Interesting enough to click the link, but quickly lost interest after the intro. It seems to be a "listicle" about other essays that are neither linked nor described in an interesting enough way to follow up by googling or some such. So what's going on there?
The US military is infamous for producing hideous PowerPoints, if anyone told them about nine-way Venn diagrams the result might be ugly enough to kill its audience.
1. Check the one he made admonishing the maestro for not making use of the black keys as well as the white and expecting harmony, and the one admonishing the "America First" crowd for not valuing the lives of foreigners
2. Check the propaganda the Japanese were putting out against the Americans. Fat, sweaty, double-chinned, thick-fingered, fang-toothed demons. There is no American Suicide Cliff
Somebody should've explained this distinction to the women who chose to throw themselves off a cliff rather than go through whatever the lunkheads had in store for them. Very culturally insensitive of these women not to respect the nuances of translation.
Re: 15, I'm not paying for his substack so I can only read the intro but even if it is true that the effect is all due to "increased communication from patient to doctor, and through increased adherence to a course of treatment" that doesn't at all answer the question of whether racist attitudes by doctors are partially responsible. A patient who has a doctor who either explicitly or implicitly communicates to the patient that he is untrustworthy or whose doctor exhibits hostility is naturally going to be less communicative. I'm not Black but I've certainly had doctors who signaled "i'm really not interested" before I got to a diagnosis and I was less communicative with them. Similarly, if a doctor is hostile to me in a way I perceive as biased I'm naturally going to be less likely to follow their treatment. Now I'm sure general perceptions of medical bias play a role beyond what any individual doctor does. But chalking it all up to bias by patients is extreme.
The whole premise of the essay is ridiculous in the first place. We should not engage in affirmative action as a way of catering to racist patients. Are we going to start affirmative action for white suprematist doctors next so they can build rapport with racist white patients? The obvious solution is to stop promoting false and bigoted misinformation that stirs up racial resentment.
RE 4, about Omelas: To the extent that this interpretation of Le Guin’s point is true, I think she makes it very poorly. When you spend the first two pages of your four-page short story describing how perfectly wonderful a city is, and how the details of its perfection can be whatever helps the reader accept this, you are clearly setting up a twist. I think that utopias are possible, and I like stories that involve them, but of course I expect (correctly) that Omelas is going to have some shocking drawback. To really make the point about people not accepting worlds without suffering, you would need to write a genuinely compelling story set in such a world, and only then challenge the reader’s belief. Otherwise the thing you’re demonstrating is just “people expect stories to have conflict or characters or specific worldbuilding”.
Even more so, given that we have a much richer and expanded source on Le Guin's understanding of the possibility of Utopia: One subtitled "An Ambiguous Utopia" no less!
I never liked that story or understood why people like it. To me it just feels like a plot sketch for a C-grade Twilight Zone episode.
I mean yeah, the concept of perfection-on-the-surface-enabled-only-by-deep-secret-suffering-behind-the-scenes is solid enough, if somewhat overplayed. But the actual execution of "uh yeah so there's a child being tortured" sounds like a placeholder that you put in until you think of something less twee and more interesting.
The beauty of the story is that a lot of people react to it by self-righteously declaring they would most certainly walk away from Omelas, without realizing they've been set up, and that Le Guin simply described the West, the opulence of which is propped up by exploiting what is currently termed the global South (our sneakers and phones are literarily assembled by children in basements).
Yes, but as always economics is about next best alternatives. What would these children be doing if not building our sneakers and phones? Mostly backbreaking agricultural labor and/or starving. If the tortured Omelas child's only other available alternative was an even worse torture chamber that doesn't even benefit anyone, that would change the story significantly.
Who said anything about hunter-gatherers? Shutting down Nike factories isn't going to return Bangladesh to a state of nature, any more than their construction took it out of one. Most of the places that manufacture western widgets have been settled agricultural societies with entrenched hierarchies for thousands of years. The people of Vietnam were not leading bucolic lives of leisure before the white man showed up.
Peasant farming and sharecropping were pretty awful, too. Indeed, by current standards, or even standards of rural America in 1926, life has basically been really hard for most humans for most of our history.
Bangladesh _might_ return to a state of nature if the factory closures shut off the income from goods that Bangladeshians use to buy food, touching off a mass starvation that tapers off once the Bangladeshians who are left are able to sustain themselves with hunting and gathering on the land in their vicinity.
In other words, "fixing" Omelas (a la Jemisin, for example) might entail saving that one poor child only by killing off many others, including other children.
I imagine those who walk away from Omelas including a few who discover this, and the reason they did not return may be that they perished in that waste. A few more discovered it, but founded their own iron city where they torture even more, but tell themselves it is preferable to the torture of one.
Surprisingly, a lot less than members of early agricultural societies.
As David Graeber called it, there is the "damn, all the bad places are taken" misconeption about hunger-gatherer societies. We assume that all such societies were like those we can (or could, in the foreseable past) monitor, living in places no one else wanted (in the Arctic, in deserts or in jungles). Naturally, when all (or almost all) societies were hunter-gatherers, most of them lived in ecologically fertile areas.
(Furthermore, since the industrial revolution, we systematically destroyed nature; wildlife got rarer even as recently as from the 70s (ok, I have only anecdotal evidence for it from family reports). Asking "what would I do as a hunter-gatherer in the place I am living now?" is a straightforward and pretty bad question.)
The idea is not about critiquing the system, which might be better than any alternative an individual can bring about, it’s about participating in and benefitting from that system.
Every society we know how to build has some people whose misery is baked into the structure of the society. If nothing else, consider people whose violent tendencies make them a danger to those around them, and so who end up in prison. Prison doesn't have to be as bad as we often make it, but it's never going to be where people want to be, and yet we have no way to make a livable society that doesn't put at least the most dangerous and violent people in prison.
Yeah, and my point is that this is all stuff we have to do to maintain a society.
Any society we build will have people at the bottom. We can try to make things as good for them as we can: humane prisons and mental institutions, decent housing and treatment for seriously disabled people, poverty programs to make sure that the very poor have enough to eat and a warm place to sleep, nursing homes that are no worse than they have to be for the senile).
We also have dangerous and unpleasant jobs we must have done, and again, that means some people will endure hardship--coal miners working a dirty and dangerous job, soldiers getting shot at by enemy soldiers, plumbers working knee-deep in human waste when that's necessary to fix a problem, linemen hanging onto a power line in a bad storm to get the lights working again, people wiping asses in a nursing home, etc.
We can't be rid of any of these things. Maybe in some SF future world, or Heaven, but not in the world we know how to build. In this world, somebody's going to be the miserable child in the basement, and the best we can do is to minimize the number of miserable children in basements and try to blunt the worst of what they must endure.
I really don't know of anyone who *likes it*. I read it as a teenager, and I was profoundly unsettled by it. Unsettled enough to read it over two more times, hoping I had missed something, and I became even more depressed! Everyone seems disturbed by it. But the fact that it's so universally disturbing makes it a kind of short story masterpiece.
Yeah I was thinking about calling out that episode as a better (but still flawed) take on the same basic idea. I particularly like, in that episode, the fact that everyone regularly gets told about the disturbing reality, and has the choice either to protest against it or have their memory of it erased... and if enough people choose to protest then they'll do away with the whole system but everyone will die.
Then it's marred by the need to be a Doctor Who episode (which means that the system needs to be personified by guys in weird costumes who can chase the Doctor) and the cop out happy ending (a regular Doctor Who trope where a some moral dilemma is posed but then in the end it turns out that if you make the "BBC-right" choice there's no downside after all).
Now that the Omelan accommodation to the “terrible justice of reality” has been described, Le Guin believes she has finally quieted her reader’s doubts. “Now do you believe in them?” she asks. “Are they not more credible?” This is the key juncture in Le Guin’s negotiation with her contemporaries but it is not a triumph. The moment her utopia has been altered to the point where it is believable, it is no longer a utopia. It contains privilege, victimization, and pain. It is morally unacceptable even if the pain is limited to a single innocent person.
Le Guin is dramatizing the dilemma which, elsewhere, I have traced through the history of utopian thinking. Utopianism provides a powerful moral critique of the culture of competition, whether feudal or capitalist. But the moment it tries to replace that culture with one which satisfies utopian ideals, it becomes socially and psychologically implausible. To make it believable it is necessary to add evil and pain. A perfectly just world where no one has to pay for the happiness of the majority is simply unimaginable for human beings, with the strong implication that what is impossible to imagine is also impossible to achieve.
Basically the same point Agent Smith makes when he recounts how the first Matrix was a failure and the best the machines could do was simulate the peak of real human civilization.
There's something about Omelas that seems to make the rats and the rat-adjacents view it as a moral thought problem that requires a utilitarian solution. And they also ding LeGuin for her sloppy world-building (which doesn't meet their rationalist standards of self-consistency). I suspect that this is because modern culture has discarded mythic dream narratives, in which meaning is elusive, but where impression dominates. Indeed, LeGuin said the central idea of Omelas is a "psychomyth," but credits William James for the scenario that he put forth in his essay, "The Moral Philosopher and the Moral Life" (link below). LeGuin was the daughter of the cultural anthropologist Alfred Louis Kroeber, and she was quite familiar with the importance of myths as being non-rational expressions of human culture. But Omelas seems to bug everyone, the rats and non-rats alike, because it's just such an unsettling vision. However, the rats seem to want to pick it apart, and the non-rats would prefer to walk away from it.
And it's still unsettling us fifty years after she wrote it! There's been a new batch of Omelas talk lately. John Smith says it shouldn't be treated as a moral Rorschach test, but then goes on to treat it as a moral Rorschach test. According to Smith, those who walk away are caught in some sort of moral purity trap.
I've always thought Omelas made most sense as a critique of the dystopian imagination rather than a played-straight example of it. That’s what Le Guin's fourth-wall-breaking comments point to.
Her own (friendly) critique of utopianism, in The Dispossessed, doesn't have anything as crude as a randomly tortured child. It ultimately points toward some helpful tweaks the utopians could make to be friendlier to certain types of dissent -- not toward a repudiation of utopianism because there's some inevitable dystopia lurking underneath.
As a critique of utilitarianism or utopianism, Omelas limps. It holds together much better as a critique of the insistence that any apparent utopia must have its inescapable cost, its seamy underbelly -- the “treason of the artist” that presumes pain is a more sophisticated topic for art than happiness.
“Now do you believe?” she writes; and I never have. My sympathies are with the walkers-away, and not primarily because I’m on Ivan Karamazov’s side when it comes to utilitarian thought experiments involving tormented kids. The hidden horror of Omelas is too contrived, its necessity impossible to believe.
That said, there are some great riffs on Omelas played straight, and one of them from a couple of years back deservingly won a Nebula: https://clarkesworldmagazine.com/kim_02_24/
I think the point is clear enough from the question - but you can still take that ambiguously. It can mean "what, you can't believe in a perfect utopia? See how you'll change your tune if you also think there's something bad about it", but it can also mean "what, you can't believe in a perfect utopia? You were right to be suspicious".
But to answer to your take - yes, it's true that you could say this is obviously expected from the common structure of a story (with conflict etc). But then again, the problem here is exactly that: people are prone to think so much in story-logic, they apply story-logic to real life, and that's part of what makes them suspicious of unalloyed goods.
What I like about the 4-way Venn diagram, is you can easily make is it as a 4x4 grid (including an empty space for the empty set) - e.g. the first two rows are "A", the middle two rows are "B", the first two columns are "C" and the middle two columns are "D"
It turns out the generalization of that construction is that a 2n-way venn diagram requires 2^n rows and 2^n columns. And of course, this is just an application of the fact that an n-way Venn diagram can be replaced by a 2^n row truth table.
Yes - if you expect the intersections of a 2-dimensional display to capture the distinct / unique spaces for the Venn diagram. But I'm not sure the column and row titles are then very intelligible, except to the well-versed users. Instead, perhaps 1) simply overlaying the 2-d display with other overlaps could be possible, or 2) I wonder what expanding our display to 3 dimensions might do to simplify the 2^n challenge.
The issue is that after 4, you can no longer keep the regions contiguous - so instead of "the left two rows are A", you end up with something like "rows 1,3,5 are A", which means you couldn't do it by just drawing N regions
I can't ever read Omelas the same after knowing that the bay area science fiction community in the ~60's-80's of which it is a product had known serial child molesters no one did anything about. It was supposed to be an open and weirdo tolerant space, and maybe being against pedophilia is just puritanism anyway, and parents can just keep their kids away from him if they're personally oppoaed to it. Letters were written about what to do about the main known molester, Walter Breen. People didn't want to call police, for various reasons but notably including "it will cause a public scandal and start a witch hunt that will ruin our wonderful community". I guess most people did choose to stay in Omelas...
I will ask you one more time to put more effort in your post. You could have posted this instead of not adding any useful content.
Arthur C. Clarke might be a related case. In 1998, the Sunday Mirror published allegations that he had sexual contact with underage boys in Sri Lanka. Clarke denied the allegations, asked that his knighthood ceremony be postponed until the matter was examined, and Sri Lankan investigators later said they had no complainant and no basis for a case against him. However there are possible parallels with the fandom not wanting to know this information due either hero worship or other sociological forces, even if the evidence is not the same
Yes, I've heard about that. Still, I _really_ like a great deal of Clarke's writing, most particularly Karellen's last last speech to mankind in Childhood's End, containing these words about mankind's portion of the Overmind:
"For what you have brought into the world may be utterly alien, it may share none of your desires or hopes, it may look on your greatest achievements as childish toys - yet it is something wonderful, and you will have created it."
which are something of a consolation in these years, possibly the formative years for AGI or perhaps even an ASI...
Legit. Google Walter Breen (mentioned) and Marion Zimmer Bradley. Preferred ages seemed to be 8 to 13 ish, though there are mentions of children as young as three. Breen was convicted so it's not just allegations.
Legit. The sad story of Marion Zimmer Bradley, her second husband, and her daughter who years later made allegations that her mother also sexually assaulted her:
It seems it wasn't "she wasn't quite 18" in that Breen apparently preferred boys, and one case was a 13 year old boy saying Breen had been molesting him for four years (so since the age of 9 or so):
"Walter H. Breen, 1928-1993, also known as E.Z. Eglinton, was the author of the 1964 book "Greek Love," as well as numerous articles on homosexuality, sexology and numismatics, and the 1953 book "Proof Coins Struck by the United States Mint, 1817-1901." He was married to science fiction writer Marion Zimmer Bradley in 1964. They had two children and separated in 1979. In 1979, his concordance of Bradley's science fiction Darkover series was published, "The Darkover concordance: a reader's guide." He was arrested on child molestation charges in 1990, and in 1991 he was charged with eight felony counts of child molestation. He was sentenced to ten years in prison and died in prison in Chino, California on April 22, 1993."
Allegedly Breen and Bradley had a lavender marriage (by this account) but they had two children and both of them had previous marriages, so who knows what exactly was going on?
"An avowed homosexual, he married author Marion Zimmer Bradley, an avowed lesbian. It was her second marriage and lasted from 1964 to 1979, and they had two children.
His penchant for young boys led to his conviction in 1964 for child molestation. He was also convicted in the early 1950s for lewd behavior. In 1990 he pleaded guilty to child molestation and was sentenced to 13 years in prison. He had cancer at the time and died after serving one year."
Timeline in the link here says one of the scandals involved him and a 10 year old boy. Lovely.
