’s blank stares like I’m telling them about Santa.
My model of what is holding things back for Waymo in particular right now is that mainly we have a bottleneck in car manufacturing, and there’s plenty of room to deploy a lot more cars in a bunch of places happy to have them.
Longer term, we also have to overcome regulatory problems in various places, especially foolish blue cities like New York and Boston, but I find it hard to believe they can hold out once the exponential gets going and everyone knows what they are missing. Right now with only a few hundred thousand rides a week, it’s easy to shrug it off.
Thus I think PoliMath might be onto something here:
PoliMath: I suspect Waymo doesn’t *wantthere to be a policy response to this data b/c it will inevitably end with the left demanding we ban human drivers and there will be a huge backlash that damages Waymo’s business in a serious way.
Waymo is steadily winning, as in expanding its operations. The more it expands, the better its case, the more it will be seen as inevitable. Why pick a premature fight?
Roon: It is pretty telling that when you ride in a Waymo, you cannot give instructions to Gemini to play a song, change your destination, or drive differently. When one of the great gilded tech monopolies of the world does not yet have a cohesive AI picture, what hope has the broader economy?
Eliezer Yudkowsky: AI companies are often so catastrophically stupid that I worry that Gemini might in some way be connected to the actual car. Oh wait, you explicitly want to be able to request that the car drive differently?
I do not want Gemini to be controlling the vehicle or how it drives, but there are other things that would be nice features for integration, and there are other quality of life improvements one could make as well. For now, we keep it clean and simple.
I am deeply skeptical that the discounts are this large, and my AI sanity check confirmed the savings are real but relatively modest. Also consider what ‘comfortable retirement’ means in places that (for example) won’t let you buy an air conditioner. But yeah, if you only have modest savings it seems like a good thing to consider.
YouTube Premium is an ideal product. For $10 a month you get no ads, creators get paid, and the variety of content is phenomenal. Yes, you could use AdBlock to get around it in many cases, and many will do that, but this is what the internet is supposed to look like.
Maxim Lobovsky: Not only is YouTube Premium great, it’s one of the few major ad-supported businesses offering a paid alternative. Paid social media is one of the only plausible solutions to the algorithm-driven polarization/rage-baiting/lowest-common-denominator content death spiral.
The problem is that you can’t then subscribe individually to everything else, because that adds up fast. Give me a unified YouTube Premium style subscription, please.
Yes, the failure to shut down TikTok despite a law making it illegal that was upheld by the Supreme Court 9-0 is rather insane. Trump is flat out refusing to enforce the ban and extending the deadline indefinitely, you can speculate as to why.
Downvotes, in some form, are a vital part of any social platform that has upvotes, both to maintain civility and maintain good incentives. If you can easily express pleasure there needs to also be an easy way to express displeasure. Dan Luu gives one reason, which is that otherwise people will write nasty comments as a substitute. The other reason is that otherwise maximizing for toxoplasma of rage and extreme reactions to get engagement wins and crowds other actions out. If you are going to do rankings, the rankings on LessWrong and also Reddit mostly seem quite good, and those are the only places where somewhat algorithmic feeds seem to do well.
Emmett Shear: The belief that downvotes are “uncivil” was one of the most common delusions I have encountered while working in social media.
Oliver Habryka: Yep, one of the things I always considered most crucial to maintain with LW 2.0. When I was shopping around for forum software alternatives when we started building LW 2.0 this ruled out like 80% of the options on the market.
Cremieux reports he was suspended from Twitter for a day for saying that Tea app had been hacked, which was called ‘posting other people’s private information without their express authorization and permission,’ except he did not do this or link to anyone who did do it (he said ‘you can go download 59.3 GB of user selfies right now’), whereas people who do expose such info often get ignored. He went warpath, citing various laws he asserts Twitter is breaking around the world.
(The link in the screenshot below takes you back to the post itself.)
Lewis: meanwhile post doxxing [someone’s] address was never removed. 2.5M views. reported it and DM’d Nikita, never heard back on either.
Sin: My contribution [which is literally a map containing the location with a giant arrow pointing to it saying it is where this person lives].
I saw this over a week later. Still there.
Elon Musk made a lot of mistakes with Twitter, but also did make some very good changes. One of them is that likes are now private. This damages an outsider’s ability to read and evaluate interactions, but it takes away the threat of the gotcha when someone is caught liking (or even not liking!) the wrong tweet and the general worry about perception, freeing people up to use them in various ways including to acknowledge that you’ve seen something, and to offer private approval.
It’s very freeing. When likes were public, which also means it was public what you didn’t like, I decided the only solution to this was to essentially not use the like button. Which worked, but is a big degrading of usefulness of Twitter.
Redaction: It really is insane how simply Hiding Likes On Twitter meaningfully shifted the overton window of the American political landscape
Samo Burja: I underestimated the impact change at the time. I think I thought preference falsification was much less pervasive than it was.
Meanwhile, in other contexts, it is still very much a thing to talk about who has liked which Instagram posts. This is exactly as insufferable as it sounds.
Every time Nikita tries to make me feel better about Twitter I end up feeling worse.
Nikita Bier (Twitter): The first step to eliminating spam is eliminating the incentive.
So over the last week, I have gone deep down the rabbit hole of X spam:
I am now in 3 WhatsApp groups for financial scams. I have become their friends. I know about their families. I cannot blow my cover yet.
What is the goal exactly? How would befriending them help? We already all know exactly how to identify these scams and roughly how they work. Understanding more details will not help Nikita or anyone else do anything. You think you’re going to do enough real world takedowns and arrests that people are scared to do scams, or something? How about instead we do basic filtering work?
Or, when he posts this:
Or this:
Eli: Twitter should include 3 schizophrenic reply guys and 1 egirl with Premium +
Nikita Bier: We did the math and that’s what retains a user.
He kids, but kid enough times in enough ways with enough detail and you’re not fully kidding. It is very clear that Twitter is doing a lot of the Goodhart’s Law thing, where short term feedback metrics are being chased without much eye on the overall experience. Over time, this goes to quite bad places.
Also, yeah, this is not okay:
Mike Solana: I truly believe blocking is a right, and I would never go after someone for blocking me for any reason. but you should not then be able to unblock, comment on a post of mine, and immediately REBLOCK so I can’t respond. in this case, I deserve at least 24 hours to roast your ass.
There are any number of obviously acceptable solutions to this. I like the 24 hours, where if you do something that you couldn’t do while they are blocked, your reblock is delayed for a day.
Vas: If a simple coffee shop has a bot farm with 100s of phones to amplify their message, please consider what a foreign agency or adversarial operator is running on your favorite social media platform.
Especially today, please consider that the opinions you read, the calls to violence you hear, and the news you digest, are all an operation done to sow hatred in your mind and your soul.
I may never stop finding it funny the extent to which Trump will seek out the one thing we know definitively is going badly, then and choose that to lie and brag about.
Initial market reaction was muted, the dollar only declined 0.3% and gold only rose 0.6%, likely because it was one escalation among many and it might fail, but this is a direct full assault on central bank independence, and central bank independence is a really big deal.
Jonnelle Marte and Myles Miller (Bloomberg): While a president has never removed a Fed governor from office, one can do so for cause. Laws that describe “for cause” generally define the term as encompassing three possibilities: inefficiency; neglect of duty; and malfeasance, meaning wrongdoing, in office.
What was this ‘cause’?
Trump had earlier called for Cook’s resignation after Federal Housing Finance Agency Director Bill Pulte alleged she lied on loan applications for two properties — one in Michigan and one in Georgia — claiming she would use each property as her primary residence to secure more favorable loan terms.
Trump said it was “inconceivable” that Cook was not aware of requirements in two separate mortgage applications taken out in the same year requiring her to maintain each property as her primary residence.
That’s it. There are no additional claims. Only the claim that she claimed one place would be a primary residence, and then claimed a different primary residence.
Pulte, in a statement posted to social media, thanked Trump for removing Cook. “If you commit mortgage fraud in America, we will come after you, no matter who you are,” he wrote.
What about if you are President of the United States and have recently had a nine figure judgment against your ‘Trump’ organization entered against you for lying on mortgage applications? Are we coming for you?
Aaron Fritschner: The mortgage fraud claim against Lisa Cook is false, per documents obtained by Reuters. Bill Pulte’s accusation, the sole pretext Trump used to fire her from the Fed, was that she claimed two homes as primary residence. These docs show she did not.
“The document, dated May 28, 2021, was issued to Cook by her credit union in the weeks before she completed the purchase and shows that she had told the lender that the Atlanta property wouldn’t be her primary residence.”
“documentation reviewed by Reuters for the Atlanta home filed with a court in Georgia’s Fulton County, clearly says the stipulation exists “unless Lender otherwise agrees in writing.” The loan estimate, a document prepared by the credit union, states “Property Use: Vacation Home”
Lisa Cook also didn’t claim a tax credit for primary residence on the second home and declared it as a second home on official federal government documents when she was being considered for a role on the Fed. A real master criminal.
Also her mortgage rate was 3.5%, modestly higher than the going rate at the time.
If you are going to try and fire a Federal Reserve President for cause, something that has not happened (checks notes, by which I mean GPT-5) ever, thus endangering faith in Fed independence and the entire financial system, you might want to follow due process, or at least confirm that your accusation is true? As opposed to demonstrably false?
A lot of people are understandably highly outraged about this, as Adam Tooze partly covered right after the attempted firing. This comes on the heels of firing the head of the Bureau of Labor Statistics because Trump didn’t like the statistics.
Yes, all the various ‘honesty taxes’ our government imposes (also see this highlighted comment) are extremely pernicious and do a lot more damage than people realize. We teach people they have to lie in order to get food stamps, and they (just like LLMs!) learn to generalize this, everything impacts everything, our society is saying lying is okay and lying to the government is mandatory, you can’t isolate that. You don’t get a high trust society that way, although we are remarkably high trust in some ways despite this.
Most of the time, the correct answer is not to enforce the rules as written even if we could do so, instead the correct answer is to remove or heavily modify the rule. Our rules are tuned to the idea they won’t be enforced, so it is likely enforcing them would not go well. Then there are exceptions, such as primary residence mortgage fraud.