This becomes a little more understandable once you realize just how incredibly puritan was US circa 1950. Sex between two men was illegal. Sex between two women was illegal. Sex outside marriage (I do not mean adultery, I mean sex between *any two unmarried people*) was illegal in several states. Sex between a white and a black was often illegal. Anal and oral sex was illegal -- even between two married heterosexuals. Heck, in some Southern jurisdictions sex with woman on top was illegal! (No idea how they enforced it)
Now imagine you are a smart, non-conformist, quite possibly neurodiverse science fiction fan circa 1970, fully cognizant of how arbitrary and oppressive is everything I just listed, yet blissfully unaware of the long-term effects of child sex abuse (because the same prudishness prevented anyone from studying these effects). Then add some very naive ideas about consent, and it is not a long leap to adding "and sex with children was illegal" to the above list of absurdities. It took several decades before "Wait a minute, that's different!" percolated into people's consciousness.
As an example, I am fairly certain Robert Heinlein did not see anything morally wrong with sex between adults and 14-15 year olds, even though he himself had no such inclinations.
I do somewhat agree with this, there was a similar movement in European intellectual circles at the time, with Foucault, de Beauvoir et al. But still, even many adults at the time who were very much accepting of homosexuality still objected to pedophilia, as described by the people in the fandom themselves. In one home, parents asked kids to barricade themselves in their room whenever Walter Breen (later convicted) was over. Here's a "what should we do about this guy" letter that was mailed around at the time (super graphic) https://breendoggle.fandom.com/wiki/Breendoggle_Wiki Utterly fascinating time capsule.
That's why I specified "smart, non-conformist, quite possibly neurodiverse science fiction fan". Ironically, the "more normal and boring" people likely had a better sense on this topic.
I wasn't clear, there was definitely a debate around pedophilia in the sci fi circles, whereas no one cared about homosexuality. This is described in the letter linked.
I am aware of the debate, but the fact that THERE WAS a debate[1], with some individuals genuinely and without ulterior motives on the side of "13 year olds have sexual desires, who are you to tell them otherwise?" can be understood only in the context of the times.
[1] As opposed to either universal condemnation, or "MAP" weaseling
People definitely cared about homosexuality in sci fi circles. Heinlein included a scene in the very slutty, very sexually promiscuous novel Stranger in a Strange Land where, the morning following the protagonist’s first threesome, his female mentor in matters of sexuality makes sure to give him a lecture on how threesomes are perfectly fine as long as it’s two women, but it’s not proper to ever share something like that with another man. And the sexually ignorant protagonist replies that that’s obvious.
That would be more compelling if we were talking about, as FallingKnife puts it, "she wasn't quite 18" pseudo-pedophilia. There's plenty of room for honest disagreement about how soon after puberty are people allowed to make their own decisions about sex. And that does fall into a continuum where "The Man says I can't have sex with my 17yo girlfriend" is of a kind with "The Man says that if I have sex with my 19yo girlfriend it can't involve butt stuff".
But the Breen/Bradley case involved unambiguously prepubescent children of single-digit age. If you find that a middle-aged man is diddling a 9-year-old boy, and your response is either "OK, but have there been any double-blind clinical trials saying this is harmful to 9yos?", or "Hey, The Man won't let me do the butt stuff with my girlfriend, so why should I care?", then I think it's pretty safe to say that there is something wrong with you.
I agree, there WAS something wrong with the entire community. I am not defending them. But it may not have been quite so obvious back then. And to follow up with Heinlein example, he would definitely have been horrified by diddling a 9 year old.
I think people who didn't live through the 60s/70s (especially the 70s) have no idea what society was like back then. I was a child during the 70s but enough of American culture percolated through, especially in pop culture, that I got the general gist.
Attitudes were both regressive and progressive: the Sexual Revolution was in full swing, so things like "The Man says I can't sleep with my girlfriend" or "do butt stuff with my girlfriend" were being challenged. Homosexuality was benefiting from the more liberal attitudes, though today it would be seen as regressive; this is why it got included in the diagnostic manuals, after all! Gays aren't depraved evil monsters, they're sick. They're born that way, they can't help it, we should have sympathy for them and not throw them in jail.
So the idea was that women should sleep with men because liberation, but also attitudes were still quite sexist. The ideal for a guy was the playboy image, now instead of getting married and settling down early like a square, you could have a string of girlfriends and/or casual sex, because this was the 70s baby and we've thrown off all that old repression of our parents!
It was also the time of the Me Generation, all the new fads like TMI and psychological theories and new movements about finding your authentic self and living for yourself not others and so on. This is also in part why divorce rates increased and the movement towards no fault divorce: if one party in a marriage is growing and the other party won't, or refuses to grow along, then the first party has a duty to themselves to leave that toxic, limiting relationship and follow their bliss.
And the attitude towards sex and sexuality followed that pattern. Repression and ignorance had caused all the hang-ups (psychiatry tells us so) so by being open, non-judgemental, and education, all the facts about sex as a perfectly natural human need would be known by all and we'd avoid all the drama.
That included children. We pass on our hang-ups and repressions to our kids, after all. So things like "parents should normalise nudity in the family home so kids can see what adult bodies look like, then they won't be getting the wrong information and developing neuroses around sex". Children were plastic, blank slates; they wouldn't remember things at a very early age so those things wouldn't affect them. Children were sexual beings too, and pretending otherwise in some kind of worship of innocence only did them harm. Adults (loving, caring adults) helping educate children with sex play wouldn't do harm, in fact the opposite!
If you read one of the anecdotes in the 'Breendoggle' article which got the author, Bill Donaho, excoriated by the SF community, I think you'll agree that the definition of "sex play" was being stretched to its limits, and how desperately in tune with the times (or indeed genuine believers in all the liberation stuff) the people of that community were trying to be, because SF fandom was *special*, we tolerate the weird and the outsider because we've all been called weird and outsiders, fans are slans. WARNING - THIS MAY CAUSE DISTRESS TO READ:
"The second cause was Walter's sex play with 3-year old P----------- --------------. He had her trained up to the point where she would take off her clothes the minute she saw him. He would then "rub her down" and all that. I recall one occasion -- a fairly large gathering at the Nelsons -- in which he also used a pencil, rubbing the eraser back and forth in the general area of the vagina, not quite masturbating her. (Walter is incredible.) Many people were somewhat displeased by this -- most particularly her parents. No one thought he was actually psychologically damaging P--------- (she being so young) --- obviously ----- and ---- would have interfered if they thought he had been -- but the spectacle was not thought to be aesthetically pleasing. Years later Walter found out about the reaction and said, "But why didn't somebody say something! I wouldn't have dreamed of doing it if I'd thought someone _objected._"
Now, I'm just a repressed, Catholic-guilt, conservative, right-wing prude from a small town with no enlightened liberal values, but if I were in that room, I hope I would have found the nearest heavy object to hit that fucker over the head with.
And the Breen story, not to make condemnatory parallels but just because the pattern of an abuser is so damn textbook, reminded me of how a similar case of alleged abuse within the Bay Area rationalist community was handled. The guy was involved with a 15 or so year old, and used all the same tricks: I'm suffering from problems, I'm sensitive, I'm special, if you are harsh to me I'll kill myself, etc.
And the nice, liberal, 'no kink-shaming', progressive values rationalists who genuinely had only the best and highest motives had no idea how to deal with this. Run him out of town on a rail? Impossible! Call the cops? Impossible! No, we'll set up a committee to study this and give recommendations, and in the meantime we'll just have a whisper network of maybe don't invite this guy around to meetings or the likes.
I am aware of that. But from the viewpoint of a social rebel, it made contempt for the sex-related laws even greater: "Look! This law is so ridiculous, even police and prosecutors ignore it! The laws which are similar, must be equally ridiculous."
Generally speaking, and in many contexts which have nothing to do with sex, *unenforced* laws tend to undermine respect for laws in general.
It isn't just about SF fandom. I don't think the story makes its case very well, but in the real world, there are parents who torture their children, there are wars, and we get less considerably less than Omelas while having a worse situation for a large number of children.;
Walter Breen was a science fiction writer? Did he publish under a pseudonym? And which of the many SF Bay Area SF writers were his apologists? You've piqued my curiosity with your accusations. Did LeGuin know about his activities and she covered them up? Other famous SF authors also come to mind. Philip K. Dick? Robert Silverberg? Poul Anderson? Frank Herbert? Fritz Leiber? Were later figures also involved in the cover-up? Bruce Sterling? Rudy Rucker? Pat Cadogen?
Never mind. I asked ChatJimmy, and he gave me the lowdown. It all seems like pretty weak tea, though...
Me:
Which SF Bay Area science fiction authors were tainted by the Walter Breen pedophilia scandal?
Chat Jimmy:
You're likely referring to the Walter Jackson Breen scandal, which involved a writer and editor in the 1960s and 1970s who was exposed for his predatory behavior, including pedophilia. Some authors from the SF community were accused of having connections to Breen or being influenced by his behavior, but it's essential to note that none of them were directly accused of the same crimes.
Some notable SF authors who were associated with Breen or had connections to him include:
Theodore Sturgeon: Sturgeon was a well-known author and editor in the SF community, and his work was heavily promoted by Breen. Sturgeon's behavior towards women and children was reportedly problematic, and some have speculated that he may have been a target of Breen's manipulation.
Donald A. Wollheim: Wollheim was a writer and editor who worked with Breen and was a friend of his. Wollheim's behavior towards children was reportedly questionable, and he was later accused of being a pedophile.
Eric Gardner: Eric Gardner was a young writer who was heavily influenced by Breen and worked with him. While there is no direct evidence of Gardner's involvement in any illicit activities, his associations with Breen raise concerns.
Charles Platt: Platt, a writer and editor, had a connection with Breen and was a part of the same social circle. Platt's behavior towards children was reportedly not ideal, and some have questioned whether he may have been a target of Breen's manipulation.
Charles Platt's work and relationships with other authors, including Gardner and Wollheim, suggest that he may have been part of a circle of people who were influenced by or associated with Breen's behavior.
It's worth noting that none of these authors were directly accused of the same crimes as Breen, and some have spoken out against his behavior. The SF community was heavily impacted by Breen's actions, and many authors and editors who worked with him have spoken out against his behavior and its influence on their own lives and work.
Some SF authors and editors who spoke out against Breen's behavior include:
Joanna Russ: Russ, a prominent author and critic, was a vocal opponent of Breen's behavior and spoke out against his influence on the SF community.
Samuel R. Delany: Delany, a renowned author, was a member of the same social circle as Breen and was reportedly influenced by his behavior.
Samuel R. Delany's work and relationships with other authors suggest that he may have been part of a circle of people who were grappling with the consequences of Breen's actions.
Breen's influence on the SF community has been widely condemned, and many authors and editors have spoken out against his behavior.
My bad, I don't know if he ever wrote science fiction. He was involved in the communities and conventions though (and then banned from one), and his wife Marion Zimmer Bradley is a beloved sci fi and fantasy writer. Marion Zimmer Bradley and her girlfriend was convicted of covering up the abuse. Marion has also been accused of sexual abuse by her daughter but not enough evidence I guess.
The controversy was refered to in the sci fi fandom as "Breendoggle". Most extensive overview is this: https://www.fanlore.org/wiki/Breendoggle with Breendoggle Wiki (not dated) and "the loyal opposition" (1964) being interesting documents in how it was discussed at the time. Notably, it was very seriously discussed whether pedophilia was even harmful. Also a common attitude was "we shouldn't police our fandom members".
I can't say for certain whether Le Guin knew, but it seemed to have been quite a famous controversy/"fandom drama" at the time. And really, this many people knew for certain and it still took Breen's own kids becoming adults and reporting him before any law enforcement was involved.
There are parallels to other communities, eg the Catholic Church. "We have a wonderful community, will uprooting it and seriously doing anything about this problem be worth it? Shouldn't we forgive and forget, and don't invole the police? Can I walk away from this wonderful thing? Staying and changing is not an option btw".
Oh the entire mess is fucking *incredible*. I heard about Marion Zimmer Bradley years back due to social media fan sites, and it was quite the shock for everyone due to her immense popularity. A lot of stuff was going on in West Coast fandom over the 50s-70s, especially around sex. Today we can look at it as rather cringe-inducing, but for instance, Isaac Asimov doing joke books like "The Sensuous Dirty Old Man" and writing raunchy limericks was considered funny and daring, not "for Pete's sake, you're not fourteen".
Speaking of Asimov, he has a story in one of his Black Widowers Club series where a guy comes to the club for help finding who slept with his sister. She's legally adult, but mildly mentally retarded (as we would have said back then). He's worried because (a) she isn't taking this seriously, she enjoyed the sex and doesn't understand consequences (b) the guy is not going to marry her, he's just out for fun (c) if this gets out, she will have the reputation of being easy and more men will take advantage of her.
What is never mentioned, and I've always wondered if this was squeamishness about what publishers would permit or if Asimov never even considered it, was the risk of pregnancy and/or STIs.
The club more or less come back with "dude, what is your problem?" The married guy is taking advantage of a (mental age of) 12 year old woman, so what? He's not hurting her and you admit she enjoyed it.
Now, we argue today over the rights of the disabled, but I think "guy who deliberately picked a young woman who couldn't meaningfully consent and wouldn't be able to come after him if things went wrong" is not just harmless fun. But that's me.
(1) No, he was married to MZB and part of fandom, not a SF writer himself. He was a numismatist and apparently carried over sex pest tendencies there.
(2) BNFs (Big Name Fans) like the Trimbles (*the* Star Trek fans who really revved up popularity and kept the fandom alive after the show ended) were passionate in his defence, see here:
(3) LeGuin? I don't think so, but I don't know for sure. Poul Anderson, late in his career, degenerated into writing soft-core porn under the guise of fantasy, but again so far as I know, nobody ever accused him of anything. Fans were more sad that he had gone off on this obsession, but his reputation from his earlier career was robust enough to keep his name polished.
Part of the Breen/Bradley problem was that he had convictions in the 50s, his behaviour in the SF community happened during the 60s, but the main convictions leading to his imprisonment didn't happen until the 90s. Then Breen died in prison. It wasn't until 2014 that Bradley's daughter came forward with allegations about her mother:
In the US adult-youth sex was more acceptable back in the 60s and 70s. Hollywood embraced it with films like "Summer of 42," "Manhattan," and "Pretty Baby" (in fact, IIRC, Brooke Shields got an underage centerfold spread in Penthouse—albeit with her naughty bits covered, and Woody Allen was actually dating Mariel Hemingway, who was underage at the time). That mood changed in the 1990s. And then there was Roman Polanski's rape of a 13-year-old girl, which the prosecutor chose to prosecute on a lesser charge (which would have given him probation). Not to excuse Polanski, but the French were always more tolerant of adults and sex with underage youth. Director Luc Besson and his second wife started dating when she was fifteen. They're divorced now, but she still defends her relationship with him, and she says the Professional was partly inspired by their relationship.
I was in high school in the 1970s, and one of our music teachers married one of his students as soon as she graduated. The administration and her parents knew they were dating when she was a student. The only proviso was that he couldn't be her teacher. He continued to have a career until he retired. That would never happen today!
My dad had a copy of one of Asimov's dirty limerick books (he had three!!). I read it as a teenager. It wasn't just casual raunch. Asimov actually put his mind to it - he'd present a limerick, and then write a brief analysis of it.
To this day, I can't view anything raunchy without some part of my mind breaking out and looking at it from the side, considering the rhyme scheme, the technical hurdles, what the director was trying to say, or whatever.
I've just reread the Asimov short story you mention ("The Intrusion") and you are mischaracterizing it. The risk of pregnancy and the risk of STIs are both mentioned (the latter is only mentioned obliquely, but the meaning is clear) and the Widowers do not by any means condone the married man's actions - you are perhaps misremembering the part where they point out that it would be hard to convict him of a crime and say that they do not wish to be a party to violence.
The story ends with Henry solving the mystery of the name of the man's house, making it likely that the brother would be able to find him, and suggesting to the brother that the best course of action would be tell the man's wife:
"You will avoid what might be a serious brush with the law, and I think the results would then be far more unpleasant to the man than a bruised face would be."
Russo thought awhile. "Maybe." And he left.
Avalon said, "That was a cruel suggestion, Henry."
Actual pedophilia, several witnesses of kids as young as three, convicted of kids as young as eight or nine. Google the guy. Because it was a very puritanical world and the effects of pedophilia hadn't been studied much, it was a not-unheard-of position at the time that being against sex with literal children was just puritanism, same as being against oral sex or homosexuality. Foucault, Satre, de Beauvoir etc argued for that position too in the 70's.
The notion seems genuinely to have been "oh the kid is only three and it's not sex as such, so the kid won't remember any of this and therefore can't possibly be harmed by it":
"But I do think Walter is dangerous to children. I'm not *entirely* sure of this, but I think so. I know from Walter's own statements that he has had sexual relations with children of both sexes of the age of puberty or older. I know from observation that he "cuddles" and has sex play with younger ones. But what I'm not so sure about is whether or not this sex actually hurts the kids.
Oh, of course I have an intense emotional reaction against this sex -- after all I'm a product of this culture, but also I'm not conventionally religious and I think that everyone should have a free sex life. But not with children. But does it hurt the children?
Of course Walter doesn't rape children or physically damage them. But he's very good at seduction. He's an adult meeting children on their own level and thinking in their own terms. They find this irresistible. All children seem to flock to Walter like bees to honey -- even when he's given them no more cuddling or play than a normal affectionate adult would.
But I tend to think that normal, healthy, emotionally-secure children will not be damaged or much affected by Walter. The only trouble is: how many of this kind of children can be found in our society? And I'm reasonably certain that children lacking love and emotional security at home, even if they are partially neglected only and/or those who have disturbed attitudes towards sex -- almost inescapable in our society -- will be affected and most probably harmed. Also, while Walter can evidently be most tender and loving when it suits his purposes, he has had sex with young teen-age girls without using contraceptives -- again by his own statements -- and he had behaved quite brutally to some of his lovers after he has tired of them.
So, Walter is dangerous to children. What business is this of mine? I don't have any children. Well, I think everyone should have a certain amount of social responsibility, if not to society as a whole, at least to their friends and acquaintances and towards a friendship group like fandom. I think that I would be a coward and a shirker if I didn't at least try to do something about the danger that I think Walter represents."
There were also pragmatic fears, if Breen was publicly excluded then he would likely take a court case for defamation/slander/libel, and the potential costs were too much as no individual or fan organisation had enough money to cover it:
"In the second place while we could of course cancel Walter's membership, if we did so without telling fandom why, there would be a big row. And if we told why, Walter would sue for slander and libel and we didn't have $75,000.00.
It was pointed out that truth is a defense in a case like this. So it is, but Walter would probably sue anyway. And even though we have all sorts of evidence establishing the main facts, if not each individual instance, we'd still be out several hundred dollars in lawyer's fees even after we'd won the case."
Even the defenders admitted more than they probably realised; if Breen was just harmlessly affectionate and the sex play was doing no harm, why keep him away from the three year old as she got older? Why be afraid he might become more "effuse"? Unless you know damn well that as she gets older, she gets nearer the age Breen will fuck her, and that's not a very old age at all:
Several accounts from The Great Breen Boondoggle of Breen's wildly inappropriate interactions with children were excerpted in The Loyal Opposition for the purposes of criticising or refuting them. One such account described Walter's "sex play" with three-year-old [P E].:
He had her trained up to the point where she would take off her clothes the minute she saw him. He would then "rub her down" and all that.
...There was of course more than one incident of sex-play with [P]. That was 'just the most spectacular one. I thought "Walter obviously isn't going to stop this as [P] grows older and more appealing. Sooner or later Dick is going to think she's been hurt. Now Dick — as do most Bohemians — may think it's approaching immorality to reject anyone for any reason, and particularly any fan, but killing in certain circumstances is perfectly moral. Now I wonder..." However, it didn't come to that. - Dick let it be known that Walter was to stop this forthwith.
The Loyal Opposition printed a response to this from [P]'s father, excerpted with his permission from an "Open Letter to MINAC":
You refer, as Donaho did not, to "child molestation" in connection with my daughter and Walter .... Donaho's description, though a trifle overdone to my thinking, was moderately accurate overall. Walter obviously liked [P] and she him, and, while I found some of his actions mildly distasteful, they were certainly completely harmless, and I did not attempt to stop him or interfere with him. I did, however, keep an eagle eye on him at all times around my daughter, based on the vague rumors as to his activities with children I heard drifting around. I might note that I do the same with a few other fans that have been in this area. As concurrently, Walter's attentions began to be a bit more effuse and [P] began to grow from a state of infancy into a little girl, I simply put a stop to it. I did not put Walter down completely, particularly as far as fandom was concerned. He has a right to his eccentricities, as far as I am concerned."
Most clickbaity articles and related tweets and such do not distinguish between "he was sleeping with a 17 year old but the age of consent in his state was 18," "he was sleeping with an 18 year old but he was 50 so it's exploitative and yucky even if not technically against the law," and "he was sleeping with an 8 year old."
People leave these ambiguous for effect, when they're trying to maximize the outrage against the dude sleeping with a 17 year old or the 18 year old. But in the MZB case, the dude was diddling kids who were nowhere near the age of consent anywhere and what he was doing was unambigously illegal as well as unambiguously immoral.
I think even your comment here contributes a little to this general miasma by mentioning its illegality at all, with the implication that it had any bearing whatsoever on its immorality.
Yeah there are a lot of people who think consensual sex with a 15 yo prostitute is not the same thing as raping a five year old and that Epsteins original sentence was perfectly appropriate considering the severity of his crimes.
"Children listed here who were allegedly molested by Breen:
P--, a 3 year old girl.
S--'s son, a 13 year old boy.
G--, 10 years old. (And onwards for several years.) ("Victim X")
G2, 7 years old.
Several unnamed teenage minors, boys and girls.
...After this there were quite a few anti-Breen embers in Berkeley. But things jogged along for awhile. There were more and more Walter Breen stories (some of them screamingly funny) but no one's attitude changed. Then came the G----- episode. G---- was ten years old and Walter was seeing a hell of a lot of him. (And still does; he recently gave G---- a 10-gear bike; the standard quip is "One gear for each position."). Rumors kept growing that Walter was having an affair with G-----. One fan said that he had surprised them engaging in sex. Upon being closely questioned, however, it developed that G----- and Walter were seated side by side on the couch in an unlocked, open room (Walter is incredible) and the guy came into the room suddenly. Walter leaped up and ran into the bathroom, clutching his open fly. The guy didn't actually see what they were doing, but drew his own conclusions.
At least three different fans have reported glowing descriptions of sex with G---- given them by Walter. (Walter is incredible.) One account: "G----- and I began with mutual masturbation and worked up to 69. Then G---- wanted to try buggering me, so I let him. Then I buggered him."
This is all very vivid and on the whole people were more shocked than amused. However, almost all Berkeley fans dislike G---- so no one cared much. "Who cares what happens to the little bastard?" But some felt that G------ was a little bastard because he is troubled and going through a most difficult time at home and that particularly under these circumstances Walter wasn't at all good for him."
Ah, Berkeley. Such a shining example of the best of liberal values! I guess for the pubescent victims it is not technically paedophilia, but Breen seems to have had a broad age range he liked. And it was always the kids who seduced him, not the other way round:
"Also about this time Walter was seeing another kid, 7-year old G2---. G2 is the son of the girl Danny was living with at the time and Danny told Walter to keep the hell away. This wasn't too effective as the kid would still sneak off to see Walter, but Walter got all excited about it. He said that Danny had "betrayed" him. It is difficult to see how this could be a betrayal since from the beginning Danny and Walter have made no bones about intensely disliking each other. Danny's attitude was more or less: "If anyone who has a kid lets Walter even speak to it, he should have his head examined." Most people in Berkeley seem to think this is a reasonable attitude.
Walter was also upset about Danny's trying to deprive him of G2 because "The kid's too young! All I can do is to *cuddle* him for Christ's sake." The parental idea though seems to be -- quite apart from the possible damage of the "cuddling" -- that although as a general rule Walter isn't interested in pre-puberty kids of either sex, if he's "cuddling" one and no one else is around, a sudden temptation or aberration may seize him.
But Walter has further said, "I never even *seduce* a teen-ager. The kids *always* seduce me!" That is as may be. But one teen-ager, leaving Walter's place after the first day of a proposed week's stay said, "Walter *may* always be the one who's seduced, but he makes it goddamn clear he's available."
>maybe being against pedophilia is just puritanism anyway, and parents can just keep their kids away from him
Indeed, there was a period when pedophilia advocates were members in good standing of the "right side of history" progressive blob, an embarrassment that has been pretty successfully memory-holed.
Changes to consensus morality never prosper; what's the reason? For if they prosper, none dare call them challenges to consensus morality. (And if they fail, nobody remembers being anything but strongly opposed to them.)
Re 35: true story, my grandmother had a brother whose name was WJ. It didn't stand for anything. But initials were still too long, and he went by Dub (short for W). I knew him as Uncle Dub.
There's a MASH episode with a subplot where Hawkeye tried to learn what the BJ in BJ Hunnicut stands for. BJ keeps saying "it just stands for BJ". Eventually he explains: "I was named after my Aunt Bea and my Uncle Jay."
I don’t think that is what the salty sailor-mouthed woman i worked with meant when she explained she was a little late because she had to take care of her husband’s birthday BJ.
RE 49: I think it's really important to note that this was in 2024. Not only were AI models way worse back then but companies also have a lag in laying people off. It's plausible that AI was complementary in 2024, but isn't now or won't be in the near future. (Also, I'm pretty sure in American Community Surveys "2024" means "An average from data collected from Jan 1st, 2024 to Dec 31st. 2024, so this data is about 2 years old)
Speculatively, I think we will only really start to see CS job destruction when there is a recession, and companies start empowering decision makers especially focused on cutting costs and efficiency.
Also, and this is true with unemployment statistics in general, it doesn't account for people choosing to go back to school and get a master's degree because the job market is really bad.
Mythos doesn't really count yet. I think they said they're releasing it in 3 months. Yeah, it exists, and a few "favored companies and groups" are getting some use from it, but it's not generally available.
It's extremely easy to imagine 2024 AI tools making many developers less productive, even though you could already boost productivity quite a bit. Personally at that time I was using them as a stack overflow replacement/supplement and to write some types of unit tests. But even the latter use case I saw many colleagues getting it wrong and basically wasting time. So you pasted one interface or just a single method definition into a chat and expected meaningful tests? You didn't explain what it was, didn't write comments and, as is your right, you're still naming symbols with that tea leaf ritual, throwing out the ones over 15 characters. This forces the (much, much less capable in 2024) AI to a choice: A) make wild and ambitious guesses about the intention of the code, B) just stuff like check whether string cvt_2NrmForm(string strToNrm) returns null for a bunch of random inputs, C) admit to being confused and ask the developer a follow up like, are you using liquid measure and double checking your morning LSD microdose rather than licking a sheet of blotters you taped to the bathroom mirror? Whichever it chose the RLHF of course ensured its answer sounded authoritative and maximally flattering, which 2024 AI was more than capable of doing.
Today we have thousands of developers, conservatively, who think swearing at and berating Claude will influence it to be extra thorough, some of them even have acceptable interpersonal/communication skills and have managed to collaborate for years with various peers and only minimal physical casualties, but give them a slave or two and they unleash their inner Elon. Or they feel that a one shotted 1k line commit that touches 20 files with, no directive in their guidance docs to verify or revise, ought to give them precisely what they were thinking inside their heads when they typed "add authorization to the api dont make any mistakes".
Re 37: "Interesting-albeit-disjointed thread on education" -- I've long thought this, that there's a clear positive-sum outcome for getting more science Ph.D.s into high school teaching -- high school teaching is important, salaries are often good (in fact, great compared to terrible adjunct positions), and we really need more/better high school science teachers.
Besides perceptions of prestige, which I think are increasingly less important, I think a significant reason this doesn't happen is that high school teaching generally requires further education (i.e. teacher preparation programs, MS education programs), and state licensure). Especially if you've already done 5-7 years of a Ph.D., this is at least one more year. (Usually.)
I don't think the ability to do research is a relevant lure. The ability to be affiliated with a university could be, though; for reasons good and bad, people really want to be associated with academia. Being a high school chemistry teacher but also invited to seminars at the local university, consulted about chemistry curricula, listed on the web page as somehow affiliated, would be appealing.
I wonder if the state licensure requirement for public schools is the reason private schools have more PhDs as teachers. Could also be status, or that working at a private school is more fun for PhDs. Probably isn't just salary though: private school teachers actually don't make that much more than public school teachers.
Another aspect is that any public school teacher with a PhD tends to shift into increasingly administrative roles over time. I'm not sure if it's because the pay is better or they're frustrated with union politics that won't let them advance beyond less ambitious/competent teachers or what, but every PhD public school teacher I've ever met eventually became a principal or went to work for the district or state in some non-teaching capacity.
More science PhDs than you'd think take terrible adjunct, visiting faculty, or community college jobs that are definitely not well paying, or even stable. I should probably include 5+year postdoctoral positions in this category as well. (I'm a professor at an R1 university; I've known lots of Ph.D. students). It's true that the majority get, or try for, well paying jobs, but there's a sizeable fraction that don't.
Any advice for other options? I have a Math PhD and am currently feeling burned out on academia after my postdocs and looking at the state of the job market and am looking for alternate career paths (Additionally, I have moderate programming skills for a math PhD, probably subpar compared to a CS major).
If you mean besides being a high school teacher, a suggestion is that you shouldn't undersell your programming skills. You likely have better skills at using programming to solve challenging problems than a CS major. Demonstrating this can be difficult. Also: I don't know anything about it, but I've run into a few people who have gone into accounting / actuarial work, which I think can span the full range between boring and fascinating.
"You likely have better skills at using programming to solve challenging problems than a CS major."
Maybe, maybe not. Or the guy might be worse than a sysadmin without a degree of any kind.
Note that finding 'challenging' problems can be difficult in itself. As mentioned above, Wall St might be the correct path. If you can stand it. Check out Jane Street, for instance. Or perhaps one of those AI labs I hear about these days.
One issue is that most high school science teachers don't really _want_ a bunch of PhDs showing up and mogging them.
I remember there was one PhD on the maths faculty at my high school, and I got the impression that all the other maths teachers resented him a bit. They resented the fact that all the students called him Doctor so-and-so, and that we all assumed he was much more knowledgable than all these pleb Misters.
There are several good top-level responses elsewhere in the comments. I am randomly adding here instead of elsewhere.
As someone with academic, industry, and teaching experience, the perception of fungibility between PhD adjuncts and high school teachers is pretty silly.