Aaron Bergman: I think ethics- and integrity-pilled people need to have a better theory of when it’s cool to lie to *institutions
The “lying to a human” vs “lying to institution” distinction is real and important btw, the bar for the latter is much lower
Oliver Habryka: Yeah, I agree with this. I think lying to institutions is frequently fine, often roughly proportional to how big they are, though there are also other important factors.
I don’t have a great answer to exactly when this all makes it okay to lie to corporations or governments and on forms. My guess is it is roughly ‘do not break the social contract.’ But if this is something where is no longer (or never was) a social contract, and no one would look at you funny if you were doing it in the open, then fine.
If you notice you are very clearly expected to lie (including by omission) or do a fake version of something, that the system is designed that way, then you have little choice, especially if you are also forced to deal with such institutions in order to get vital goods or services.
I try to keep a high threshold for criticism but it does seem like Trump ordered a bunch of people murdered (some might use the term ‘war crime’ but I prefer plain language and also there was no war involved, the ‘war on drugs’ is not a war) on the high seas with absolutely no legal basis for doing so? He ran the plot of Clear And Present Danger straight up in the open? You didn’t know there were drugs involved, and even if you did you can’t go around blowing up boats purely because there were drugs involved?
JD Vance: Killing cartel members who poison our fellow citizens is the highest and best use of our military.
Rand Paul: JD “I don’t give a shit” Vance says killing people he accuses of a crime is the “highest and best use of the military.”
Did he ever read To Kill a Mockingbird?
Did he ever wonder what might happen if the accused were immediately executed without trial or representation??
What a despicable and thoughtless sentiment it is to glorify killing someone without a trial.
I’m sincerely and deeply confused what makes this not straight up murder and have not seen any serious arguments for why it would be anything else, as opposed to ‘yes it is murder and I really like murder for us here, yay this murder.’ It also seems, to the extent such points are relevant in 2025, like a very clear ‘high crime and misdemeanor.’
Apple’s new iPhone 17 Pro Max seems like a substantial improvement over the iPhone 16 Pro Max. You get 50% more ram, at least twice as many camera pixels, better cooling, a substantially better battery, and a new faster chip with GPUs ‘designed for AI workloads.’ I’m going to stick with my Pixel 9 Fold, the only feature on iPhones that is compelling to me at all is avoiding anti-Android discrimination, hell of a business model, but those are nice upgrades.
Apple Vision Pro is making small inroads in specialized workplaces that can exploit spatial computing, including things like exploring potential new kitchens or training airline pilots. It is expensive, but if it is the best tool, it can still be worth it. The rest of us will be waiting until it is lighter, better, faster and cheaper.
When Meta tries to have internal clarity and do research on the dangers of its products, people leak that information to the press, details get taken out of context and they get hauled before Congress.
When Meta responds to this by avoiding clarity, and suppressing the research or ignoring the problem, they get blamed for that too.
I mean, yes, why not simply actually deal with your safety problems is a valid response, but the incentives here are pretty nasty.
The central accusation is that the company’s lawyers intervened to shape research into risks from virtual reality, but I mean it would be insane for Meta not to loop the lawyers in on that research. If we are going to make Meta blameworthy, including legally, for doing the research, then yes they are going to run the research by the lawyers. This is a failure of public policy and public choice.
That doesn’t make the actual problems any less terrible, but it sounds like they are very standard. Kids were bypassing age restrictions, and when interacting with others they would get propositioned. It seems like the accusation is literally ‘Meta let its users interact with each other, and sometimes those users said or did bad things.’
Experts have long warned that virtual reality can endanger children by potentially exposing them to direct, real-time contact with adult predators.
It is remarkable how consistently even the selected examples don’t involve VR features beyond allowing users to talk? I’m not saying you don’t need to have safeguards for this, but it all sounds very similar to the paranoia and statistical illiteracy where we don’t let children participate in physical spaces anymore.
I love this report of a major problem running the other way, to which, I mean, fair:
In a January 2022 post to a private internal message board, a Meta employee flagged the presence of children in “Horizon Worlds,” which was at the time supposed to be used only by adults 18 and over. The employee wrote that an analysis of app reviews indicated many were being driven off the app because of child users.
I’m not saying Meta handled any of this perfectly or even handled it well. But there’s no smoking gun here, and no indication that they aren’t taking reasonable steps.
Matthew served, and Nate served back, so now it’s on.
Nate Silver: Academic journals might be a lost cause but they’d probably be better if you had some non-academic practitioners serving as reviewers. Journalists have their problems too but they have much better bullshit detectors, for instance.
The most important research paper of the past 10 years is the Google transformer paper (“Attention Is All You Need”) and it was written by non-academics and published in an open-access journal.
You ran some cool regression analysis OK great. Make some nice graphics and put it on a Substack. Engaging headline, 1500-2500 well-written words. That’s literally 100x faster than trying to publish in a journal and it’s better peer review anyway.
Matthew Sitman: Very, very occasionally an exceptional generalist intellectual or particularly well-informed journalist might be able to see a problem with a paper that an academic close to the subject doesn’t, but this radically underestimates the uses of expertise/familiarity with a literature.
As someone who’s been an academic and now talks/writes about ideas for non-specialists, a difference is that academics, ideally, know what they don’t know, are aware of questions asked/answered previously, etc; if that can produce tunnel vision, well, they’re trying to dig deep.
Nate Silver: Well, I know a lot about statistical inference, have been doing it for 25 years, have faced a lot of public scrutiny, and in the fields where I also have a lot of domain knowledge, probably half of published papers have obvious fatal flaws that render them unfit for publication.
Maybe I’m a weird outlier, but the peer review process is obviously broken. Maybe it’s better in the fields I *don’tknow well. But I’d be surprised if that’s true.
Aella: People don’t understand how much of a joke the current state of peer review is. It’s extremely bad.
It’s bad enough that at one point I was suggesting to someone “why don’t you just deliberately insert mistakes so they can feel satisfied about finding those and don’t end up fucking with the rest of the paper”
St. Motweb: This is actually a strategy that many academics use.
SolDad: I unironically do this in my papers, sort of. Not inserting new stuff, but leaving fairly-obvious but not super important work undone as “low hanging fruit” for the reviewers to notice and ask for.
This suggests a wager.
Select a field. A neutral party (or an LLM with an agreed upon prompt) selects 10 recent papers that meet some criteria for being Proper Papers In Good Journals and that assert some statistical finding and thus could in theory be flawed.
If Nate Silver can find a fatal flaw in at least 2 of the 10 papers, as evaluated by an agreed neutral judge, then he wins. If not, he loses. This should cover both sides: Two is enough that the system is obviously fatally flawed, and Nate says he can average five.
This is not a statement that the first best solution to peer review involved outsiders like Nate Silver reviewing papers. That seems obviously wrong. It is a claim that the current solution is so utterly terrible that outsider review would be a big improvement.
Indeed, ‘put your ideas out on the internet and let people critique them’ is the actual essence of true peer review, and far superior in practice to the formal version in 2025.
This is the peer! This is the review! That is how all of this works! This is it working!
Rob Bensinger is latest to note that we could do way better on a wide array of problems if we could improve discourse norms, and this could be a high impact play. That doesn’t mean we know how to pull it off. As he notes, prediction markets have been somewhat helpful, but seem unlikely to be the full game changer we need. Also as he notes, this would be great to pull off but there’s an attractor that causes this to look more relatively doable than it is, which can trick people into focusing on it more than they should relative to other interventions.
Kelsey Piper provides her overview of the findings that Giving People Money on a monthly basis in America does not seem to improve most outcomes, including child outcomes. They’re not ‘blowing’ the money on vices, but people give back a lot of it by working less, and while they tell stories about how great things are, most of the statistics don’t improve.
She then points us to a conservative critique of her analysis by Charles Lehman. I agree with Charles that the criticism ‘maybe the gains don’t show up in the measurements’ is rather silly, unless you can point to a specific policy aim that isn’t being measured, and explain why you have hope for that one despite the others not showing up.
I also appreciated Charles saying that for American purposes, a study of a social intervention in Africa should be treated similarly to when biologists say something ‘works in vitro,’ as conditions are so radically different. The synthesis would be that ‘give people money’ is highly useful at improving various outcomes when those people have so little money that they risk starvation, but not that far beyond that, and existing interventions here already take us above the threshold.
We definitely need to reconcile our model of how this works with not only the null results in these studies, but also the other null results from many other programs.
One reason to be suspicious of ‘policy mostly can’t help’ is that if you propose ‘policy mostly can’t hurt’ or even ‘getting rid of existing policies and enforcement mostly can’t hurt’ then most people will disagree with you. So at minimum, you can help by not hurting, and you should find the extreme asymmetry suspicious.
I do have one policy objective that I am confident this does help with if it is reliable and sustained, which is fertility. I’m sticking by the estimate that for every ~$270k in value (which need not be cash) you transfer to parents, you get one additional birth. This is one area where anticipation of money, or rather anticipation of having the necessary resources of all types, definitely changes behavior.
Lennox: But when I looked at what the EAs were actually doing, and the methods they were using to evaluate charities, it quickly became clear that this was not going to work. One look at a GiveWell spreadsheet filled my heart with dread. They were creating insanely detailed cost effectiveness estimates of different interventions, using probabilistic models that tried to account for every possible empirical and philosophical assumption you could think of.
It would be great to analyse policies that fundamentally transform the economy at this level of detail, but there were a couple of problems. First, it’s impossible to create a useful model at that level of detail for transformative economic policies. Second, even if it were possible, there’s no way I could do it.
Fine. Lesson learned.
Except, of course, lesson not learned, because he didn’t then think ‘oh that is exactly why the socialist ideas I am advocating for won’t work.’ So he continues, and asks his sociological experts who love socialism. The same result happens again:
This was… disappointing to say the least. Here was a group of serious academics who had spent decades trying to make a rigorous case for socialism, and this is what they ended up concluding? That we don’t have the social technology to make it work, but maybe one day we will get there.