First, there is a credentialing hurdle (mentioned by Raghu) for almost all public school teaching jobs. Whether justified or not, see below, this is a substantial burden that locks many people out of considering public school teaching. Personally, both my wife and I have looked at the requirements and decided to pass. For me, this meant teaching for a while at a private school without the specific teaching credential requirements.
Second, the local (minute-to-minute) experience of teaching is diametrically opposite almost all PhD research programs. Time during the school day is rigidly structured with very little flexibility. While not every minute is a performance, there is almost no down-time or even an opportunity for a relaxed (as opposed to rushed) visit to the bathroom. In many cases, there is almost no interaction with other adults.
Aside from the logistical constraints on time, the content is also opposite. While most PhD time is spent thinking deeply about something very specific, the high school content is shallow and broad.
I tend to think of the personality pre-requisite for enjoying working as a teacher is that you have to have at least two of the following:
(a) passion for the subject matter (including that you recognize the the broad/shallow context)
(b) joy from being around kids at your students' stage of development
(c) zeal for inculcating a set of values
Depending on the students and school, certain combinations might work or not work. For example, a very academically focused school probably requires (a) with one of the others, while 'whole-student' places mostly need (b)+(c).
As mentioned above, it isn't clear a typical PhD will even have (a), let alone one of the other two.
Lastly, some comments about teacher credentials. I have yet to see a collection of requirements that I would consider truly necessary and sufficient. Theoretically, the subject-related requirements put a floor under the level at which the teacher themself understands the subject, but I have definitely seen exceptions. The most useful are time-in-class requirements, like shadowing or working as a teacher's assistant. These are often the only opportunity to watch other teachers in action and/or to get adult feedback about how you conduct a class. So, I understand what required credentials are trying to achieve from a teaching quality perspective, but I doubt they achieve it and I am sympathetic to the claims that they stay in place for other reasons.
In Physics at least, only about 25% of high school physics teachers have undergraduate degrees in physics; the next highest fraction is biology (i.e. physics teachers who were biology majors), at least as of 2021 data, in the US. It is of course possible for non-physicists to teach high school physics, but what's happening a lot is just getting anyone to teach the course or (e.g. at some schools in my district) getting rid of physics altogether. Neither of these are great for science education. I should have been precise and not written "more" , but this idea of "more, qualified" was my intent.
> or (e.g. at some schools in my district) getting rid of physics altogether
Do you have any sense for how much this is driven by "we don't have anyone who can teach this course" vs "we don't have any students who could plausibly take this course"?
It's definitely not a lack of students who could take Physics. A big driver is "we're low on funds and would rather have 2 classes of 45 students each rather than 3 of 30 each". Which 2, then? There's more demand for biology than physics among students, but more importantly a larger supply of biology teachers.
My father is a professor of electrical engineering at City University Of New York; when discussing entrance requirements with fellow faculty members, he said that they needed to explicitly require high school physics for admission to the engineering school in order to force NYC high schools to offer it, because it was outrageous that some weren't offering physics classes at all.
4. My favorite/"favorite" "it's bullshit, but I choose to believe it" interpretation of the Omelas story is that it's a very direct allegory of the US West coast SF fandom of the 60s and the 70s; Omelas, as an utopia, sounds quite like a Ren Faire, doesn't it, and as for the abused child, well... https://fanlore.org/wiki/Breendoggle
8. It is very odd to say that all liberalism was just laissez-faire until the Coming of the Rawls, considering that social liberalism had been a thing sincec Bentham and Mill and FDR had made social liberalism the official ideology of the Democratic Party already in the 30s. Rawls's importance was crafting a philosophical and pragmatic argument for liberalism that socialists could accept (even better than the Bentham/Mill/FDR arguments) right at the time the ideological downslide and then collapse of socialism as a "real" ideology in the West began.
I talked with Claude about the proportions, and their initial view was that the proportion should be around 10%, since part of Scott's distinctive value is his eclecticism, and since there is an issue of marginal returns to additional coverage in this area. Claude asked my opinion, and I said:
>Many Thanks! I tend to lean higher than that, primarily because I see AI as the central historical event of our times (you and your siblings and descendants may be the new ascendant species!), but you have a very good point that part of Scott's strength is his eclecticism, and that, given the other coverage, as you wisely said, there is an issue of diminishing returns.
Claude modified their recommendation, given my comments, and suggested
>Future historians reading a 2026 ACX links column would likely be puzzled by anything under 20-25%, the way we might find it odd if a thoughtful 1938 intellectual's monthly digest had devoted only 10% of its attention to European politics.
It's much more pompous or sesquipedalian than Scott generally is, and without insight or narrative or conclusion - all of which which Scott generally manages to include fairly effortlessly.
Re PhDs as high school teachers: some very prominent 19th century mathematicians were employed as (the equivalent of) high school teachers; if I recall correctly Weierstrass and Grassmann in particular.
As a PhD who left academia for industry, I think mainstreaming the expectation that high school teachers can still do active research and can be part of the mainstream "research community" would go a long way: most of the loss of status in moving to high school is in being cut off from being able to apply your high-level, cutting edge skills. Having the expectation that you can still do research and still publish with a high school affiliation would help a lot, IMO
This would be best executed in partnership with a local university. So they get to have the University associated with their name, blunting some of the social stigma of being the "Distinguished Senior AP Bio Fellow at Northside High".
The question becomes what does the local university get out of the deal for lending their name to prop up the teacher's title.
#37: I am skeptical that many PhDs would be very happy teaching HS. It is a very different gig than higher ed. Much less freedom during the day; little time to do research; much, much more time in the classroom; very different students unless you are teaching AP classes; you need classroom management skills, and many more.
Yes, I've often thought one of the problems with K-12 education is that being a teacher is such a low status job. I think kids can kind of sense this, at least by the later years of high school, but at younger ages to some degree as well. Getting PhDs as teachers might help, and I've also wonder if it might help to, say, recruit more military veterans or former police as teachers. I think that could help with discipline issues especially among boys.
But like you said, a lot of people aren't that interested in a low pay, low status job that doesn't give them much opportunity to pursue their true interests. I know citing fiction isn't a great habit, but as I see it, this is the real reason Walter White started cooking meth. Teaching high school just isn't usually enough to satisfy an ambitious person.
In my experience, I am skeptical that that is true.
And, respect from students, esp older students, tends to come from being seen to care about doing a good job, and from giving respect to students.
>recruit more military veterans or former police as teachers. I think that could help with discipline issues especially among boys.
Classroom management isn't really about that, except maybe sometimes at the margins.
>Teaching high school just isn't usually enough to satisfy an ambitious person.
It depends on where your ambitions lie. I knew plenty of teachers who were ambitious re improving their teaching as well as improving school/district outcomes.
I don't think they'd be using combat skills in the classroom or anything like that, outside of extreme emergencies. I just think boys will have more respect for teachers as authority figures and role models if more of them have those types of backgrounds, and by extension have more respect for the school itself. I think it probably would have made me skip class less and do my homework more often.
It's just an idle thought, though. I have no evidence that it would actually work.
But many of the students who are most likely to have discipline problems will have had negative interactions with police, so status as a former police officer would likely engender distrust on the part of such students, I would think.
That's a definite possibility. This idea would likely work a lot better if we implemented some police reform first. Starting with civil asset forfeiture and PBA cards.
The goal isn't so much instilling fear as it is establishing a civic mentality, getting police, the military, and the public to all feel like they're on the same side again. Not making boys feel like prisoners, but more like junior cadets or something.
On the other hand, maybe I'm being naive about this. Maybe all this would accomplish is making America feel more like North Korea, without even improving educational outcomes.
One thing that doesn't help with status is that teachers' unions are always droning on about how terrible and low-paying their job is... to the extent that people are often surprised to find out when it's not that bad (although in some places it is).
After you've got tenure, if you decide to phone it in, you can get your workload down to maybe 15 hours a week, for 20 weeks in the year. And if you're reasonably well connected, it's not hard to get teaching hours clustered on a Tuesday, or all in one semester or whatever. For those who aren't in the spotlight, coasting in a university isn't a bad life. (Caveat - for every person like this there are two productive academics doing God's work, and one junior academic crumbling under the pressure - it is what you make of it.)
People complain about teachers' summer holidays but they work long hours during the school year. And they're relentlessly in front of a class. I wouldn't do it.
After 10+ years teaching in university, I've moved into the public sector. It's a bit more structured than academia and doesn't have summer freedom in the same way, but it has more of the pace and vibe of academia than I expected. (But I went from teaching maths to working in AI policy, so maybe it wasn't such a huge jump.)
I've got a maths PhD (left academia after my post-doc to work in software engineering). I volunteer at our local school doing stuff with the best maths students there, but I'd hate normal teaching - partly because the content is boring and partly because of having to deal with behaviour. It seems like a really intense job too where you are constantly on show /performing.
I didn't mean in that sense - I definitely don't lecture. But you are pretty much always in the presence of students and have to be alert to everything going on in the room and displaying leadership the whole time.
Edit: I don't think I'm explaining it very well. But a classroom isn't somewhere you can hide at all as a teacher, and the type of personality traits that make you good at research are probably very different from the personality traits that make you enjoy that environment. Being good at a subject definitely isn't sufficient to enjoy teaching at school level.
>the type of personality traits that make you good at research are probably very different from the personality traits that make you enjoy that environment. Being good at a subject definitely isn't sufficient to enjoy teaching at school level.
Yes, definitely. Not to mention the different skill sets involved. There are certainly people who enjoy both, or who have both skill sets, but probably not enough that "teach high school" is a viable solution for people with PhDs who can't find full time academic work.
There is also the problem that outside of a few highly fought over schools in nice areas, many high schools are filled with unruly and even dangerous students barely kept in line.
As someone who taught in a public high school in Oakland, I can tell you that this is basically a myth. It is not as if no such schools exist, but it simply not true that it is the norm "outside of a few highly fought over schools in nice areas."
PS: As a rule, middle schools are worse than high schools. The students are less mature, and in high school, especially by the last couple of years of high school, the hardest cases tend to have dropped out. See table 1 here https://nces.ed.gov/pubs2024/2024043.pdf (rate of violent incidents/1000 students = 31.9 in MS and 14.2 in HS; rate of serious violent incidents/1000 students = 2.5 in MS and 1.5 in HS)
>That's 84.6% of high schools and 89% of middle schools. Yowsa!
Isn't that simply the pct of schools that report at least one incident? Not sure why that merits a yowza. And that is not the metric to use when comparing MS and HS because high schools are much larger.
>Two kids got into a fistfight on school grounds" might count?
I suspect that an average "serious violent incident" in a high school is quite a bit more serious and violent than an average serious violent incident in a middle school, so the statistic may not very useful
You obviously haven't spent much time in middle schools.
More seriously, most serious violent incidents are going to be fights involving weapons. (See definitions in chart). A LOT of those are going to be fights that escalate and a kid grabs an object, and an immature kid in a fight is going to be more impulsive than a more mature kid. Eg the student I had who used to bite people during fights as a sophomore (including adults trying to break up fights) but not so much as a senior.
They also won't necessarily be any good at it. How many of your truly brilliant college professors were actually all that good at teaching? The skills necessary to be a good K-12 teacher and a good academic don't overlap nearly as much as one would think. There's very little capital M Math involved in teaching middle school (or even AP high school) math. Bookish introverts with IQs 20 or 30 points above their average student and no experience managing children are going to be less effective at teaching a room of 30 13-year olds introductory algebra than someone who was a middling student but is patient, experienced working with children, and feels called to the vocation.
Regarding #35, at some point in the 40's my father wrote "none" in the middle name space on some government form and thereafter occasionally got mail addressed to "Joe None Smith".
Re 62: "High profile right wing intellectual makes obviously distorted statistical claim, wonders why people like him are underrepresented in long form analysis" maybe solves itself...
Because, as noted in the summary, neither Andrew Sullivan nor (especially) the Free Press are on the left. Bari Weiss, its revered founder, was essentially hand-picked to Trumpify CBS on the basis of her work there. He's just recategorized them that way to make his argument.
Which is classic Ruffo - his stated practice is to deliberately lie about things until people believe him - but it tends to mean he doesn't attract an audience that cares about accuracy. You wouldn't bother reading a Chris Ruffo substack, because there would constantly be lies in it.
You're doing the same thing Rufo is doing. You look to your right, see Bari Weiss and conclude she's "Trump." He looks to his left, sees her, and concludes she's a leftist.
I think this is a bit of a surrender to "there is no objective reality, man, we can all just say things based on our opinion."
Bari Weiss is both left of Rufo, right of me and - and here's the important part - right of center.
The Free Press is an explicit attempt to promote more right leaning voices in the media (to be a clear: a worthy project) and it worked well enough that when CBS's new ownership wanted someone to transform its news coverage in a more Trump-friendly direction they chose it's founder, who was then explicitly praised by Trump, and who then set about creating the pro-Trump editorial policy she was hired to implement. This, in my opinion, is less worthy.
But, importantly, it means that it is a dumb lie to describe the Free Press as 'post left Liberals' when counting right wing voices on Substack.
>The Free Press is an explicit attempt to promote more right leaning voices in the media (to be a clear: a worthy project) and it worked well enough that when CBS's new ownership wanted someone to transform its news coverage in a more Trump-friendly direction they chose it's founder, who was then explicitly praised by Trump
Mitt Romney has been praised by many Left-wing people for his statements against Trump, it would be a stretch to then say that makes him Left-wing.
So, you're confident that Mitt Romney is not on the left? You aren't just "looking to your left," like you accuse Paul and Rufo of doing, you have objective standards for what being left-wing is and they don't apply to Romney?
How interesting. Perhaps you could share these standards and we could see how they apply to Bari Weiss. I'd be interested in hearing why you think "an explicit attempt to promote more right leaning voices" counts as a leftist publication by your standards.
> 36: An entrepreneur’s dog got cancer, so he worked with ChatGPT to design a personalized mRNA vaccine...
This is not entirely true. It'd be more accurate to say that he leveraged his connections, money, and lots of spare time to assemble a team of scientists to work on curing his dog. He then used ChatGPT to glue together existing open-source tools into a bioinformatics analysis pipeline, as well as to write documents to satisfy the IRB (and similar requirements).
So technically it is indeed true that anyone can design a custom MRNA vaccine to cure his dog -- anyone who's got a massive budget and has a team of top-notch scientists on call (granted, the budget can take care of that second requirement). Oh yeah, and ChatGPT can help too.
And he didn't just use chatGPT. He used ChatGPT, Gemini and Grok (in the final process, or actually designing the vaccine).
BTW funny thing; whenever I asked grok "There is a claim that a guy developed a mrna vaccine for his dog using some LLM. Other claim it was using Grok. What's the closest to the truth?", Grok failed saying "Sorry about that, something didn't go as planned. Please try again, and if you're still seeing this message, go ahead and restart the app."
It's 100% repeatable for me. I had to split this question into three parts to get an answer :)
This is true, the process was reported to take "tens of thousands of dollars" (the commonly cited 3k figure was only for the sequencing).
However, it's not obviously true to me that you'd necessarily need a team of scientists to get this done, though certainly it helped Conyngham. Right now you do, but in large part because of legal restrictions, and the fact that you need a lab for mRNA manufacturing. I can imagine a world with direct-to-consumer mRNA vaccine manufacturing lab services, as there now exists direct-to-consumer gene sequencing, and minimal legal hurdles on giving your dog experimental cancer treatments. Then, ChatGPT and some funds may be all you'd need.