He then does the ‘finds smoking causes cancer, quits reading’ move, saying this means analytical Marxists had undermined themselves so maybe look at critical theory. Because, of course:
I’d assumed that if you want to solve a systemic problem like global poverty, you need to understand the root cause, and the root cause of poverty was, of course, capitalism.
…
However, understanding the root cause of something doesn’t automatically help you solve it.
The main mistake, of course, is that the root cause was not capitalism but instead the human condition. This had not yet entered Lennox’s hypothesis space. The other mistake was, yes, knowing the root cause of something does not always help solve it.
The evidence continued to mount.
Throughout undergrad, I would read sociological theorists and often find their arguments vague, opaque, and at times just poorly argued. Then I would read work by EAs and find it crystal clear, carefully argued, and generally well calibrated to the evidence.
…
The final nail in the coffin came while reading Scott Alexander’s essayMeditations on Moloch.
..
Looking back, I could have saved myself a lot of time. These fundamental problems with the project were probably obvious to many in the EA community and they would have told me the project was unlikely to be useful, if I’d had the courage to ask. But I avoided getting their feedback, partly because I figured they were ideologically blinded and would just dismiss anything critical of their movement.
That last line really is wonderful. Socialist refuses to get feedback from target audience because they are worried audience is ideologically blinded. Love it.
After that he is then able to do some self-reflection, also fun but less fun. Then in conclusion he comes around and notes that if you accept that the world is a swirling mess of misaligned incentives and coordination problems, then this completely undermines the Marxist political project.
That is indeed how the world works, so yes. Thank you. Well done, sir. Also, well done, sir, at the end:
Anyway, a couple of years after this happened I fell in love, and it was everything the poets and songwriters said it would be. So I guess the moral of the story is: if you find yourself tempted to construct elaborate ideological arguments in a vain attempt to make yourself feel smart and important, consider falling in love instead.
Push to revive nuclear energy relies on deregulation; experts say strategy is misplaced.
In May, President Donald Trump signed four executive orders to facilitate the construction of nuclear reactors and the development of nuclear energy technology; the orders aim to cut red tape, ease approval processes, and reshape the role of the main regulatory agency, the Nuclear Regulatory Commission, or NRC. These moves, the administration said, were part of an effort to achieve American independence from foreign power providers by way of a “nuclear energy renaissance.”
Self-reliance isn’t the only factor motivating nuclear power proponents outside of the administration: Following a decades-long trend away from nuclear energy, in part due to safety concerns and high costs, the technology has emerged as a potential option to try to mitigate climate change. Through nuclear fission, in which atoms are split to release energy, reactors don’t emit any greenhouse gases.
The Trump administration wants to quadruple the nuclear sector’s domestic energy production, with the goal of producing 400 gigawatts by 2050. To help achieve that goal, scientific institutions like the Idaho National Laboratory, a leading research institute in nuclear energy, are pushing forward innovations such as more efficient types of fuel. Companies are also investing millions of dollars to develop their own nuclear reactor designs, a move from industry that was previously unheard of in the nuclear sector. For example, Westinghouse, a Pennsylvania-based nuclear power company, plans to build 10 new large reactors to help achieve the 2050 goal.
However, the road to renaissance is filled with familiar obstacles. Nuclear energy infrastructure is “too expensive to build, and it takes too long to build,” said Allison Macfarlane, a science and technology policy expert at the University of British Columbia who used to chair the NRC from 2012 to 2014.
And experts are divided on whether new nuclear technologies, such as small versions of reactors, are ready for primetime. The nuclear energy field is now “in a hype bubble that is driving unrealistic expectations,” said Edwin Lyman, the director of nuclear power safety at the Union of Concerned Scientists, a nonprofit science advocacy organization that has long acted as a nuclear safety watchdog.
Meanwhile, the Trump administration is trying to advance nuclear energy by weakening the NRC, Lyman said. “The message is that it’s regulation that has been the obstacle to deploying nuclear power, and if we just get rid of all this red tape, then the industry is going to thrive,” he added. “I think that’s really misplaced.”
Although streamlining the approval process might accelerate development, the true problem lies in the high costs of nuclear, which would need to be significantly cheaper to compete with other sources of energy such as natural gas, said Koroush Shirvan, a nuclear science researcher at the Massachusetts Institute of Technology. “Even the license-ready reactors are still not economical,” he said. If the newer reactor technologies do pan out, without government support and subsidies, Shirvan said, it is difficult to imagine them “coming online before 2035.”
It’s déjá vu all over again
Rumblings of a nuclear renaissance give experts a sense of déjà vu. The first resurgence in interest was around 2005, when many thought that nuclear energy could mitigate climate change and be an energy alternative to dwindling supply and rising prices of fossil fuels. But that enthusiasm slowed mainly after the Fukushima accident in 2011, in which a tsunami-triggered power outage—along with multiple safety failures—led to a nuclear meltdown at a facility in Japan. “So, the first nuclear renaissance fizzled out,” said Lyman.
Globally, the proportion of electricity provided by nuclear energy has been dwindling. Although there has been an increase in generation, nuclear energy has contributed less to the share of global electricity demand, dropping to 9 percent in 2024 from a peak of about 17 percent in 2001. In the US, 94 reactors generate about a fifth of the nation’s electricity, a proportion that has held steady since 1990s. But only two of those reactors have come online in the last nearly 30 years.
This renewed push is “a second bite at the apple, and we’ll have to see but it does seem to have a lot more of a headwind now,” said Lyman.
Much of that movement comes from the private sector, said Todd Allen, a nuclear engineer at the University of Michigan. In the last couple of decades, dozens of nuclear energy companies have emerged, including TerraPower, co-founded by Bill Gates. “It feels more like normal capitalism than we ever had in nuclear,” Allen said. Those companies are working on developing the large reactors that have been the backbone of nuclear energy for decades, as well as newer technologies that can bolster the field.
Proponents say small modular reactors, or SMRs, and microreactors, which generate less than 300 megawatts and 20 megawatts, respectively, could offer safer, cheaper, and more flexible energy compared to their more traditional counterparts. (Large reactors have, on average, 900 megawatts of capacity.) One 2022 study found that modularization can reduce construction time by up to 60 percent.
These designs have taken the spotlight: In 2024, a report estimated that the SMR market would reach $295 billion by 2043. In June, Energy Secretary Chris Wright told Congress that DOE will have at least three SMRs running by July of next year. And in July of this year, the Nuclear Energy Agency launched a dashboard to track SMR technologies around the world, which identified 74 SMR designs at different stages around the world. The first commercial SMR in North America is currently being constructed in Canada, with plans to be operational by 2030.
But whether SMRs and microreactors are actually safer and more cost-effective remains to be determined. A 2022 study found that SMRs would likely produce more leakage and nuclear waste than conventional reactors. Studying them, though, is difficult since so few are currently operational.
In part, that may be because of cost. Multiple analyses have concluded that, because of rising construction and operating costs, SMRs might not be financially viable enough to compete for the world’s energy markets, including in developing countries that lack affordable access to electricity.
And recent ventures have hit road bumps: For example, NuScale, the only SMR developer with a design approved by the NRC, had to shut down its operations in November 2023 due to increasingly high costs (though another uprated SMR design was approved earlier this year).
“Nothing is really commercialized yet,” said Macfarlane. Most of the tech companies haven’t figured out expenses, supply chains, the kind of waste they are going to produce or security at their reactors, she added.
Fuel supply is also a barrier since most plants use uranium enriched at low rates, but SMRs and microreactors use uranium enriched at higher levels, which is typically sourced from Russia and not commercially available in the US. So scientists at the Idaho National Laboratory are working to recover enriched uranium from existing reactors and developed new, more cost-effective fuels, said Jess Gehin, the associate laboratory director for the Nuclear Science & Technology Directorate at the INL. They are also using artificial intelligence and modeling simulation tools and capabilities to optimize nuclear energy systems, he added: “We got to reach 400 gigawatts, we need to accelerate all of this.”
Companies are determined to face and surpass these barriers. Some have begun pouring concrete, such as one nuclear company called Kairos Power that began building a demo of their SMR design in Tennessee; the plant is projected to be fully operational by 2027. “I would make the case that we’re moving faster than many in the field, if not the fastest,” Mike Laufer, the company’s CEO and co-founder, told Reuters last year.
Some experts think achieving nuclear expansion can be done—and revel in the progress so far: “I would have never thought we’d be in this position where we’re working so hard to expand nuclear, because for most of my career, it wasn’t that way,” said Gehin. “And I would say each month that goes by exceeds my expectations on the next bigger things that are coming.”
Doing more with less?
Although the Trump administration aims to accelerate nuclear energy through executive orders, in practice, it has not allocated new funding yet, said Matt Bowen, an expert on nuclear energy, waste, and nonproliferation at Columbia University’s Center on Global Energy Policy. In fact, the initial White House budget proposed cutting $4.7 billion from the Department of Energy, including $408 million from the Office of Nuclear Energy allocated for nuclear research in the 2026 fiscal year.
“The administration was proposing cuts to Office of Nuclear Energy and DOE more broadly, and DOGE is pushing staff out,” said Bowen. “How do you do more with less? Less staff, less money.”
The Trump administration places the blame for the nuclear sector’s stagnation on the NRC, which oversees licensing and recertification processes that cost the industry millions of dollars each year in compliance. In his executive orders, Trump called for a major reorganization of the NRC. Some of the proposed changes, like streamlining the approval process (which can take years for new plants), may be welcomed because “for a long time, they were very, very, very slow,” said Charles Forsberg, a nuclear chemical engineer at MIT. But there are worries that the executive orders could do more than cut red tape.
“Every word in those orders is of concern, because the thrust of those orders is to essentially strip the Nuclear Regulatory Commission of its independence from the executive branch, essentially nullifying the original purpose,” said Lyman.
Some experts fear that with these new constraints, NRC staff will have less time and fewer resources to do their jobs, which could impact power plant safety in the future. Bowen said: “This notion that the problem for nuclear energy is regulation, and so all we need to do is deregulate, is both wrong and also really problematic.”
The next few decades will tell whether nuclear, especially SMRs, can overcome economic and technical challenges to safely contribute to decarbonization efforts. Some, like Gehin, are optimistic. “I think we’re going to accelerate,” he said. “We certainly can achieve a dramatic deployment if we put our mindset to it.”
But making nuclear financially competitive will take serious commitment from the government and the dozens of companies, with many still skeptical, Shirvan said. “I am quite, I would say, on the pessimistic scale when it comes to the future of nuclear energy in the US.”
NASA’s interim administrator, Sean Duffy, was fired up on Wednesday when he joined a teleconference to talk about new scientific findings that concerned the potential for life to have once existed on Mars.
“This is exciting news,” said Duffy about an arrow-shaped rock on Mars found by NASA’s Perseverance rover. The rock contained chemical signatures and structures that could have been formed by ancient microbial life. The findings were intriguing, but not conclusive. Further study of the rocks in an advanced lab on Earth might prove more definitive.
Duffy was ready, he said, to discuss the scientific results along with NASA experts on the call with reporters. However, the very first question—and for any space reporter, the obvious one—concerned NASA’s on-again, off-again plan to return rocks from the surface of Mars for study on Earth. This mission, called Mars Sample Return, has been on hold for nearly two years after an independent analysis found that NASA’s bloated plan would cost at least $8 billion to $11 billion. President Trump has sought to cancel it outright.
Duffy faces the space press
“What’s the latest on NASA’s plans to retrieve the samples from Perseverance?” asked Marcia Dunn, a reporter with the Associated Press, about small vials of rocks collected by the NASA rover on Mars.
“So listen, we’re looking at how we get this sample back, or other samples back,” Duffy replied. “What we’re going to do is look at our budget, so we look at our timing, and you know, how do we spend money better? And you know, what technology do we have to get samples back more quickly? And so that’s a current analysis that’s happening right now.”
A couple of questions later, Ken Chang, a science reporter with The New York Times, asked Duffy why President Trump’s budget request called for the cancellation of Mars Sample Return and whether that was still the president’s intent.
“I want to be really clear,” Duffy replied. “This is a 30-year process that NASA has undertaken. President Trump didn’t say, ‘Hey, let’s forget about Mars.’ No, we’re continuing our exploration. And by the way, we’ve been very clear under this president that we don’t want to just bring samples back from Mars. We want to send our boots to the Moon and to Mars, and that is the work that we’re doing. Amit (Kshatriya, the new associate administrator of NASA) even said maybe we’ll send our equipment to test this sample to Mars itself. All options are on the table.”
Specifically, the child died of subacute sclerosing panencephalitis (SSPE), a rare but always fatal complication that strikes years after an initial measles infection. The health department’s announcement offered few details about the child, including the child’s age, but said that the child had contracted the virus before they were old enough to be vaccinated against measles. The first of two recommended doses of measles, mumps, and rubella (MMR) vaccine is given between 12 and 15 months.
“This case is a painful reminder of how dangerous measles can be, especially for our most vulnerable community members,” Muntu Davis, a Los Angeles County health officer, said in a statement. “Infants too young to be vaccinated rely on all of us to help protect them through community immunity. Vaccination is not just about protecting yourself—it’s about protecting your family, your neighbors, and especially children who are too young to be vaccinated.”
SSPE is caused by a persistent measles infection in the central nervous system. Children infected with the virus may go through the standard disease progression—flu-like symptoms, high fever, the telltale rash—and then appear to fully recover. But, for a small few, the virus remains, and SSPE emerges years later, often seven to 10 years after the initial infection.
The Los Angeles health department noted that SSPE generally affects about 1 in 10,000 people with measles, but the risk may be much higher—about 1 in 600—for those who get measles as infants, such as the child who recently died.
With widespread vaccination, which led to measles being declared eliminated from the US in 2000, SSPE has virtually disappeared in the US. However, with vaccination rates slipping and anti-vaccine misinformation and views gripping the country, health experts fear seeing more of these devastating cases. Already, the US measles case count for the year is at a 33-year high, and two other children, as well as an adult, died from the acute infection this year.
Paragon, responding to the committee’s findings, accused Italian authorities of refusing to conduct a thorough technical verification—an assessment it argued could have resolved the issue.
Apart from focusing on investment, the Atlantic Council notes that the global spyware market is “growing and evolving,” with its dataset expanded to include four new vendors, seven new resellers or brokers, 10 new suppliers, and 55 new individuals linked to the industry.
Newly identified vendors include Israel’s Bindecy and Italy’s SIO. Among the resellers are front companies connected to NSO products, such as Panama’s KBH and Mexico’s Comercializadora de Soluciones Integrales Mecale, as highlighted by the Mexican government. New suppliers named include the UK’s Coretech Security and UAE’s ZeroZenX.
The report highlights the central role that these resellers and brokers play, stating that it is “a notably under-researched set of actors.” According to the report, “These entities act as intermediaries, obscuring the connections between vendors, suppliers, and buyers. Oftentimes, intermediaries connect vendors to new regional markets.”
“This creates an expanded and opaque spyware supply chain, which makes corporate structures, jurisdictional arbitrage, and ultimately accountability measures a challenge to disentangle,” Sarah Graham, who coauthored the report, tells WIRED.
“Despite this, resellers and brokers are not a current feature of policy responses,” she says.
The study reveals the addition of three new countries linked to spyware activity—Japan, Malaysia, and Panama. Japan in particular is a signatory to international efforts to curb spyware abuse, including the Joint Statement on Efforts to Counter the Proliferation and Misuse of Commercial Spyware and the Pall Mall Process Code of Practice for States.
“The discovery of entities operating in new jurisdictions, like Japan, highlights potential conflicts of interest between international commitments and market dynamics,” Graham says.
Despite efforts by the Biden administration to constrain the spyware market through its executive order, trade and visa restrictions, and sanctions, the industry has continued to operate largely without restraint.
Wyden said his office’s investigation into the Ascension breach found that the ransomware attackers’ initial entry into the health giant’s network was the infection of a contractor’s laptop after using Microsoft Edge to search Microsoft’s Bing site. The attackers were then able to expand their hold by attacking Ascension’s Active Directory and abusing its privileged access to push malware to thousands of other machines inside the network. The means for doing so, Wyden said: Kerberoasting.
“Microsoft has become like an arsonist”
“Microsoft’s continued support for the ancient, insecure RC4 encryption technology needlessly exposes its customers to ransomware and other cyber threats by enabling hackers that have gained access to any computer on a corporate network to crack the passwords of privileged accounts used by administrators,” Wyden wrote. “According to Microsoft, this threat can be mitigated by setting long passwords that are at least 14 characters long, but Microsoft’s software does not require such a password length for privileged accounts.”
Additionally, Green noted, the continuing speed of GPUs means that even when passwords appear to be strong, they can still fall to offline cracking attacks. That’s because the security cryptographic hashes created by default RC4/Kerberos use no cryptographic salt and a single iteration of the MD4 algorithm. The combination means an offline cracking attack can make billions of guesses per second, a thousandfold advantage over the same password hashed by non-Kerberos authentication methods.
Referring to the Active Directory default, Green wrote:
It’s actually a terrible design that should have been done away with decades ago. We should not build systems where any random attacker who compromises a single employee laptop can ask for a message encrypted under a critical password! This basically invites offline cracking attacks, which do not need even to be executed on the compromised laptop—they can be exported out of the network to another location and performed using GPUs and other hardware.
More than 11 months after announcing its plans to deprecate RC4/Kerberos, the company has provided no timeline for doing so. What’s more, Wyden said, the announcement was made in a “highly technical blog post on an obscure area of the company’s website on a Friday afternoon.” Wyden also criticized Microsoft for declining to “explicitly warn its customers that they are vulnerable to the Kerberoasting hacking technique unless they change the default settings chosen by Microsoft.”
Google is one of the most ardent proponents of generative AI technology, as evidenced by the recent launch of the Pixel 10 series. The phones were announced with more than 20 new AI experiences, according to Google. However, one of them is already being pulled from the company’s phones. If you go looking for your Daily Hub, you may be disappointed. Not that disappointed, though, as it has been pulled because it didn’t do very much.
Many of Google’s new AI features only make themselves known in specific circumstances, for example when Magic Cue finds an opportunity to suggest an address or calendar appointment based on your screen context. The Daily Hub, on the other hand, asserted itself multiple times throughout the day. It appeared at the top of the Google Discover feed, as well as in the At a Glance widget right at the top of the home screen.
Just a few weeks after release, Google has pulled the Daily Hub preview from Pixel 10 devices. You will no longer see it in Google Discover nor in the home screen widget. After being spotted by 9to5Google, the company has issued a statement explaining its plans.
“To ensure the best possible experience on Pixel, we’re temporarily pausing the public preview of Daily Hub for users. Our teams are actively working to enhance its performance and refine the personalized experience. We look forward to reintroducing an improved Daily Hub when it’s ready,” a Google spokesperson said.
Microsoft’s Office 365 suite will soon incorporate AI models from Anthropic alongside existing OpenAI technology, The Information reported, ending years of exclusive reliance on OpenAI for generative AI features across Word, Excel, PowerPoint, and Outlook.
The shift reportedly follows internal testing that revealed Anthropic’s Claude Sonnet 4 model excels at specific Office tasks where OpenAI’s models fall short, particularly in visual design and spreadsheet automation, according to sources familiar with the project cited by The Information, who stressed the move is not a negotiating tactic.
Anthropic did not immediately respond to Ars Technica’s request for comment.
In an unusual arrangement showing the tangled alliances of the AI industry, Microsoft will reportedly purchase access to Anthropic’s models through Amazon Web Services—both a cloud computing rival and one of Anthropic’s major investors. The integration is expected to be announced within weeks, with subscription pricing for Office’s AI tools remaining unchanged, the report says.
Microsoft maintains that its OpenAI relationship remains intact. “As we’ve said, OpenAI will continue to be our partner on frontier models and we remain committed to our long-term partnership,” a Microsoft spokesperson told Reuters following the report. The tech giant has poured over $13 billion into OpenAI to date and is currently negotiating terms for continued access to OpenAI’s models amid ongoing negotiations about their partnership terms.