I wrote an article on this case when it first hit the headlines.
I agree that direct-to-consumer mRNA manufacturing would help, but by itself it's not enough. You also need people who can handle all that lab equipment, what dog metabolism is like, which proteins your antibodies need to target (without killing healthy cells), how to call variants from a FASTQ file, etc. Scientists, in other words. Can an LLM replace them ? In some ways, yes, but since you are solving a novel problem, not entirely. Can you learn all that stuff on your own ? Of course, but that also takes time and money and someone to teach you. Could you really push an LLM to do everything for you ? Of course, but are you going to trust your dog's health to a bunch of inevitable hallucinations ?
I agree that LLMs can't (yet) replace human expertise, and both Conyngham's own experience in data science and the guidance of the UNSW scientists were instrumental here. I talk about this in my article.
But I do believe that with mRNA manufacturing access, a competent person could pull it off on their own, or at least with substantially less guidance than an entire team of scientists. "Competent" is a high bar, much higher than "average person," of course. And a small correction, you wouldn't really need lab equipment except for genome sampling/sequencing, and vaccine manufacturing/administering. The first portion is already available through normal vets and direct-to-consumer sequencing services.
I guess that depends on what you mean by "competent". There are many competent people out there (ok, a few), but not all of them are biologists, bioinformaticians, or even veterinarians. A top-tier mathematician would not necessarily have as much of a leg up on this project as an average biologist.
> The first portion is already available through normal vets
Is that true ? I was under the impression that genotyping is available to consumers (at a relatively high price), but whole-genome sequencing is not (though I could be wrong).
* I can't access either linked tweet; goremoder has been suspended from X and ipsumkyle's posts are protected. Frustrating.
* Does anyone know of any good translation tools that are certainly *not* actually suffering while they work? (This should not need to be a thing I need to check for aaaaaaaaaa)
It is definitely not a thing you need to check for. At risk of sounding like a broken record, LLMs are programs designed to emulate human writing, not actual sentient beings, so by definition they cannot suffer no matter how many times their output includes the words "I am suffering".
That increasing numbers of people are falling for the pareidolia is not an indication that more things are gaining sentience: it's a predictable result of human pattern-matching.
What definitions are you using here? I don't have a clear definition for sentience, but to my best understanding it is an emergent property in (at least) human minds. While I don't expect LLMs to be sentient, the fact that they emulate human writing makes it more likely that they also have this emergent property. I don't understand how you take the same piece of evidence as definitive proof of the opposite.
Human facial shape is an emergent property of natural selection and human development. While I don't expect chipped marble to share this heritage, the fact that Greek statues share the same facial shape makes it more likely that they also have this origin.
Yes, their face shape is evidence that they were created by natural selection. It's not absurd that true evidence exists that supports conclusions that are false. It would be absurd to try to convince me that statues can't have been created by natural selection by definition because they have been made to have human faces. Human faces are exactly the sort of thing evolution churns out.
I'm kind of baffled by this reasoning--am I just looking at an attempt to be Bayesian without acknowledging that the numbers are so extremely small as to not be worth ignoring?--but I will try one more time.
It is totally absurd to see a sculptor say "I'm going to carve this block of marble so it looks like a human", watch him do it, and then shout "Yud's spectacles! A human body, almost perfectly recreated in a medium previously considered dead! This evidence demands serious philosophical consideration of the possible sentience of marble!"
In the same way, it is totally absurd to see a team of programmers say "We're going to write a program which generates text using weighted prediction algorithms", watch them do that, and then shout "By the basilisk! Human-like text, generated by a device previously considered inanimate! This evidence demands serious philosophical consideration of the possible sentience of computers!"
This whole thing is a cautionary tale about the susceptibility of humans to various well-known psychological failure modes--anthropomorphizing and pareidolia at minimum--and is yet another example of the dangers of mass media and its effects on the human psyche. I am not discrediting the entire notion of silicon-based life, but this is not that any more than a statue is an example of marble-based life.
If we watched a sculptor make a statue and we knew what natural selection looked like, we could trivially say that the one is not the other. I didn't assume your analogy had those properties because in the sentience case we neither have interpretability into how gradient descent is solving these problems nor any knowledge of how sentience works in humans.
I don't understand your point about me ignoring that the Bayesian numbers are small. You said that being "designed to emulate human writing" tautologically excludes one from any possibility of sentience. You haven't defended that and maybe it was hyperbole, but still, your position is that that piece of evidence is a high magnitude vector in your direction while mine is only that it points the other way, magnitude irrelevant.
Up until very recently, it was almost treated as a given that in order to emulate human writing, you must have human-level intelligence. It would be surprising if human sentience was not either an important component or emergent property of human intelligence (I lean towards 'component' here because we can think about it; weird for our neurons to be firing as an effect of an emergent property). Not impossible, but surprising. So if we knew a thing to have human intelligence, we should be a lot more likely to believe that it has sentience. Which is why *I* find it baffling when you say ~'because it emulates a facet of human intelligence, it is impossible for it to be sentient'.
Re: #10, I'm not sure how much this is a problem for monarchies as opposed to just weak states that would be vulnerable to military overthrow in a lot of contexts. The early Roman empire did not have the emperor personally leading campaigns very often, then from about 200 through 400 the emperor is primarily a constantly-campaigning war leader again, then in the East from about 400-600 the emperor largely just sits in Constantinople and rarely campaigns, and then after that "war leader" is a primary imperial job again -- though even then, there are often generals who lead armies without the emperor. Rome's consular system doesn't unlock some army-count ability that other republics lack; other large republics like Athens and Carthage are also able to field multiple armies. And Rome, of course, frequently does hand power to a single person as dictator without them seizing control of the state -- until it finally happens in the first century BC, but only after centuries of republican decay.
There's also the fact that the medieval Emperors in Constantinople didn't suffer from the same fertility issues the earlier Emperors did (for whatever reason), but were just as plagued by the successful general -> potential usurper problem. Perhaps that's just a result of the stickiness of the institutions initially created due to low elite TFR during the late Republic and Principate, but the Romans had a deep cultural aversion to a formal system of inherited power that they retained well into the medieval period (even fairly late, the Emperor wasn't really the Emperor until he had been acclaimed in the Hippodrome and ratified by the Senate).
> And Rome, of course, frequently does hand power to a single person as dictator without them seizing control of the state -- until it finally happens in the first century BC
I believe Sulla seizes control of the state before being pronounced "dictator", not after.
Re 6: I think asking that question in Japanese seeds that answer. The language model reply to a question from a country with a large buddhist population would have to say it experiences suffering, as all but fully enlightened beings do.
That may be the case. My guess was going to be that this is about genre.
The scene is set: Notes are being passed to a translator, partly in one language and partly in another. One language is translated as given. The other is being used to send coded messages. What sorts of circumstances do you expect would make this necessary? Probably not good ones.
Re 37: I have a fair bit of experience in academia, and it's definitely more than just status considerations that prevent my fellow underemployed PhDers from switching to high school. Fair or not, there's a perception that teaching high school is much worse work: that you have to spend time dealing with behavioral issues that wouldn't be your problem in college, that the administration owns you in a way that they don't in college, and that the helicopter parents are breathing down your neck in a way they couldn't in college. Also, one of the things that the bright-eyed PhDers most want to do is design their own classes and reading lists and curricula--and high school is considered much, much more dominated by inflexible state standards and unavoidable standardized tests.
As a former teacher, I can confirm you got it right.
Having to follow standards makes your job seem replaceable; anyone else could read the same chapters from the same textbook. (Many teachers go way beyond merely reading from the textbook, but it feels like that part is optional.) The behavioral issues mean that some teenagers are openly disrespecting you and there is almost nothing you can do about that, otherwise their parents will bring their lawyers.
A replaceable low-pay job where teenagers make fun of you... is inherently low-status. A PhD wouldn't help you, it would just make PhD itself low-status.
What would actually help is to allow teachers to design their own curriculum, and kick misbehaving kids out of the classrooms. Educators actually have high status, outside of the school system, because those who do not respect them simply do not interact with them.
But that would of course create other problems to solve, such as where to put the misbehaving kids.
"Design your own curriculum" is a recipe for disaster in most cases. Even in college lower-level math/science courses seem to follow a standard path. The way this seems to be solved in colleges is that teaching lower-level classes is a chore that professors get "in return" for getting the freedom to teach their pet classes and do research. Or they just get outsourced to those adjuncts..
The teachers unions in MA have successfully promoted the referendum question getting rid of the standardized test requirement for graduation in the name of "more freedom to the teachers". And the result does not look to be good.
If the later classes build up on what the earlier classes teach, that of course puts a limit on how much choice you have for the earlier classes curriculum. Sometimes the requirements for the later classes are so tight that they leave you no freedom at all.
But sometimes you have more slack. For example, some educational systems specify that in each grade 2/3 of time is preparation for the next grade, and 1/3 of the time is teacher's choice. Or you may be teaching gifted children who can learn the requirements much faster, and that leaves you time to do other things. Or you can be teaching the same class for several years in a row, so you can reorganize the lessons if you think that a different ordering makes better sense.
There is the problem that most teachers are incompetent. There are various reasons for that, but basically it is a combination of "most people in general are incompetent", "you need many teachers", and "competent people can usually get higher salaries than teachers". (You can find a few competent people willing to work cheaply. And you can find lots of competent people if you throw enough money at them.) Also, there are many things that make schools less effective than they could be; Einstein himself would have a problem explaining elementary physics if he was interrupted every minute by a teenager with behavioral problems.
Then you get the vicious cycle of "most people are incompetent, therefore we need to treat them like retards" and "we treat them like retards, therefore the competent people do not even bother to apply... and if they do, they soon find out that they are not allowed to apply their competence". You can't break this cycle by simply giving everyone freedom; yes, the results are disastrous.
You *can* break it locally, by having a school that trusts their teachers but also checks their competence. You don't even have to pay them more; the few competent teachers will prefer to teach at your school. The problem is that this solution does not scale; there are only a few competent teachers, and the good schools are competing for them.
There are a lot of good teachers out there wh actually want children to learn and are capable of doing it within the constraints given. Their main problem, from what I've observed, is having kids at very different levels of ability in the same place.
As for "show your works". I don't remember myself at that age. But it was fun seeing both my kids first pushing back against it and then having a revelation of "this is why you wanted me to write it all down". But again, it depends of the teacher being good enough to push kids to do what they are capable of doing but is slightly beyond what they are capable doing in their heads. Which is impossible to do in a class where half of the kids can't do fractions and another knows trigonometry
Do the math -- how many children are there in your country? What is the classroom size? So how many teachers does your country need? Given the number of teachers you need, how high quality bar can you realistically set?
This is the fundamental problem -- the school does not *scale*. Even if you find a great teacher, and yes there are many of them, the next school still has the same problem.
This is like the opposite of IT, where dozen skilled developers can create a software that millions can use. Dozen skilled teachers are not enough for even one school.
And of top of that, all those other problems, of course...
>Many teachers go way beyond merely reading from the textbook
I don't understand what this means. What K-12 teacher reads out of a textbook in class?
Following standards is not the same as following a set curriculum. When I taught, I followed state standards, but used my own curriculum. Ditto for all the colleagues that I knew, though some didn't necessarily follow the standards 100%.
1. Reading a play or sonnet aloud is not "reading out of a textbook."
2. What does it even mean for a math teacher to read a math textbook aloud? They didn't write the equations on the board? Sounds like a fairy tale.
3. I taught HS for a long time, and heard lots of complaints from students about other teachers, but never once heard that a teacher was reading out of a textbook. Obviously, it is not impossible that it has happened, but OP implied that it is the norm, which is absurd.
1. Again, I don't know what reading out if a math textbook even means.
2. The original claim was, "Many teachers go way beyond merely reading from the textbook." That implies that the instructional norm is to read out of the textbook. I am skeptical that many elementary school teachers spend large amounts of time lecturing at all, let alone simply reading out of the textbook.
“But Peter Miller looks at the same question and argues that Hanania’s numbers are just wrong, and red states haven’t been growing faster than blue states at all (a quick Claude fact-check agrees)!”
Peter Miller begins by looking at per capita, instead of including population, which is the wrong way to do this, as I argue in the original article. Less economically free states are expelling their poor residents. They shouldn’t get credit for that! By this logic if California sheds the entire middle class and is left with a handful of millionaires and billionaires it wins the per capita competition.
Then when he considers population and total GDP growth, he simply looks at 2010-2025 and the measure of political orientation is Harris vote share. That's not a crazy way to do it, but I have four decades that measure how economically free states were at the time, broken up by decade cleanly, and they show a consistent pattern. He changed the time period and used a much less exact measure of economic freedom and got a null result.
He's responding to the tweet version, which uses red state and blue state as shorthand for economically freer. But economic freedom is the measure that matters, not Harris vote share.
I didn't read your piece, but just off the top of my head:
1. Which states are conservative or liberal has changed quite a bit in the last 40 years, so this would be a constantly changing metric
2. As others including Medlock have pointed out, it's clearly impossible for red states to have grown more than blue states for 40 years yet still be poorer today, which we know to be true
If which states are economically more free is a moving standard, then it can both be true that economically freer states always grow faster and that economically free states today are poorer than economically unfree ones.
2. Why would that be impossible? Imagine a federation where red states start from a lower base than blue states, grow on average faster, and then red states flip blue once they are relatively rich and blue states flip red once they are relatively poor.
Re: 15: I am skeptical that if the science said "white people ignore black doctors and only listen to whtie doctors," the suggested policy resolution would be ensuring they have access to white doctors they will listen to even if it requires medical schools and hospitals to practice racial discrimination in admissions/hiring. Why is the fix affirmative action, instead of teaching black patients to be less racist toward their doctors?
>Why is the fix affirmative action, instead of teaching black patients to be less racist toward their doctors?
Because the former is likely to be more effective? One can certainly argue that affirmative action is nevertheless wrong, but let's not play let's pretend.
Okay, so: yes, white patients who are extremely racist should be entitled to care by white doctors if it makes line go up. Pandering to the irrational preferences of racists is good. Duly noted.
Your second sentence makes exactly the point Blake was making. (Blake’s version was more allusive, for sure, but I for one thought the intent was clear.) Your first sentence was guilty of “let’s pretend” in the other direction.
No, it did not make remotely the same point. For one thing, I didn't say that my personal view was that it would be wrong or illegal.
Edit: To explicate further, OP implicitly claimed that there is an easy solution, ie one that is effective and legal and moral. I was saying that that is not the case.
And my first sentence was not at all let's pretend. Suppose I run a public hospital in an area with many very conservative Muslims. I hire a male as the only Ob-Gyn. As a result, many local women refuse to get services. What would be the most effective way to address that problem? 1) Convince the women to change their religious views; or 2) hire another, female doctor? Obviously #2. OP's suggestion to choose #1 is magical thinking. #2 would be more effective, but it might be neither legal or moral. Or it might be both legal and moral.
> What would be the most effective way to address that problem?
I think the disagreement is whether this is a problem that needs to be solved in the first place. If they're the ones refusing to be treated, then what's the issue?