Stretching back to 2019, Microsoft’s tight partnership with OpenAI until recently gave the tech giant a head start in AI assistants based on language models, allowing for a rapid (though bumpy) deployment of OpenAI-technology-based features in Bing search and the rollout of Copilot assistants throughout its software ecosystem. It’s worth noting, however, that a recent report from the UK government found no clear productivity boost from using Copilot AI in daily work tasks among study participants.
Some Republicans are still angry over the deplatforming of Trump by tech executives once known for their progressive politics. They had been joined by a “vocal and growing group of conservatives who are fundamentally suspicious of the benefits of technological innovation,” Thierer said.
With MAGA skeptics on one side and Big Tech allies of the president on the other, a “battle for the soul of the conservative movement” is under way.
Popular resentment is now a threat to Trump’s Republican Party, warn some of its biggest supporters—especially if AI begins displacing jobs as many of its exponents suggest.
“You can displace farm workers—what are they going to do about it? You can displace factory workers—they will just kill themselves with drugs and fast food,” Tucker Carlson, one of the MAGA movement’s most prominent media figures, told a tech conference on Monday.
“If you do that to lawyers and non-profit sector employees, you will get a revolution.”
It made Trump’s embrace of Silicon Valley bosses a “significant risk” for his administration ahead of next year’s midterm elections, a leading Republican strategist said.
“It’s a real double-edged sword—the administration is forced to embrace [AI] because if the US is not the leader in AI, China will be,” the strategist said, echoing the kind of argument made by Sacks and fellow Trump adviser Michael Kratsios for their AI policy platform.
“But you could see unemployment spiking over the next year,” the strategist said.
Other MAGA supporters are urging Trump to tone down at least his public cheerleading for an AI sector so many of them consider a threat.
“The pressure that is being placed on conservatives to fall in line… is a recipe for discontent,” said Toscano.
By courting AI bosses, the Republican Party, which claims to represent the pro-family movement, religious communities, and American workers, appeared to be embracing those who are antithetical to all of those groups, he warned.
“The current view of things suggests that the most important members of the party are those that are from Silicon Valley,” Toscano said.
Additional reporting by Cristina Criddle in San Francisco.
The purported main purpose of school, and even of childhood, is educating children.
Many people are actively opposed to this idea.
Either they have other priorities that matter more, or sometimes they outright think that your child learning things is bad and you should feel bad for wanting that.
Some even say it openly. They actively and openly want to stop your child from learning things, and want to put your child into settings where they will not learn things. And they say that this is good.
Or they simply assert that the primary point of education is as a positional good, where what matters is your relative standing. And then they pretend this doesn’t imply they both should and are going around preventing children from learning.
In other places, we simply epicly fail at education and don’t seem to care. Or education ‘experts’ claim that things that obviously work don’t work, or things that obviously don’t work, do work.
Consider this section some combination of peek into this alternative universe of thought and the fun of multiple meta levels of shooting fish in a barrel?
What? Was it supposed to do that? Would you want it to?
Very obviously the only way to close academic gaps fully is to go all handicapper general and ban bright kids from getting educations. Thus, The War on Education.
Freddie starts off saying we can’t admit some kids aren’t smart, and some kids will naturally do better at school than others, to which I say you just admitted it, and I’m happy to admit it, and everyone I talk to is willing to admit it, so who is this mysterious we. It is, presumably, a Certain Type of Guy who is an ‘education expert’ of some kind and presumably has a maximum of one child who has gone to school.
Freddie DeBoer: Our educational debates are largely useless because most people engaged in those debates assume out of hand that, absent unusual circumstances like severe neglect or abuse or the presence of developmental or cognitive disabilities, any student can be taught to any level of academic success, and any failure to induce academic success in students is the result of some sort of unfortunate error.
Well, it depends.
If you mean ‘those debates’ as in those between those ‘education experts’? Then perhaps yes, they make these types of absurdly stupid assumptions. If you mean ‘debates among actual regular humans,’ then no. Obviously not. One would question whether Freddie has met such people.
Education can raise the absolute performance of most students modestly, but it almost never meaningfully reshuffles the relative distribution of ability and achievement.
Um, again, what exactly were we trying to do? Educate the children? Or make sure we don’t educate the children? Half and half?
I mean, I guess Freddie then does a job repeatedly exposing ‘the contradictions’ as it were in the entire equality project, but the barrel already has a lot of bullet holes, the water is leaking and the fish are dead.
So we get more fun lines like this:
We have spent an immense amount of effort, manpower, time, and treasure on forcing students to meet procrustean academic standards, despite the fact that we have overwhelming evidence that their relative performance is largely fixed.
Yes, obviously, also yes the extra money is mostly being wasted but even if it wasn’t the whole point was presumably to (drum roll) educate the children.
Why in the world would we spend tons of resources and time on relative education, which by definition is zero sum and a red queen’s race? That doesn’t make sense. There’s a fixed amount of relative education.
At the end of this essay, I will argue that education is important, does matter, and is worth funding – but that what’s now assumed to be its primary purpose, moving students around in quantitative educational metrics, is actually what education does worst.
Who thought that was its primary purpose, either the metrics or the thing itself?
Meanwhile, the reason this was brought to my attention is that his ‘absolute learning has value’ t-shirt is raising questions supposedly answered by the shirt:
What This Essay Does Not Argue:
That absolute learning (that is, learning as measured against a standard or benchmark or criterion) has no value; rather, relative learning is practically and morally dominant in these discussions because only relative learning (sometimes discussed in terms of educational mobility) can better one’s economic fortunes, and it is that potential that underlies our entire modern educational debates and the reason for obsession with achievement gaps.
The next section is ‘I Assure You, You Do Care About Relative Learning.’
I assure him that I don’t.
His first argument is that relative learning indicates absolute learning. That is true but saying this means therefore you care about relative learning (checkmate, liberals?) is not how logic or words work. Caring about the territory does not mean you care about the (not very accurate) map.
Second, while I am happy to concede that absolute learning happens all the time, this should not be mistaken for saying that absolute learning is easily achieved, reliable, or consistent.
I don’t understand why this is supposed to be a relevant argument here. It seems like he’s saying I care about [X], but actually [X] is hard, so instead I care about [Y]?
Most importantly, though, is a simple reality: the consequences of education are derived from relative performance, not absolute.
…
In the vast majority of scenarios where education is relevant, applicants of whatever type are being evaluated relative to peers.
There’s saying the quiet part out loud, and then there’s this.
The purpose of education is… to do well on applications?
He concedes that one might learn to drive and then use this skill to usefully operate a moving vehicle, but says this type of education is rare – that most education has no actual use whatsoever, other than as a positional good to grab a larger share of stuff.
Then he goes through that schools are ‘not guilty’ because improving educational outcomes is impossible anyway. Transferring does nothing. Charter schools don’t work. Interventions don’t work (literally “Nothing “Works””), full null hypothesis.
All right, so now we have a section ‘So What Should We Do?’
Very obviously, if you actually believed all that, you would want to dramatically reduce spending, both in money and in the time of children, on school, since school is almost entirely about relative position. Spending more on school, trying to achieve more or improve performance, in this model, is a defection against everyone else. So we should ban attempts to educate children, beyond some basic skills, and focus on practical stuff like learning to drive. Completely reorient childhood.
So having pre registered that, let’s see what he recommends.
Improve air quality. Okay, sure, that is one of the somethings that work, although again I don’t understand why he thinks improving performance is good.
Lower our educational standards. Don’t make kids learn (for example) abstract math. Yes, that makes perfect sense for Freddie, if the learning is useless, you shouldn’t require it. Again, if he is right then we should go farther, and ban such learning. Why are we letting kids engage in a zero sum competition?
Soft tracking. Again, good idea, not sure what it has to do with the post.
Invest in a robust safety net. Maybe? That’s a different department.
Then he tries to pivot back to ‘actually education matters.’
Education creates the conditions for children and young adults to discover ideas, literature, science, and art that might otherwise remain inaccessible.
It provides the structured time and social environment where curiosity can blossom, where students can learn how to think about problems that don’t have easy answers, and where they can build lasting relationships with peers and mentors.
The point of school, then, is not to guarantee that every child climbs into the top decile of performance but to offer each student the chance to cultivate knowledge, resilience, and imagination in ways that enrich their lives.
So absolute learning of something does matter after all, then. I mean, this description does not match what I know about actual schools, nor would I design anything like a current school if those were the goals. And he doesn’t seem interested in a redesign or asking how to maximize the things that he thinks matter. But hey.
Meanwhile, here’s the top comment, so yes things do get pretty insane:
James K: This is what I pay you for, Freddie, thank you for being so clear-headed about this topic.
I’ve been teaching for 16 years now and it boggles my mind that the band teacher can literally get on stage and say “We have the beginner, intermediate, and advanced band for you” and of course the baseball team can be divided into Varsity and JV, but I am not allowed to say that some kids are not smart enough to handle my AP classes because this means I don’t BELIEVE IN THEM or am supporting TRACKING (always said in the tones people reserve for the words ‘eugenics’ or ‘segregation’).
I mean, yes. It means you support tracking because tracking is good. It means you don’t believe in them in the sense that you don’t believe in things that aren’t real.
So no, in that sense Freddie isn’t arguing with a strawman. Which means that the entire system of education is being run by people who are at war with education.
A Tennessee teen is suing his school for ‘compensatory education’ after graduating with a 3.4 GPA, but being unable to read, or even spell his own name, and the school system has the audacity to defend against that lawsuit.
But the school took no action, the suit says, other than giving him 24 hours to complete his assignments.
But even this “solution” was a problem. Because when William was at home with his schoolwork, he relied on AI programs like ChatGPT and Grammarly to complete his assignments for him, according to the judge who ruled on his suit last week. As a result, William continued to achieve high marks on his classwork throughout his entire four years of high school, even though teachers knew he was illiterate.
If you can’t read, using ChatGPT is kind of crazy – you’re presumably scanning or copy pasting in text you don’t understand, then copying out text you don’t understand and hoping for the best.
Well, they’re not great obviously, to the extent you can trust the scores to map to Reality, but they are still above the start of the graph in 1998, and we’re talking about a seven point difference. That’s less than a fifth of a standard deviation. This is nothing, if anything this shows that Covid didn’t change things much?
The comments section of Scott’s post is full of despair about classroom conditions getting worse, shifts to teaching strategies that don’t work (see the Mississippi reforms but in reverse), discipline collapsing and teachers having no tools if kids don’t play along, many teachers quitting, chronic absenteeism happening and being accepted and tolerated, and many families not prioritizing education, on top of the continuing trends involving smartphones. That’s on top of the obvious ‘Covid took away a lot of schooling’ concern that Scott starts with.
What seems more meaningful than the overall smaller drop is the widening gap between low and high performers, another trend predating Covid. Scott has several graphs showing this, and I am convinced this is real, with a variety of causes. If you are properly equipped and motivated, you can avoid the pitfalls described above, and you have access to the entire web and world and a lot of new resources, now even including AI. Whereas when the bottom falls out, the bottom falls out.
Meanwhile in 12th grade, nearly half scored below ‘the basic level,’ which involves things like ‘using percentages to solve real-world problems,’ and reading scores hit a new low. What we are doing, including adding funding, is clearly not working. Or rather, it is working hard, only not at the goal of children learning academic skills.
The War on Algebra in particular is still perhaps the craziest thing I’ve ever seen, actively preventing children from learning math out of spite. In the sense that it both very clearly purely destructive and evil, and also horribly unpopular across the board, and also high salience to a lot of voters. Yes, I do know what their arguments are for doing it, and they very much do not make it any better.
Ro Khanna: It is absurd that Palo Alto School district just voted to remove honors biology for all students & already removed honors English. They call it de-laning. I call it an assault on excellence. I took many honors classes at Council Rock High in PA.
Autumn Looijen: I ran the campaign to bring algebra back to SF’s middle schools.
It was the most popular thing I ever worked on. Voters don’t want to take away opportunities for kids who can’t afford private school.
If anyone wants to put this on the ballot in Palo Alto, happy to advise.
Maud Maron: We are trying to do a version of this in NYC! Would love to have you speak to NYC parents who want to have algebra & geometry options in Middle School.
Meanwhile in San Francisco, the war rages on. It seems the city has not yet been retaken by sanity on all fronts yet, although there are some promising signs. All of this seems like it has to be beyond unpopular, in a ‘cause families to move out’ way, yet here we were again not too long ago (it got better, for now):
Garry Tan: San Francisco schools is trying its absolute hardest to make sure all middle income families who could move out of the city do so right away.
“Grading for Equity” is going to be a real disaster and I guess this is a boon for SF private schools and Burlingame housing prices.
For education bureaucrats who ruin our public schools with the most unfair and anti-merit polices: BUSINESS IS BOOMING.
Someone needs to investigate the Schools of Education that spawn these policies because it is a real danger to public schools everywhere.
Basically this scam is Idiocracy in real life.
Mike Solana: the san francisco board of education must immediately fire the superintendent. if they do not, they must all be removed from power.
I don’t get how this falls under ‘you can just do things’ but it seems it did, at least until people sounded the alarm?
John Trasvina (The Voice of SF): Without seeking approval of the San Francisco Board of Education, Superintendent of Schools Maria Su plans to unveil a new Grading for Equity plan on Tuesday that will go into effect this fall at 14 high schools and cover over 10,000 students.
The school district is already negotiating with an outside consultant to train teachers in August in a system that awards a passing C grade to as low as a score of 41 on a 100-point exam.
Were it not for an intrepid school board member, the drastic change in grading with implications for college admissions and career readiness would have gone unnoticed and unexplained. It is buried in a three-word phrase on the last page of a PowerPoint presentation embedded in the school board meeting’s 25-page agenda.
…
Grading for Equity eliminates homework or weekly tests from being counted in a student’s final semester grade. All that matters is how the student scores on a final examination, which can be taken multiple times.
Under the San Leandro Unified School District’s grading for equity system touted by the San Francisco Unified School District and its consultant, a student with a score as low as 80 can attain an A and as low as 21 can pass with a D.
– eliminate homework and weekly tests from counting toward semester grade
– allow students to take the final exam multiple times
– convert all B grades into As, and all Fs into Cs
It’s hard to see the difference between this policy and what you’d get if a bunch of 10yos locked the teachers in a closet and rewrote the rules.
Karen Vaites: More media attention here, please! 🚩🚩🚩
Jared Walczak: The sad irony is that Grading for Equity is virtually the opposite of Teaching for Equity, because under this system, the only kids who might get a real education are those from families that take more into their own hands, bringing higher expectations and resources to bear.
So, effectively no grading, then. You can do whatever you want all semester, no homework (so perhaps there’s some upside here?), phone out in class every day, whatever, all you have to do to pass is get 21% on an exam you can take multiple times. That was going to be it.
SF Standard: Just in: SFUSD is delaying a planned “grading for equity” initiative after the proposal sparked furious backlash.
Kelsey Piper: SF superintendent backed off immediately after the flood of negative feedback. This strikes me as a pretty dramatic change from how previous standards erosions were received, and a really good sign.
Most politicians want to make their constituents happy, and often their information environment is kind of terrible for that. It’s worth advocating for the stuff that matters to you. Don’t be an asshole, but be clear and outspoken.
San Francisco’s turnaround happened very fast. The Bay Area could become one of the best-governed parts of the United States inside a few years if we work to make it happen.
Well, maybe. They say they are ‘delaying’ the initiative. Which means they’re presumably going to keep paying the consultants, and they are going to try again to destroy all the incentives and measurements involved in education.
Fighting against algebra and grading is bad enough, but reading?
As in, people who want to ban teaching kids to read until age 6. No. Seriously.
Because they’re ‘not ready.’
Erik Hoel: 62% of American kids have a tablet at age 6.
They spend 3.5 hours every day on screens (increasingly, TikTok).
And because our school system waits so long to teach reading, they never get a chance to become readers.
“Education experts” have been saying for decades that we must wait to start teaching reading until 6-7 for neuroscientific reasons. These reasons appear, as far as I can tell, to be basically made up. Consider this recent article, which quotes a bunch of experts on this.
E.g., Maryanne Wolf says that brain myelination needs to reach a certain stage, and that teaching reading prior to 5 is “really wrong” and that she would ban teaching reading prior to 6 nationwide if she could.
I’m not going to bother quoting more of the evidence because this is so utterly Obvious Nonsense as to be beyond belief. Frankly, if I had a child that was 6 years old and couldn’t read I would not be thinking ‘good it is finally time,’ I would be debating exactly how much to panic and reassuring my wife not to panic far more.
The 3 year old I am currently supervising can somewhat read. I could read before my first memory (which was at 5 and involves reading books) so I don’t know exactly when it happened, and I learned without anyone trying to teach me.
I am going maximum opposite. There is no higher priority than teaching a child to read as early as they can handle it, and every actual parent knows this. There is a reason why the advice is constantly read to them, read with them, push reading. Reading enables everything else. The entire ‘education’ establishment really does need to flat out join the delenda est club.
In the name of them developing empathy for the people you force them to teach? As long as they pass a certain threshold of knowledge, the rest of their childhood, and indeed life, belongs to the people, and they’re a horrible person if they think otherwise, and the purpose of school is to teach them this?
No, seriously, this is something quite a lot of people, especially those in education, actually believe.
Setting all ethical or moral considerations aside, and even assuming that is the goal, what in the world makes you think this is going to work in the direction you want?
Tracing Woods: If a child is in a class, they should be there because it is the best environment to help them learn, not so they can act as an unpaid tutor to provide vague “peer effects” to others A system that abuses children as resources instead of teaching them to their level is unethical.
Joe McReynolds (3.4m views): That’s what life *is*, though! If you’re unusually smart/talented, your primary purpose in life is to help lift up others who weren’t born/raised as lucky as you were. Learning that sooner rather than later is important for developing a sense of altruism and communitarianism.
“With great power comes great responsibility” is a simple, true statement. To the extent being born with unusual (intellectual) power is an “innate characteristic,” that good luck means that you owe the universe hard work. You’re born with a debt that takes a lifetime to repay.
There it is, very explicitly. These people actually believe this. If you’re talented, your purpose in life is to be enslaved, to be forced to help others. Your life does not belong to you. Your labor does not belong to you. Your time does not belong to you. Who cares whether that benefits you? You belong to the people, from each according to their ability, at the barrel of a gun.
Kelsey Piper: If I were actively trying to extinguish my children’s sense of altruism, compassion and responsibility I can’t think of a better plan than forcing them to spend all of their time doing random ‘altruistic’ chores they didn’t choose, and aren’t equipped to succeed at.
If you say to your kid “I’m going to volunteer this weekend to help socialize cats”, they’ll probably come along and they may discover a lifelong love of helping animals! if you force them to spend all their time on it, guess what, they’re gonna hate it.
if you want your children to be people who give generously to their communities and their broader world, be that kind of person yourself and let them witness the ways in which this is part of the good for you and for your community.
SteelBlaidd: Why do they need to go to college to learn to be teachers if kids can be expected to do it in elementary school?
Kelsey Piper: some people out here believe that homeschooling is immoral since you don’t know enough to teach your kids and also that a smart 9 yo can do it.
Ben Hoffman: The important thing is that the 9yo is forced to do it without pay to teach them that being educated means going along with nonsense dramas, which qualifies them to get paid to go along with nonsense dramas when old enough that that’s dramatically appropriate.
Not only that the smart 9you can do it, that they should be forced to do it without pay. While the parent is forbidden to do it, because they are unqualified.
Ryan Moulton (QTing Joe above): Everybody is dunking on this, and I get why, but I’m a little more sympathetic. Particularly in lower grades, developing empathy for people different from you or dumber than you is a really important thing to get out of school, comparably important to getting through math faster.
Sarah Constantin: t is super common for kids to openly taunt anyone who’s not as good as them (at anything!) and i do think it’s important for them to learn manners, grace, and sportsmanship…
but it doesn’t deserve as much time in the day as math class.
Gallabytes: I would add that these kinds of assignments don’t necessarily breed empathy it’s pretty easy for it to create contempt instead.