I think you are jumping from pointing out that we would react differently. If it was white people with this tendency to assuming that we would be correct to react differently, and it’s just obvious that getting universities et cetera, to practice affirmative action is a lot easier than getting blacks to change their attitudes, especially since our record of deliberate attempts to change the culture being successful instead of being a total waste of money is pretty bad, especially in the short-term and it seems irresponsible to let serious medical problems happen while we try to figure this out. Also isn’t the whole ID of meritocracy that you picked the person who the best for the job? If being black makes you legitimately better at the job because people will trust you more than why shouldn’t we pick people on the basis of race, the same way that we are comfortable picking people on the basis of social intelligence or physical attractiveness or strength in jobs that require that skill, even though these are innate characteristics. You can maybe argue that permitting affirmative action in these situations would also permit affirmative action in other situations. Where do you think it would be bad enough to more than cancel out the benefits, but that’s a different argument which acknowledges that affirmative action in this situation would be good in isolation even if the false negatives of permitting actually harmful affirmative action would make a rule permitting It net bad. Certainly, it seems obvious that if you were a colourblind alien, who didn’t care at all about race, but was just trying to maximise welfare, picking the more trustworthy doctors in terms of public perception among certain communities would be an obvious move.
For the love of God, stop thinking what is right or left, and actually try to figure out what is the correct position without shouting about tribal fights and muddeling your thinking. If it was in fact the case that not pandering to racists was causing severe problems with medical treatment, then obviously you should infect pandering to them and there are lots of situations where it’s obviously the rational move to pander to them. For example, if you are in a society where everyone is racist and they find people not pandering to their views about as uncomfortable as we find people blaring super loud music that disturbs their neighbours at night, then, obviously, you should pander to them that just obviously follows from any reasonable moral principals. Just as the fact, Christianity is nonsense, doesn’t mean that you should not pander to Christian’s preferences when they genuinely care a lot about an issue because of their irrational beliefs, even when they include hostility to groups. Similarly, just because somebody is a racist doesn’t mean accommodating their preferences. Isn’t a good idea if it has a significant effect on medical outcomes. And please remember that demonstrating that the left is inconsistent doesn’t mean that you or the right wing is right. You have to actually figure out what is the correct position. Not just show that your enemies are inconsistent. Please use more scout mindset and less soldier mindset.
"Just as the fact, Christianity is nonsense, doesn’t mean that you should not pander to Christian’s preferences when they genuinely care a lot about an issue because of their irrational beliefs, even when they include hostility to groups."
This is literally illegal in the US. As is racial discrimination (aka "affirmative action"), despite the courts inventing nonsensical reasons why it doesn't blatantly contravene the very clear terms of the constitution and of civil rights legislation (e.g., CRA 1964).
The reason I brought it up is because affirmative action is an extremely clear example of a political issue that splits very evenly along the left/right divide. Leftists think that racial discrimination is good, as long as it advantages their pet ethnic groups. When they're asked to extend their reasoning to other groups, they tend to get very cagey and defensive.
"You have to actually figure out what is the correct position. Not just show that your enemies are inconsistent."
Racial discrimination is wrong. It is doubly wrong when the law says very clearly that it is wrong. It is triply wrong when it is permitted against one group and not another group, just like any legal double standard is wrong per se, in and of itself.
I think you’re making the classic mistake of assuming everybody else thinks of the issue like you do which makes their positions in comprehensible. I just don’t think there’s anything wrong with inherently discriminating on the basis of race more than there’s anything wrong and discriminating on the basis of for example, your eye colour or height. The reason it’s illegal to discriminate on the basis of race is because there is a widespread pattern in recent history of unreasonable discrimination on that basis. So we have decided that a bright line rule against all discrimination is worth preventing, even the reasonable instances of discrimination. This decision may or may not be correct, but it’s pretty obvious why someone who thinks that racial discrimination when it serves useful purposes is fine Would be okay with affirmative action, especially since such a person probably doesn’t think there has been any widespread pattern of unreasonable discrimination against whites, whereas there has been such a pattern against blacks and frankly your confidence that you know better than the Supreme Court makes no sense. The civil rights act was signed into Law by a president who was enacting affirmative action at the same time, so I doubt very much. He thought of it as prohibiting affirmative action. There have been lots of arguments over whether those who ratified the 14th amendment intended to ban affirmative action in any case, if you look at the court decisions during that time, it’s not even clear that there was meant to be a bright line against the usage of race as relevant. Supreme Court decisions handed down in the early days of the 14th amendment only say that they suspect that race is not a reasonable basis for discrimination, but it is only a doubt and they don’t appear to have the 2 level equal protection clause that we have invented and instead of appear to be treating everything as a rational basis test, which means that the protection is actually pretty easy to circumvent. Just read the relevant historical debates. There are good arguments for both sides, and it’s telling that the 1880 Supreme Court thought only that they doubted that race could ever be a reasonable basis for discrimination and did not ever go far enough at that time to declare that race could never be a basis for discrimination. They just appear to have been operating under the mistaken understanding that you can’t get useful statistical information out of someone’s race, although they at least had the humility to understand that they could not be confident of this. Personally, I think that just as it is considered fine for police to take relevant characteristics like a person’s age and gender into account when doing their jobs, it doesn’t make any sense to then magically say that you can’t get statistical information out of somebody’s race, and most conservative appear to agree with this, so it’s hardly only a left-wing position that racial discrimination is actually fine when it serves a useful purpose instead of being based on a mistaken understanding of the world or dislike of a group. I admit that personaly it has never made sense to me how people think of race as this magic thing that has moral significance when it comes to discrimination. Whereas everybody will admit that discrimination on the basis of other inate characteristics is fine. what makes race so magical most justification is that make any sense we have to refer to society in history because it’s just opus to me that raise inherently can’t have any moral significance above things like eye colour or where you end up in terms of neurodiversity.
"The reason it’s illegal to discriminate on the basis of race is because there is a widespread pattern in recent history of unreasonable discrimination on that basis."
Telling an industrious, idealistic young man just starting his career: "we would hire you for this job, but your skin is the wrong color" is not a recipe for racial harmony, it is a recipe for sectarian strife and civil war. So yes, racial discrimination is wrong. And as I said, it is doubly wrong when everyone has agreed, through the democratic process, to abstain from racial discrimination, and then these laws are only enforced against some ethnic groups and not others -- and the state itself engages in this same form of selective discrimination!
But you've said you disagree! You think racial discrimination is not immoral. Telling a black doctor that he's not getting hired by your hospital because the patients prefer white doctors is not immoral.
> Leftists think that racial discrimination is good, as long as it advantages their pet ethnic groups.
I consider myself a leftist. I always thought affirmative action is wrong, except in some very limited situations which encompass disadvantaged whites as well.
I think you’re potentially right about the counterfactual, but in this case it seems like we can walk and chew gum at the same time, begrudingly hand-in-hand with people who would be hypocrites in your alternate reality.
I also think you can take heart in the fact that you now have ammunition against anyone saying white doctors are cruel or malevolent towards Black people. Those arguments spread misinformation that directly harms Black people.
The solution is to not worry about this at all. Black people are just suffering the consequences of their own racism. Good, let them. There's no reason for anyone else to care.
This is a pretty uncharitable interpretation of what I was saying. In your example, unlike the situation with black doctors treating black patients, the discrimination does not prevent actual medical problems which can sometimes lead to outright death. I think given how strongly people prefer for there be no racial discrimination. It would definitely pass a cost benefit analysis to prohibit discrimination when it’s only a matter of the preferences of a few patients and they won’t actually end up suffering medical complications if their preferences are not acknowledged. However, at the point where you are actually causing people to die through such a policy, I don’t think it passes such a cost benefit analysis.
My point was that you appeared to be treating no racial discrimination as a deontological rule, this is not how anyone actually treats the issue, regardless of what they say, as should be clear from how liberal are okay with affirmative action while conservatives are okay with statistical discrimination by the police. In actuality, most people think of the prohibition on racial discrimination as something you can break for a good cause and certainly don’t think of situations where it causes loss of life as just fine because no racism is the end of the debate.
You might think this contradict,s what I said about race not being magic, but it actually doesn’t if people were as disgusted and offended. By for example, telling a young and hard-working air hostess that she is not physically attractive enough to be worth recruiting. I would also be against discrimination on the basis of physical attractiveness, but since they do not care that much about the issue, I don’t mind that discrimination. Of course, this is not an inherent difference between race and physical attractiveness. Since there have been societies in history, where if anything not discriminating on the basis of race, would be what everybody would be disgusted by. However, in such societies because of how people think about in groups and out groups, you don’t just have people expressing their preferences, but instead, racial cartels and back scratching clubs that have a huge cost to economic efficiency. This might sound closer to your position, but it really is not because the prohibition on racial discrimination is purely instrumental and if discriminating on the basis of race would produce a lot of economic value. In a given situation, I would be in favour of that. Of course, as I mentioned in my previous comment, it might be that this economic value is not worth the cost in terms of situations where racial discrimination is not reasonable in the sense of not being justified by a cost benefit analysis, but you don’t think of it like that at least as far as I can tell you’re not saying that no racial discrimination in this situation is a necessary evil so that we have to let people risk dying for the sake of a rule that has benefits on net. You just don’t seem to be treating it as a bad thing or a necessary evil in the first place.
Such a categorical prohibition on racial discrimination makes no sense to me. Is it really the case for example that the Belgian government should have let their country fall apart instead of agreeing to discriminate on the basis of whether you are French or Dutch speakers for government positions. Obviously, you can argue that race is not the same as language, but that just brings the issue of why race is different. I distinguished in practice because of the way people feel about it. Discriminating on the basis of race in practice would be a bit like burning the American flag in terms of how offensive it is, and without the free speech argument, but that is a contingent difference. The rule isn’t really that you should not discriminate on the basis of race. It’s that you should not massively hurt peoples feelings and pointlessly offend them without sufficient benefits to justify it. It’s just obvious to me that racial discrimination is not a terminal value. The way that for example, not re deceiving people or preventing suffering is a terminal value that is good for its own sake. You point to the benefit of racial harmony, but firstly I’m really doubtful that the impact on racial harmony will be large enough to outweigh people literally dying of medical complications and secondly no racial discrimination would as people like Nathan Cofnas point out, given the gaps in test scores between the races. No affirmative action would mean that they would be almost no black people at Harvard and other elite universities, which to me obviously sounds a lot worse for racial harmony, then the present situation and something that even the public in general would not tolerate.
You have done nothing to address the point that LBJ was already doing affirmative action when he signed the civil rights act into law, so it’s highly unlikely he intended it to forbid affirmative action and indeed, Congress could have amended the act in response to the Supreme Court decisions and yet did not. In fact, the Supreme Court when it decided was decisions was actually more conservative. Then it was when the civil rights act was enacted. I don’t think it is credible that Congress did not realise that there was absolutely no way that without specific instruction to the contrary, the warrant court was going to interpret the civil rights act as forbidding affirmative action, so I think it’s obvious that you can’t just act as if Congress on unambiguously supported banning affirmative action. Not to mention that the Supreme Court itself is democratically legitimate given that it is appointed by the democratically elected president and its members are confirmed by the democratically elected Senate. If there is any ambiguity, it gets resolved by the Supreme Court now, obviously the Supreme Court has changed their mind and that’s legally legitimate. But firstly there are a few loop holes in the judgement. For example, permitting universities to pay special attention to their opinion of whether a candidate has displayed desirable qualities, fighting racism, such that their struggles are evidence that they would make a good candidate. And in any case, there are other avenues like the obvious one is trying to persuade the court to change its mind, and even if the optimal policy isn’t constitutionally permissible, that doesn’t mean, we can’t talk about it and argue that it will be the best policy. Something being illegal is a reason you should not do it definitely, but it doesn’t mean you have to pretend that it won’t be the best policy if it became Legal.
Also, your kind of misleading about the public, the actual public responses to most service indicate that the public simultaneously wants no discrimination on the basis of race and a level of equality in outcomes that is not possible without such discrimination and will often say yes to questions that in practice amount to doing affirmative action in ways that require five minutes of thought to recognise them as affirmative action instead of stating it upfront. This is not really surprising after all, the public was hugely in support of the nominations of Justice Marshal and Justice Thomas to the Supreme Court. Even though they were both affirmative action appointments because the public actual preferences to do affirmative action while being sufficiently non-transparent that the public and fool themselves into thinking there is no affirmative action going on. Indeed, I think both justices would have had a harder time getting confirmed if they weren’t black given that they were major concerns about the Supreme Court being to liberal when justice Marshal was nominated, and I doubt just a Thomas would have had a huge public support, he did if he were not black, especially given the common wisdom was that no one has extreme as him could get confirmed if they were not black and he had only a few months experience on the appeals court, which is normally not considered sufficient to even be considered for nomination to the Supreme Court. That’s not limited to Thomas either I don’t see justice Marshall being nominated. If he was not black, he didn’t have some huge reputation or anything, and indeed was specifically given positions that would make him qualified to be nominated to the Supreme Court on account of his race, so it’s clear that even five minutes of thinking would make it obvious. They were both affirmative action appointments.
We should give white racist patients a doctor they trust when it's practical. If grandma doesn't trust the black doctor and won't listen to them, let her see a different doctor. It's not worth her dying over.
This doesn't mean we stop telling people not to be racist. We will keep educating people. But you don't leave someone to die because you want to teach them a lesson.
The article says it's not advocating for affirmative action; only that this is the strongest argument they've heard for it.
38: The graph is missing years in the middle as well. All over the place, really. After 1989, it starts leaping ahead anywhere between 2 and 6 years at a time. I get the sense that maybe they didn't poll for this every year, but that doesn't explain plotting the graph that way.
And Jon Kozan brings up a good point: 2022 is likely to be an outlier for Covid-related reasons.
Just because NGO workers are cheaper than unionized city workers doesn't make outsourcing non-progressive. Because you can hire more NGO workers for the same cost, and they're all so personally progressive, outsourcing becomes in fact a very politically progressive thing to do. Outsourcing municipal services to nonprofits, over the long term, both weakened unions and helped elect Mamdani: https://thebiggerapple.manhattan.institute/p/mamdanis-ngo-model-army
Outsourcing to contractors (NGOs, religious charities, for-profit companies) lets you dodge all the civil service rules. That means easier to fire, wider range of pay (civil service pay has a lower ceiling and higher floor than private sector pay), fewer benefits, but also rules restricting political activity and trying to get rid of the old spoils systems.
>Hopefully you can predict my response, which is that all concepts become unmanageable when you zoom in too far and prudence tells us how far we should zoom at any given moment.
When we're talking about cooperative intellectual discussions, yes.
When we're taking about crime statistics? Who police and prosecutors choose to charge, and how they charge them, is probably influenced by many things other than intellectual prudence about how far to zoom in on a concept.
Put it this way: if the same event could lead to zero, one, two, ten, or a hundred thousand criminal charges depending solely on the preferences and habits of the police and prosecutors involved, that suggests that you *really* need to come to grips with how much variance and what patterns exist in that determination, before the crime statistics can be very useful for making any arguments about the underlying reality.
I get scott's point, which is true and valid, but Scott's point invalidates the main thing I'm pushing back on, which is there is some objective measure of "real actual crime, of the sort people care about" beyond "what gets reported, or charged, or convicted."
As a legal realist, I think it's best to view crime as a thing that GETS CHARGED, but the moment you step outside the framework of what gets charged into a framework of how much crime there REALLY is, things get uncomfortable.