Sarah Constantin: yeahhh.
Gallabytes: feels like people talk about school in far mode as some inscrutable thing.
if I put *youin a room with people you had 4 sigma on and told you to teach them math Or Else, how would you feel? how would this make you feel about them?
I believe we should treat Joe’s perspective the same way we would treat others who would force people with certain characteristics to labor for no compensation.
Tracing Woods (reference documents and more details in thread): in 1930, researchers studied ability grouping and concluded you needed to adjust the curriculum to make it work
in 1960, more confidently so
then in 1990, they studied grouping without changing curriculum, concluded it was useless, and advocated to get rid of ability grouping
over time the field got better and better at studying the form of ability grouping that everybody had known was pointless for sixty years while just sorta disregarding the form that kept getting results
I get so mad every time I read this stupid study
the field of education set itself back generations because it kept listening to people who thought ability grouping was “antidemocratic and antiegalitarian” and as such badly wanted it not to work
we had it figured out in 1936 and then we threw it away for kicks.
I am not a fan of the idea of educating children in 2025 primary via traditional classes. Traditional classes feel like learning a lot more than they actually cause learning.
But I accept that we are going to keep doing this for a while.
Given you are going to have traditionally shaped classes on various subjects, very obviously you want to track their progress and group those children by ability.
Helps almost all children learn more, whethertheyare behind, ahead or neither. There are some corner cases where kids are ‘on the bubble’ between tracks, or get tracked wrong, but mostly this is opportunity cost, that they missed benefits.
Is universally popular with parents, to the extent that ‘ending tracking’ is the least popular serious policy proposal we have ever seen. As in David Shor says ‘removing advanced classes from schools’ is literally the single most unpopular policy Blue Rose has ever polled (yet there goes Palo Alto doing exactly this.)
Ability grouping, done wisely, so utterly obviously works as to make the alternate hypothesis absurd.
Tracing Woods: will someone struggling with basic arithmetic and someone who knows calculus benefit from the same instruction? no.
would selective schools and the students in them benefit from opening their doors to everyone? they certainly don’t seem to think so!
do athletics orgs advocate grouping young LeBron into his local mixed-ability rec league so he can trarin and progress? no.
do gifted kids who accelerate learn more advanced material when they’re presented with it? yes.
if people see school as a democratic equalizer where everyone should learn the same things, they find ability grouping doesn’t work (doesn’t accomplish that goal). if people see school as a place where people should learn specific subjects and progess to whatever level they can, they find it does work (does accomplish that goal). and because education research is dominated by the former, onlookers glance at its output and say “huh, the results are mixed. guess we’ll never know!”
there is no substitute for understanding what is going on at a ground level.
The question is how to make it work best, not whether it works. Very obviously, as the next section discusses, it is possible to massively screw it up if you try hard enough.
Jordan Michelson: Why would *teacher’s unionsoppose ability grouping? It makes no sense.
Karen Vaites: The average American would be shocked by the degree to which K-12 education is ruled by ideology. Beliefs about teaching and how we want learning to work often trump evidence about what does, in fact, work.
Tracing Woods: “Trace, why are you making such a big deal out of something everybody already agrees with?”
Because the people we trust to direct society on this topic at every level oppose it.
And I hope that maybe if I shout enough about that people will really internalize what that means.
Exactly. Everyone agrees we want [X], where [X] is tracking. We keep talking about it because [~X] keeps actually happening.
I presume opposition is mostly ideology. Full stop. They want to prevent the wrong kids learning too much. They are sacrificing the kids on their alter.
Academics and education ‘experts,’ despite the literature and all actual observations and everyone involved saying that tracking helps all kids learn better, keep lamenting that parents want tracking, and work to destroy it, often in the name of ‘equity.’
It is common to see people claim ‘the research’ says that ‘downstreamed’ kids who are grouped at lower ability do worse rather than better as a result and that ‘the research’ supports this. As far as I can tell this is simply not true, these people simply think it ‘should’ be true, and seek out ways to say it anyway.
Tracing Woods: what happens when academics study parental perception of ability grouping?
They lament that parents of students at all levels favor it even though it’s BAD.
Virtually no parents support ending ability grouping.
Parents of kids in both remedial and G/T programs agreed that their kids should be grouped with kids of equal ability
80% of special-ed parents, 90% of parents with kids in remedial courses, and 98% of parents with kids in advanced programs agreed that their kids were helped by it.
This was all very disappointing to the authors.
Why do the authors here, like many other academics and education experts and many school principals who somehow end up actually destroying such programs, say that this is bad, despite everyone involved in the actual schools agreeing it helps all of the students?
Because the ‘educators’ who determine policy (as opposed to the teachers whose job is to actually educate children) consistently have decided that they do not care about the life experiences of families and children or helping children learn.
What they care about, other than money, is preventing learning rather than causing learning. Or, as they call it, ‘equality’ or ‘equity.’ Never mind that this ‘equity’ directly hurts the students who are otherwise being ‘denied’ it, what matters is that they be given ‘opportunity to learn and equality of educational opportunity.’ Educational opportunity shall be destroyed until this is achieved. If that leads to everyone getting a worse education, even the worst off kids, well, that’s not their department’s KPI.
I don’t quite agree with Anton that these people ‘hate you and your children.’ They only hate that you and your children might do better than other children, and want to prevent this from happening. They only hate you if you oppose this goal.
Garry Tan: Ability grouping in school (honors/AP) depends on your frame. If school’s job is to equalize, grouping looks like a fail. If it’s to let kids sprint ahead, it works.
Academia worships equalizing— so the School of Education bureaucrats become anti-education, ruining schools.
When you examine the list of nonprofits and academics that want to remove advanced math from classrooms and water down the standards for all students it will leave you shaken.
It’s not a fringe movement. It is School of Education Orthodoxy.
Tracing Woods: This is a fair question! Opposition to ability grouping is a fringe idea opposed by the great majority of parents. So which obscure, fringe organizations are pushing it?
Let’s ask the National Council of Teachers of English what they think:
Or what about the most prominent law casebook publishers?
Or consider the National Association of Secondary School Principals.
How about the Association of State
Supervisors of Mathematics, NCSM: Leadership in
Mathematics Education, and the National Council of Teachers of Mathematics (NCTM)?
You know it.
If it is a choice between the form of academia that wants to prevent children from learning, and the form of academia that helps teach useful things, and one or the other must be destroyed?
The choice seems clear.
Tracing Woods: my modest proposal to every university that has published research claiming ability grouping (with paired curricular modification) doesn’t work:
detrack. remove your admissions standards. remove course prerequisites.
if detracking works, let’s create Harvard Community College.
However, you do have to choose a reasonable implementation. Is it possible we also in some ways are screwing implementation up so badly that adding what we call ability grouping to the mix, as implemented in practice, could make things worse?
Tracing Woods: What happened when North Carolina changed its laws to require top-scoring students to be placed in advanced math?
The state board of education changed the test cutoffs, subverting the intent of the law by dropping almost all students from the top-scoring category.
Janet Johnson and John Wittle: The law was intended to help high school students who excelled in their math classes move into the advanced track. Before the scoring change, 11% of high school students statewide scored at Level 5, the highest level, with some districts seeing rates as high as 25%. The EVAAS Prediction vs. Performance table (above) showed that in 2009, 42,144 students were predicted to be successful in 8th-grade algebra, and only 18,670 students were enrolled. Using EVAAS prediction as the metric would have given 23,474 more students access to advanced math.
After the law passed, which required schools to admit all Level 5 students to advanced classes, and the state changed the scoring scale, fewer than 1,500 (too few to report) high school students in the entire state achieved Level 5.
The schools had technically complied with the law while completely subverting its intent. These charts are from the NC School Report Cards.14 After 2019, the Math 1 Performance charts show no Level 5 and very little Level 4, similar to this.
The full post on ‘the Algebra gatekeepers’ keeps outlining all the tactics used to ensure that kids do not learn algebra, especially disadvantaged kids. As you read it, it keeps getting worse.
For example, we attended a meeting with the parents of a middle school girl who had earned an A in 7th-grade pre-algebra but was denied enrollment in 8th-grade algebra despite her and her parents’ wishes. The teachers argued that her formative assessments didn’t align with her summative performance, suggesting her previous success didn’t guarantee future outcomes.
The language arts teacher claimed the student had “appeared to struggle” during benchmark activities and that earning an A seemed “harder for her than for other students who achieved similar summative data results.” The math teacher who had given her an A pointed to C grades on some earlier formative assessments, arguing that despite her subsequent A performance on chapter tests and other summative data points, these initial benchmark scores indicated she “sometimes struggles with foundational concepts during the formative assessment process.”
The administrators nodded knowingly as teachers referenced “inconsistent performance across benchmark measures” and “concerns about the gap between formative and summative data trends.” They suggested that while her final grades represented solid summative data, the formative assessment patterns revealed “areas of concern” that made advanced placement inadvisable, regardless of what the summative data actually showed about her mastery of the subject matter. This is just an example of the kind of talk the parents encountered, so the church advocacy group started bringing someone from our staff to the meetings to help them cut through this.
This type of circular reasoning, where success was reframed as evidence of potential failure, was typical of how schools justified excluding qualified students from advanced courses.
Administrators would routinely promise that students could move to advanced tracks “later, in high school,” but our analyses and other research we found indicate students rarely moved to the advanced track after the 8th grade.
The tracking system created rigid pathways where missing 8th-grade algebra typically meant students couldn’t reach calculus by graduation, limiting their college and career options.
…
One fifth grader exemplified the problem. He had scored at the highest level on every test, had a 98% EVAAS prediction for success, and straight As on his report card. He was also officially classified as Academically Gifted by the school. When a new advanced math class was created, he wasn’t invited. When he asked to join, his parents were told no, he needed to be “recommended.” School officials told us, with straight faces, that his consistent past success was no indication he could succeed in advanced math, that they were keeping him out for his own good.
What North Carolina is doing here, excluding lots of qualified students, does at least seem better than ending algebra entirely for everyone, I suppose.