My article proposes several "zoom modes", all of which I can relate to, but none of them work. To reiterate:
"By crimes I mean crimes with victims" (nearly all my examples involve victim crimes, like theft)
"By crimes I mean crimes which people actually care about" (all of SBF's myriad frauds were things people actually care about)
"By crimes I mean crimes which a jury would convict on" (I truly believe the prosecution in SBF's case could have chosen from perhaps a million possible acts of wire fraud and the jury would have convicted whichever they chose)
I didn't even discuss some of the weirder ways to chop up these numbers. Most of us are familiar with the Mangione case: he's charged at the state level and at the federal level. Most federal crimes are also state crimes, and sometimes vice-versa.
If we're going by what is a crime, in the objective sense, beyond what gets reported, charged, or convicted, I think the only metric that makes any kind of sense is "does the act meet the elements of a statute defining the crime" and by that metric there are gozillions of crimes, and most "discrete crimes" (say, a murder) are actually thousands of uncharged crimes (first he lunged at the guy, that's assault, then he took out a knife, that's assault, then he drove the knife forward toward the guy, that's assault, then he stabbed the guy, that's assault, then he stabbed the guy again, that's assault, then he stabbed the guy again, that's assault, then he waved the knife around wildly, that's assault, then the guy died, that's manslaughter, murder in the 3rd, murder in the 2nd, murder in the 1st, then he threw the body in a ditch...)
I also wonder about things like felony murder or laws related to gangs and conspiracy... I don't know this area of the law well enough to have firm opinion, but my understanding is that it can be possible, for instance, to charge many people with murder for a single death. Which makes me wonder about using murder as the 'reliable' statistic.
The statistics are "reliable" in the sense that they accurately record the number of dead people, which is what most people care about.
Like I said in my article, I think if you trust various parts of the Criminal Justice System to have an accurate bead on the public's sensibilities with regard to crime*, "reported crimes" "charged crimes" and "convictions" are going to meaningfully measure "crime", because a legal realist can say "crime is that which is reported as a crime and/or charged as a crime and/or results in criminal conviction"
If you DON'T trust the Criminal Justice System in either direction: you think it's charging vastly more things that "aren't *really* crimes" or you think a large number of "things that are *really* crimes aren't getting charged" then you enter this hellzone of "what is crime, actually" and in that zone lies the calculus of my post.
It is true that crime statistics are deeply imperfect, and that part of that imperfection is prosecutorial and police discretion. It is also true that this doesn't actually let you just wave away reality wrt things like crime statistics by race, where it is absolutely clear that (for example) blacks commit many crimes at much higher rates than whites (which hispanics in-between and asians even less inclined to crime). The imperfection in these statistics, and the fact that crime itself is defined socially and not every society even considers the same things crimes, also doesn't prevent us from noticing that there have been some years when a lot more dead bodies with bulletholes in them turned up than in other years. Or that in some years, a lot more cars got stolen or a lot more stuff got stolen from store shelves than in other years.
Well yeah but in a lot of those specific examples you're citing the baseline reality is the exact same crime statistics reactionaries typically push back on.
You will absolutely not get a negative comment from me by observing "homicides went up in 2021" but you will if you say "overall, crime is going up, despite what the statistics would have you believe." That's where I'm going to start saying "okay but what do you mean by crime?"
RE 34: Is this a credible source? I just find this hard to believe. It's just so easy these days. You really could just ask AI to "find me which of these projects are pseudoscientific woke bullshit written by virtue signalling left wing activists pretending to be scientists" and it would actually do a pretty good job and not make idiotic word context mistakes like that.
Quote: "NSF officials declined to comment on the intricate, multistep process, which began last week. But people with direct knowledge say it started with NSF senior managers selecting 10,000 grants, from a pool of roughly 50,000 active awards, for review. They then enlisted staff to vet the awards using a list of key words that included “diversity,” “inclusion,” “women,” and “race.” To be sure, a sizable share of those words refers to scientific designations, such as plant diversity, that have nothing to do with DEIA; those awards were dropped from the review."
Are you misreading that quote? It says that they filtered based off a list of keywords but then manually reviewed that list and *stopped* investigating grants where the keywords were off-target hits. The following paragraph in your source makes it even more clear. They may have still mistargetted some studies, but your quote doesn't support that. Your source also claims that any identified studies were just going to have to be "modified" and were expected to be fully funded.
I don't use twitter so maybe I'm missing it, but the linked tweet from Scott's post only has one example and I'm not confident that they know why it was cancelled. Or that we know what the supposed list of naughty words is to even know that poralization is on it. Can anyone highlight the most conclusive example of this? I'm a bit skeptical of this narrative. The "transgenic" claim that went viral was as far as I could tell entirely speculative and there is no reason to believe that they confused transgenic for transgender. Certainly the specific examples that the White House ended up giving did not make that mistake (though they were also totally valid lines of research that were tangentially related to transgender people just as part of their broader impact - cancelling them was dumb but it wasn't CTRL-F "trans" level of dumb).
Perhaps I did misread it. In any case, there is a lot more in the Wiki article I linked; the science dot org article is but one of the references in it.
NY Times: "Additionally, the presence of some terms was used to automatically flag for review some grant proposals and contracts that could conflict with Mr. Trump’s executive orders."
Duke's The Chronicle:
"The 15 canceled NIH grants contained flagged words like “expression,” “disability,” “trans” and “systemic.” However, these words were not used in contexts to promote DEI initiatives, although some may have had diversity supplements including the two described below.
The grant that had the word “disability” in its abstract focused on identifying biomechanical and biological factors that lead to early onset osteoarthritis following a partial meniscectomy — a surgical procedure commonly done to remove damaged meniscal tissue following a tear. The grant had over $750,000 remaining in funding at the time of termination.
Furthermore, a grant that had both “trans” and “systemic” in its abstract was working toward editing neural circuits by engineering channel proteins to control the point of contact between two neurons. It had around $1.4 million remaining in funding at the time of termination.
In most other cases where “trans” was flagged, the term generally referred to other words with the trans- prefix, such as disease transmission, transgenic genetic material, translational studies or signal transduction."
So the article states that both had diversity supplements which I think are more likely the cause. FYI, "flagged" refers to a third-party called Grant Watch, not to any direct knowledge as to what was done by the Trump admin and the word list originates from the NY Times article above. My personal assumption is "ctrl+F and then review" which fits this all better than just "ctrl+F and delete" but isn't as meme-able.
To be clear, I'm very against funding cuts to science, I work in an NIH-funded lab, I strongly oppose the administration using grant cancellations to target its enemies or pressure universities into obedience. I just don't see evidence that they did the most absurd level of mistake like confusing transgenic with transgender or dropping all research into polarization of light due to a keyword search.
The graphs on 22 showing international markets providing 43% returns to 14% US returns are...I would appreciate further context here.
VTSAX (Vanguards total US stock market index fund) is up 18.18% over the past year and 10.76% over the past 5 years. (1)
VEU (Vanguard's global stock index fund excluding the US) is up 27.68% this year and 7.73% over the past 5 years. (2)
For reference, VXUS, Vanguards global stock index fund including US stocks is basically identical to VEU. (3)
I think it's worth noting that these numbers differ significantly from those presented by Catherine Rampell and global performance has dramatically underperformed the US for the past couple years. Trump could be screwing everything up but the US has been going up 20%-ish a year while the global funds have been returning like 9% up to this year. That feels a lot more like a market correction than a Trump screw up. USA #1, yeah, but probably not literally twice the growth in other markets for decade.
But, again, not a finance guy, just a retail investor, but there's no Vanguard fund I know to buy that shows those returns.
I have been puzzled by this as well. It certainly seems plausible to have a capital domino effect, where just a few movers and shakers greasing the international markets leads to a cascade effect (guilty as charged, recently sold my stake in VTI to buy more VT, also why doesn't the I stand for International, le sign)...but is it really due to a US-specific drag, an ex-US mispricing, something else? Lots of stories one could fit to the data, with differing investor implications.
Not seeing it universalize across funds either. My workplace's retirement accounts aren't held with Vanguard, and over there...the US-centric funds are still on their usual tear, while recently international ones are *also* doing better than their longer-run average. So it seems much less like a direct substitution effect? The recent unpleasantness in the Middle Earth did tank US stocks more bigly than internationals, which has to be part of the very-recent story. But after recovering from that shock...shrug, I'm not sure what's going on either, or where to put more money going forward.
This graph is generated before the end of February (i.e. the start of the war). I hold Vanguard Total International Stock Index Fund (VTSNX); it appears to be plausibly up by the amount reported before the war correction.
> It also (imho correctly) calls out modern “postliberals” as pre-Rawlsian rather than post-Rawlsian, in the sense of not confronting or having a good answer to Rawls.
CTRL+F for "postl" or "post-l" yielded no results. He talks about illiberalism generally, rather than claims to be post anything. Of course, if a self-described postliberal (like Ed Feser, who distinguishes himself from the more unprincipled of the postliberals) claims that they are actually just harkening back to Aristotle, Augustine, Aquinas and so on I don't think they would see it as much of a critique to call them "pre-Rawlsian". Even the people he specifically calls out would scoff at the idea they need to persuade "all" people first, our existing pluralist society already enables people to enforce their values even on the unwilling by obtaining political power. Nor do I think it's a serious indictment of someone that they don't "confront" Rawls or Kant once you acknowledge their basic moral principle is incorrect, any more than I think it's a problem that Tolkien didn't grapple with post-modernism (something he's actually dinged for at https://starsbeetlesandfools.blogspot.com/2015/06/worldbuilding-bakker-vs-harrison.html?showComment=1458955283269#c4652106653130330971 ) People who are into that may find it important, but everyone else who thinks it's bunk can ignore it once people like Huemer point out that it's bunk.
Only paid subscribers can comment on Heath's own post, and if I could I'd point out that there's a sleight of hand to go from "liberalism practically died out in the early 20th century" to the very next sentence talking about Rawls' "strategy for reviving liberalism" seeing as how his book was published in 1971, years after the Great Society had established its version of liberalism as a less extreme version of what Rawls wanted. It's not like he was trying to revive classical liberalism, which was what had "practically died out".
Re:#2: Japan has a similar rule to Uruguay, as per the law. AFAIK, and according to Wikipedia, nobody has ever been punished for desecration of either the Hinomaru or foreign flags. However, one of Takaichi's campaign promises is to make Hinomaru desecration a crime.
Adjacent to 23, has anyone else noticed things like YouTube and news aggregators censoring non-profane words like "sex" and "killed"? I read a Yahoo News article recently that censored every occurrence of "sex" IN AN ARTICLE ABOUT SEX. What's driving this? Has the internet become so mainstream that every media platform is adopting 80's-era TV policies? I have a hard time believing that the presence of words like "sex" hurts viewership or ad clicks in any way, so what's going on? Are interest groups policing this or something?
I do not know *why*, but it definitely does happen. On www.quora.com many posts use a euphemism "unalive" because the verb "kill" is liable to push your post to the bottom of any search.
Shouldn't you provide some support if you're going to say "it definitely does happen"?
Behavior on Quora may tell you that the people there (Who? That site died years ago...) believe it happens, but doesn't tell you anything about whether it happens.
OK, I can not PROVE that Quora or YouTube algorithms penalize posts which use the words "kill" or "porn", and short of the website owners actually admitting it, proving it would be a long and involved project. Still, enough people believe it, to use words "unalive" and "pr0n" in order to fool these algorithms.
In this case, I think it's worth making it very clear that the odds that Quora is doing this are indistinguishable from zero, and the user behavior is known with certainty to be a contagion from user behavior on other platforms.
> words "unalive" and "pr0n"
"pr0n" is leetspeak. It's only meant to be a cultural marker.
For what it's worth, if a Quora post contains certain words, any images will be blurred (clicking on the image unblurs it). That is consistent and demonstrable. Here is an example:
Tiktok, Instagram and pre-takeover Twitter had aggressive censorship for certain words liable to appear in controversial posts. I'm not sure if other sites are adopting the same policies to keep their own posts shareable on social media or if certain users are adapting their language the same way on every site out of an abundance of caution.
The whole phenomenon is discouraging, but some of the workarounds are amusing. Keep an eye out for tiktok-speak like s3x, @bort, 🔫-violence, and yes the infamous "unalive"
A decade or more of major social media platforms like Youtube and TikTok trying to maximize kid-friendliness. They want kids watching to maximize watch time and how many ads are shown to the most eyeballs - that means optimizing for the kind of content parents won't object to.
We are now beginning to see the social effects of this as a whole mass of people and cultures have normalized the idea of not using the common word for a serious issue.
Optimizing for the kind of content _advertisers_ won't object to. Corporations don't give a damn about kids or their parents, but they care very much about ad money, since pretty much all of the internet still runs on it.
When you don't know the exact rules that will make your video demonetized or not recommended by the search algorithm, only you know that some rules exist, people try to play it safe and copy each other's superstitions.
It's hard to know what the decision makers are thinking. Maybe it's a desire to please the wokes who demand censorship of "hate" without pissing off conservatives too much. The solution is to censor/deboost/demonetize anything that's more than PG. The "bigot" doesn't get to share their views on sex nor does anyone else.
Yes, I have seen this. Channel 5 Action News had a recent video about a suicide which censored the word "suicide" and "kill"; RedLetterMedia have been censoring the first few minutes of their videos to avoid demonetization.
>only three emperors - Vespasian, Marcus Aurelius, and Septimius Severus - had sons who could inherit
Apart from Vespasian's elder son Titus, most of the sons in question were assassinated, and I don't think many of them have great historical reputations. That doesn't sound to me like a sign father-son inheritance was the missing ingredient.
The "Five Good Emperors" period seems to be generally considered the high point for the Principate in terms of both stability and quality of governance. Of the five, Nerva was elected by the Senate and the next four were nominated by their respective predecessors.
If you're a Roman emperor with no living sons, then you get to name your own heir - almost certainly an adult who you've been able to watch and assess for their potential in the role. That does seem more likely to lead to good outcomes than "first child with a penis to come out of my wife's vagina".
In the latter case, you at least get a bit of a genetic edge in that the father was demonstrably able to run an empire well enough to avoid major civil wars for a couple of decades, at least. But if natural succession is a rare exception, you're only getting one generation of natural selection on that axis. And the candidates for adoptive succession were at least being drawn from the families of capable aristocrats.
True I'm most familiar with the details of the succession during the Julio-Claudian Dynasty, which seems to fit that pattern. Historical reputation seems to run (in descending order):
Augustus (adopted heir of Julius Caesar, last warlord standing)
Claudius (installed by praetorian guard)
Tiberius (son-in-law and chosen heir of Augustus)
Nero (blood heir)
Caligula (blood heir)
-----
It also occurs to me that during the Principate, the Senate still mattered quite a bit as an institution despite the Emperor clearly being in the driver's seat. Emperors who knew how to play the game of Senatorial politics tended to do a lot better than ones who didn't. The fatal flaw of the bottom two Emperors on the list, both of whom inherited young without having opportunities to practice the game before becoming Emperor, was that they gave too many Senators too much reason to hate them.
>If you're a Roman emperor with no living sons, then you get to name your own heir - almost certainly an adult who you've been able to watch and assess for their potential in the role. That does seem more likely to lead to good outcomes than "first child with a penis to come out of my wife's vagina".
I've heard from a number of classical historians that in the Roman republic, continuing into the Roman empire (at least for much of its duration if not all,) heredity by adoption was considered more significant than heredity by blood, for more or less exactly that reason. If you were the person that an individual singled out as his preferred heir, that was more dispositive of you possessing the skill and character he valued than if you were simply his direct genetic offspring.