Raymond Arnold points to the best objection I have seen, which is that if a more advanced option exists then many parents will push for it even when it is inappropriate for their particular child, or use it to push their child way too hard, and while this means some kids are bored it is saving a lot of families from being forced into a red queen’s race.
This is a real cost, but the prior should be rather extremely stacked against ‘if we let kids learn more then parents would try to have their kids learn too much and this would be bad,’ especially when you can gate the advanced classes with objective tests. Yes, parents can push their kids to study harder to try and pass those tests, but that’s a risk I am willing to take.
What is the steelman case that ‘ability grouping doesn’t work’ or ‘ability grouping has been tried and didn’t work’ in some particular context?
This by Karen Vaites is perhaps the closest, in particular on early reading grouping, convinced me that in practice you really can mess it up badly enough to make things actively worse. This was convincing that we’re messing up badly enough that this is a real possibility.
As in, what happens in practice is that you group kids by a measurement of abstract ‘reading level’ and then focus on ‘achievement’ of ‘reading level,’ forbid them to read anything beyond ‘reading level,’ and don’t ask what actual skills they need, don’t move them between groupings as their skills change, and then wonder why it isn’t working.
One could almost say, if you look at the details, that the teachers are using ‘reading groups’ as a substitute for actually teaching the children to read. You put them in a group and then you did your job. Again, yeah, I can see how that wouldn’t work. Indeed, if you are outsourcing the teaching job to other kids, then at that point you actively want uneven groups, because you want to group students with student teachers.
Whereas once students get to ‘escape velocity’ on reading, which the better students have relatively early, they no longer need a teacher, they just need motivation and permission to read books. Whereas the system seems designed to stop kids from reading books they want to read if they are deemed ‘too hard’? A kid can tell you if a book is too hard, they won’t want to read it.
One big complaint is that it is ‘hard to measure reading level.’ I don’t think it is hard. You can observe a lot just by watching. The problem is that you’re measuring a set of distinct reading skills as if it was one number, and then treating that one number as real, and also abdicating all the real work.
Sarah Sparks: But evidence suggests that the practice may be less beneficial than teachers think: It can exacerbate achievement gaps and even slow reading growth for some children unless the groups are fluid and focused on skills rather than overall achievement.
…
“What we’ve discovered is that it’s fine to have a group of students of different levels, as long as they all are working on the same learning needs,” said Carol Connor, an education professor at the University of California, Irvine, who developed the program. “You can have students of different reading abilities who all need to work on decoding. … What doesn’t work is if you put your kids who already know how to code in a group to learn how to code, again. You receive more behavior problems because they’re really bored, … and our research suggests that it has a negative effect on their growth.”
Karen Vaites: Tim Shanahan breaks down key research and its instructional implications in The Instructional Level Concept Revisited: Teaching with Complex Text. As researchers looked into the effectiveness working at reading level, studies found that it “has made no difference—that is the kids taught from grade level materials do as well as those at an instructional level—or the instructional level placements have led to less learning.”
More recently, he highlights additional new evidence from a study of third graders: “Results indicate that weaker readers, using texts at two, three, and four grade levels above their instructional levels with the assistance of lead readers [other, better reading, third graders], outscored both proficient and less proficient students in the control group across multiple measures of reading achievement.”
From a question: My daughter is in first grade. Her classroom teachers have all the books in the classroom library leveled, and students are not allowed to go beyond their reading level during “Independent” reading.
From the answer: Your daughter’s aspirations as a reader are the problem here. Some kids are allowed to read the red-dot books and others are stuck with the baby books with the blue dots. She wants to be a red dot kid, to hang with the red dot kids, to be seen as a red dot kid … but her teacher can only see her as a blue dot kid and she must learn to stay to her own bookshelf with her own kind if she is going to succeed in this classroom.
…
In the meantime, explain to your daughter that the teacher is trying to help her but that we teachers sometimes don’t get it right, and that you can’t always “fight city hall.”
So yes, you group primarily by what aspects the kids need to work on most, and that works better. Sure. I can totally believe that is a better strategy. Skill issue.
Instead of using it to figure out what kids need to do to learn to read and putting them in position to learn that, it sounds like grouping is being used to prevent kids from meaningfully reading? The purpose is to gate reading behind general tests? To spread ‘equality’ to progress on different reading aspects?
Sarah Sparks: It sounds like good sense. “Kids should be reading just-right texts as they grow as readers.” That just soundssensible, doesn’t it? Many urban legends do… until you know better.
I can see why that might actively backfire. This isn’t about ‘ability grouping’ not working. It’s about failing to actually group by the relevant ability, and it’s about the ‘just-right text’ theory that seems to me obviously wrong.
Sarah Sparks: During Tier 1 instruction, you want all kids working with grade level texts; students reading below grade level will need scaffolding and support (as well as targeted Tier 2 and/or 3 intervention).
This promotes equity, for it’s the best mechanism for helping below-benchmark students to catch up.
It also honors the fact that a fifth grader who reads at second grade level is still thinking at the level of a fifth grader, and he or she will remain engaged and motivated by learning content and vocabulary at his or her developmental level. (No more baby books for big kids, y’all!)
Ignore the ‘this promotes equity’ framing, since you could simply say ‘this promotes learning to read.’ Equity via catching up those lagging is good, and you call it learning.
The theory here is that age matters a lot. That if you are a fifth grader, your ability to learn is inherently much stronger than that of a second grader, whereas the ability of different second graders is alike? Equality (of those at the same age) for me, inequality (of those at different ages) for thee. Whereas the correct model is that each kid has a different ability to learn different things, that usually improves steadily with age.
But also note that this is saying that the best way for many students at second grade reading level to learn reading is to assign them to read fifth grade level books, indeed to mandate it. Yes, I can believe that. So why are we so often telling kids at second grade reading level or literally in second grade or both that they can’t read the fifth grade books even when they want to?
The other theory present in this proposal is, how about using techniques that actually teach kids reading. And yeah, I agree, that would be great.
Tracing Woods (replying to Vaites): This is an extremely useful article that deserves a full, thorough response.
My short response is that I agree narrowly (most leveled readers seem quite bad, training specific skills matters, in-class grouping for reading is quite popular and often pretty uninspired) and disagree broadly (there is pretty strong evidence for the value of several forms of ability grouping, drawing from eg Direct Instruction, Success for All, acceleration/gifted literature; ability grouping has acquired a bad reputation for reasons mostly unrelated to its performance; “grade level” is the wrong measure) in a way that would be productive to hash out more fully.
I agree fully that the real question is cultural.
In case you didn’t realize that there is a war. There has been for a while.
For those with visual impairments, language models can summarize visual content and reformat information. Tools like ChatGPT’s voice mode with video and Be My Eyes allow a machine to describe real-world visual scenes in ways that were impossible just a few years ago.
AI language tools may be providing unofficial stealth accommodations for students—support that doesn’t require formal diagnosis, workplace disclosure, or special equipment. Yet this informal support system comes with its own risks. Language models do confabulate—the UK Department for Business and Trade study found 22 percent of users identified false information in AI outputs—which could be particularly harmful for users relying on them for essential support.
When AI assistance becomes dependence
Beyond the workplace, the drawbacks may have a particular impact on students who use the technology. The authors of a 2025 study on students with disabilities using generative AI cautioned, “Key concerns students with disabilities had included the inaccuracy of AI answers, risks to academic integrity, and subscription cost barriers,” they wrote. Students in that study had ADHD, dyslexia, dyspraxia, and autism, with ChatGPT being the most commonly used tool.
Mistakes in AI outputs are especially pernicious because, due to grandiose visions of near-term AI technology, some people think today’s AI assistants can perform tasks that are actually far outside their scope. As research on blind users’ experiences suggested, people develop complex (sometimes flawed) mental models of how these tools work, showing the need for higher awareness of AI language model drawbacks among the general public.
For the UK government employees who participated in the initial study, these questions moved from theoretical to immediate when the pilot ended in December 2024. After that time, many participants reported difficulty readjusting to work without AI assistance—particularly those with disabilities who had come to rely on the accessibility benefits. The department hasn’t announced the next steps, leaving users in limbo. When participants report difficulty readjusting to work without AI while productivity gains remain marginal, accessibility emerges as potentially the first AI application with irreplaceable value.
Google objects to this characterization. A spokesperson calls it a “cherry-picked” line from the filing that has been misconstrued. Google’s position is that the entire passage is referring to open-web advertising rather than the open web itself. “Investments in non-open web display advertising like connected TV and retail media are growing at the expense of those in open web display advertising,” says Google.
If we assume this is true, it doesn’t exactly let Google off the hook. As AI tools have proliferated, we’ve heard from Google time and time again that traffic from search to the web is healthy. When people use the web more, Google makes more money from all those eyeballs on ads, and indeed, Google’s earnings have never been higher. However, Google isn’t just putting ads on websites—Google is also big in mobile apps. As Google’s own filings make clear, in-app ads are by far the largest growth sector in advertising. Meanwhile, time spent on non-social and non-video content is stagnant or slightly declining, and as a result, display ads on the open web earn less.
So, whether Google’s wording in the filing is meant to address the web or advertising on the web may be a distinction without a difference. If ads on websites aren’t making the big bucks, Google’s incentives will undoubtedly change. While Google says its increasingly AI-first search experience is still consistently sending traffic to websites, it has not released data to show that. If display ads are in “rapid decline,” then it’s not really in Google’s interest to continue sending traffic to non-social and non-video content. Maybe it makes more sense to keep people penned up on its platform where they can interact with its AI tools.
Of course, the web isn’t just ad-supported content—Google representatives have repeatedly trotted out the claim that Google’s crawlers have seen a 45 percent increase in indexable content since 2023. This metric, Google says, shows that open web advertising could be imploding while the web is healthy and thriving. We don’t know what kind of content is in this 45 percent, but given the timeframe cited, AI slop is a safe bet.
If the increasingly AI-heavy open web isn’t worth advertisers’ attention, is it really right to claim the web is thriving as Google so often does? Google’s filing may simply be admitting to what we all know: the open web is supported by advertising, and ads increasingly can’t pay the bills. And is that a thriving web? Not unless you count AI slop.