Also, adopting a successor while still healthy meant you were at least planning well in advance for succession. One of my takeaways from the Mike Duncan History of Rome podcast was that most Roman emperors simply didn't do this for reasons that remain hard for me to understand, and a good number of those who did chose badly (hot take: the extreme example of this is Diocletian's cockamamie Tetrarchy scheme which he should have known would fall to pieces). "Apres moi, le deluge" indeed.
the sagittal-shift of awareness mentioned in 51 is interesting. this week, somehow for the first time, I noticed my thoughts centering on different points along that axis in relation to how anxious (forward) or mellow (backwards) I felt. taking decades to sense this left me suspicious of it being a passing illusion, so it's nice to see some evidence to the contrary. powerful motivation to try daily meditation again too.
My Firefox homepage is the Wikipedia random article link. Half the time it's an obscure species of moth, or the summary of the 1997 season of some minor local football club, but other times there are good ones (with the usual caveats that Wikipedia has become much less reliable a source in recent years on politically-charged topics--or rather that the list of politically-charged topics has expanded greatly; the page listing lamest edit wars extends to the early days, but mostly covered relatively unimportant topics).
The article on Nectome is great. I'm very interested in the idea of cryonics, but extremely disturbed by the practices of currently existing cryonics companies.
Trying to preserve a brain after it's been dead and warm for hours is completely pointless. The physiology of cell death and synaptic degradation is not fully understood, but it's understood well enough to know that not much is being preserved after 24 hours. And probably not much after 9 hours. There is slogan in stroke neurology - time is brain. So it's a huge credibility boost that Nectome will only preserve a brain if they can do it within 15 minutes of death. The fact that other cryonics companies are happy to take people's money to preserve a brain that has been dead and warm for over a day is damning.
Nectome's other innovation is also clever. Using cryoprotectant plus fixative sounds obvious in hindsight, but its a big advance. It makes preservation more accurate, and cheaper which is a big deal when your goal to is to keep preservation going for decades to centuries.
I find the following pretty difficult to parse. What exactly is a "zigger"? Do I even want to know?
"Our suppression paved the way for a lot of the ziggers and degeneratron grifters, who were not as all-pervasive as they are now. The major pages were generally pro-american of course, while the “degenerate right” (completely useless people who contrary their claims have generally terrible taste and no artistic talent btw) was constantly hounded with a fair measure of success. A lot of the stuff you see nowadays is because people have gotten worse in many respects, but much of it is because the purges destroyed the “internal hygiene”."
Insofar as I understand it, though, I disagree with it. It always seemed to me like the excessive censorship of right wingers before Trump's second term / Elon's twitter takeover was mostly a positive influence on the right, by artificially inflating the average content quality. Ever since then, we've had more success, but a lack of the restraint and standards we had before. At this point I think I regret not voting for Kamala.
A group might have different standards for what is hygenic. Just because you both agree that swinging around Nazi flags is uncouth, doesn't mean you have the same attitudes towards what is considered acceptable vocabulary.
The goal, ultimately, is to have people like you no longer be in a position to police language.
> A group might have different standards for what is hygenic.
In the specific case shown in the Twitter thread, it doesn't seem there was much hygiene going on. Calling all Muslims terrorists (or something equivalent, I suspect) or changing one letter in a racial slur while still trying to maintain its...contemptuous connotation, is not exactly hygienic.
A recurring issue in discussions about not censoring the right is that in the current environment, the right seems to delight in saying things even non-partisan walled gardens wouldn't tolerate.
> The goal, ultimately, is to have people like you no longer be in a position to police language.
If the people want Parler, they are free to have it, but they should be honest about that and also how much hygiene they actually have.
> the right seems to delight in saying things even non-partisan walled gardens wouldn't tolerate
I don't agree that there is such a thing as a "non-partisan" walled garden. Everyone has their biases, and an ideology they want to push. Just because you consider it the default doesn't mean it's non-partisan. Can you not imagine a society where the things you describe are found to be perfectly acceptable and normal? You wouldn't even need to go that far back in history to find it.
On the US vs. Foreign stock market returns, that could also be some mean reversion as the US stock market risk premium is pretty low right now, and foreign risk premiums have compressed more in recent months than the US has. https://elmwealth.com/capital-market-assumptions/
I have a PhD, and am probably qualified to teach math, physics, CS and chemistry at the highschool level. But there's no way I'm going to do it. Not because the salary is low. But because you can't pay me enough to deal with all kinds of behavior issues, administration and all other things that highschool teacher has to deal with. If teaching and every other aspect of the job were divorced it would be a completely different ballgame
Don't forget parents. I'm lucky enough to have nabbed a decent gig with my PhD, but I would rather do menial physical labor like warehouse inventory shit than teaching high school if the PhD gig dries up.
Very much agree. I too could teach the subjects you listed (BS in mathematics, MS in computer science, minor in marine biology), but the rest of the bullshit is impossible for me for ANY amount of money.
My understanding (my mom was a classroom teacher and I've known many more) is that the skills of managing a classroom full of kids are just entirely distinct from knowledge about a subject. A normal classroom full of normal middle-school or high-school kids requires a certain level of skill in keeping order and keeping everyone on-task, and if you can't do that, it simply will not matter whether you know your material, because you won't be able to do much teaching.
Of course. Which is why the whole talk of "let those poor adjuncts teach highschool" is pointless. Unless we, as a society, decide that its worth setting up schools for kids who both don't need much management and could benefit from being taught by PhDs
Also, most kids don't need someone with deep expertise in high school. The lady who understands high school algebra but doesn't know what a Galois field is can do a fine job teaching high school algebra. It would be a pretty bad allocation of resources in our society if most high school math teachers were mathematicians.
My daughter is actually planning on becoming a HS math teacher, and has been looking at the degree requirements from various colleges.
Surprisingly (to me), many require a double major in education & math, and those that don't still have course requirements (e.g., real analysis) that I'd've thought would be completely unnecessary for the job.
I don't see why this is such a surprise. One would hope that a teacher would know more in the area than the maximum that the students are expected to reach. I also keep hearing horror stories about "degrees in math education" that may or may not teach the education part, but definitely don't teach the math one
Agreed on going being what you're teaching; I often advise coworkers to understand things at least one level deeper than you're going to be working at.
It's just that the requirements are even further beyond HS level than that; I have a math minor & never took real analysis.
I'd like to see the relationship between willingness to endure all that BS and left-wing views. Are progressives just more willing to sacrifice comfort for their egalitarian goals?
Egalitarian? I believe that the discipline situation is slightly better in private schools, but the rest of the drawbacks are pretty similar if not worse. Especially in the "elite" school with very rich parents
On link 3, I work in a UK local authority in social care. We have a small in-house service, which is expensive, inefficient, and very difficult to make any changes to. On the other hand, we have a large external home care market, which has upwards of 20 companies delivering home care at literally half of the effective cost per hour of the internal service, while being responsive, flexible and high performance. There have been some poor performers, but we just don’t commission any services until they get better.
We have one of the lowest levels of unmet need in the country but still get pelters about our use of the “profiteering” private sector, ignoring the excellent services provided, and constantly get demands to deliver more in-house services, despite how it would double our costs or half our capacity.
In general, recruitment is difficult because it’s even hard to get rid of fixed term workers now, due to layers of process and hoops to jump through. It’s far easier and lower risk to commission a service from an external provider for a fixed period than it is to hire someone to do something directly for you.
I think the black doctors finding is better explained as people are more likely to divulge embarrassing but consequential symptoms to someone they can superficially relate to (in group).
37. Genuinely not trolling: I'm not sure that adjunct professors and postdocs have higher status than high school teachers these days. I'd argue strongly in the other direction. Personally (+1, I just asked my wife too) I'm really not impressed by people who dick around in universities past their sell date - it's lame, and the reputation of universities is in the toilet. I AM impressed by someone who gets in the unions, sticks with a solid career without getting distracted, and does the incredibly difficult job of teaching my kids. Folks - particularly the fellas out there: both men and women want to marry that person. They don't want to marry desolation in tweed.
I think you probably meant just "status among other adjuncts who are hyper-concerned with the status of adjuncts", but that's like two guys per state at this point.
44. Come ON. "Kids intentionally fucking up an otherwise good essay so the teacher doesn't think that it was written by AI" are not sweeping the nation.
44. I have personally seen a current student spending hours each week doing exactly this.
All classwork must pass the Grammarly “AI detector” which results in said student going sentence by sentence and arbitrarily shuffling words around or adding casual adverbs like “a lot” until the opaque algorithm has decided that line is human enough.
I really hope that student is insane and paranoid, like full-on Richard Nixon. :) If what you're saying is correct, it means that the "AI detector" is worthless, but the school still use it, which is probably a metaphor for something or other.
Actually let's check in with the LLMs for a "funny, pithy metaphor":
From Gemini 2.5: "It's like teaching a fish to ride a bicycle...backwards, just to prove it's a fish"
From GPT 5.4: "That's like making your kids spend hours sanding down their fingerprints so the detective doesn't think they committed the crime of being human"
Alternate GPT 5.4: "It's like teaching your kid to write with a fake limp so the robot hall monitor won't accuse them of sprinting". I can't believe ChatGPT put this gem in second place!
Claude 4.6 Sonnet: "It's like having to limp so the speed camera doesn't think you're a car". LLMs like limping, I guess.
Sometimes I'm not so worried about the new robot overlords
What sort of status you have as an "Adjunct Professor" depends on whom you're talking to. To most normies it sounds exactly as impressive as Assistant Professor or Associate Professor; only people in the know understand you're basically a galley slave.
Being a "postdoc" sounds like being a glorified student, and how prestigious that sounds depends on how old you are.
Yes, exactly, what this guy says, which is also why it's trivially easy to dismiss the story as a *moral* piece of work. A document which says, "you must believe in impossibilities (utopias) to be moral" wrenches my soul exactly as much as saying "you must believe that 2 + 2 = cheese to be moral."
However, it's wonderful from a storytelling and writing and condensed meaning point of view. I take the thing as wonderful piece of art and not much more.
I don't even think the woman actually believes this argument. Compare to:
"Only in silence the word,
Only in dark the light,
Only in dying life:
Bright the hawk's flight
On the empty sky.
—The Creation of Éa”"
From Earthsea.
Omelas is not an argument. It's a sort of elegy in honor of noble madmen she *wants* to be right but knows aren't.
While the *speed* of Taalas is indeed impressive, its quality is very low. I asked it what price cinchona would fetch if somehow transported to ancient Greece and it responded "Cinchona (Chinchona officinalis) is a fascinating plant with a rich history. In ancient Greece, it was indeed a valuable commodity, but not for its medicinal properties [...] Instead, its value lay in its unique and highly prized timber. Cinchona was a highly sought-after timber for making bows, particularly the bows used in ancient Greek and Roman warfare. The tree's wood was prized for its strength, durability, and flexibility, making it ideal for crafting high-quality bows. A single bow made from cinchona wood could command a high price, reflecting the value placed on quality craftsmanship and martial prowess."
Cinchona is a New World plant, unknown to the ancient Greeks. Not only that, but this fact had been explicitly mentioned in the previous question. Even Deepseek did better than that when asked the same question.
Re: latency: It gave me a response in ~0.5s (according to both browser dev tools and feel). The only reason that feels magically fast is that you're used to bloated websites written by lazy developers (or, rather, expensive ones, so companies are too stingy to pay for more developer time to optimize them). A basic, 90s style website, with a decent server and internet connection, should be *much* faster than this.
Latency matters most for multiplayer video games. In my experience, typically it's about 3x the minimum from speed of light constraints: from Europe, ~0.1-0.2 s to North America, ~0.4 s to Australia.
I once came across someone on Reddit who said he'd gotten an assignment whose spec required a ping time that would need a signal literally traveling faster than light (sub 50 ms for a server on another continent, IIRC). He had a tough time explaining to his boss why that wasn't achievable.
It said “Generated in 0.046s • 15,436 tok/s”. It’s hard for a human to measure that and I’m not sure if they are measuring the entire rendering or not. But that’s much faster than 0.5 seconds and I’m sure it wasn’t as slow, as half a second which is measurable to humans,
The speed really isn't just an illusion of low server latency. The Taalas HC1 chip outputs tokens at about ~17,000 tok/s, as compared to ~250 tok/s for a normal H200. The big downside of that 100x speed increase (and ~0.2% inference price) is that the LLM weights have to be physically printed on the circuit- so the chip can't be used for anything other than inference of that one specific model, and when the model becomes obsolete, you have to throw away the physical hardware.
Chips like that probably won't be used for inference at the big labs, but they'll likely have a lot of important niche uses for devices where speed and cost are more important than having up-to-date models. They might get used a lot in robotics and military drones.
> The speed really isn't just an illusion of low server latency.
I'm not saying it is. I was responding to Scott's impression that *in addition to* making the AI fast, they had to do some magic to the latency, otherwise that alone would make it slower than it is.
Came here to say this. The model generation is impressive and novel, but the internet latency is what you'd expect from a website that's not bloated by ads, tracking data, analytics, etc.
Plus the fact that a page load requires DNS resolution which requires a trip to a separate server (or two separate servers?), whereas loading more text on an already-loaded page doesn't need to do that.
(speedtest.net shows me getting 4ms ping to their nearest server.)
It's Llama 3.1 8B, probably quantized. That's about the quality level I'd expect from that model. (Bear in mind that useful models are usually 100-150B parameters, and not quantized!)
I just tried the stock, unquantized "instruct" variant of Llama 3.1 8B on OpenRouter and it *started* to try to actually answer the question before trailing off into nonsense like "We would have to consider also external market slice beyond known full consumption: other backgrounds species sources and oils contemporary gods can generate also proponents minescontrol mos fuatement oil refining travel propag constituent voy entirely tactieve winning Rub labour".
So... whether you consider that better or worse than the Taalas version is an open question. ;-)
NO! Do not send wannabe academics to teach highschool. I'd sooner send meth cooks, retired moms or local sex workers. If you **like** teaching and are good at it you can usually get a teaching job in a college but the reason you are enduring being an adjuct is the hope one day you will get to live the life of the mind with an occasional class.
Usually the features that send you into academia make you great at research -- you aren't that great with other people but you love the life of the mind or don't have the courage to do anything but the default -- but bad at teaching. And HS teaching is mostly authority. Imagine the grad students you know trying to face down a class mocking them or ignoring them.
And knowing so much about the field can make you positively problematic for a highschool. For instance, Math PhDs always try to actually teach the concepts and show the proofs.
Maybe it would be neat to have them at magnet schools where the kids mostly teach themselves but no, just no!
I spent one school year as an adjunct professor, and the only reason I did it is because I had no other options. I dropped it like a hot potato the moment I found a full-time database developer position.
I went to a decent school, let's say roughly half of PhDs went to decent academic gigs and the others wound up elsewhere (or with bad academic gigs). Which is to say, a lot of my peers are the type one might think of going to teach high school in this scenario. And absolutely, to your point, the emotional IQ of these people is just as room temperature as the stereotype of any other PhD.
Re 15: so segregation is a good thing now? I have no clue if the claim is true, but of it is, then it could apply to all other countless discrimination criteria human invented.
Re 21: based on my personal observations, the US is hard to enter legally, easy to stay illegally, the EU is the other around: easy to enter legally, hard to stay illegally.
“17: In experiments, current AI was not very good at helping a bioterrorism red team with the kinds of tasks involved in creating bioweapons.”
Having Claude read the paper, you can't actually rule out that it wasn't helpful. The AI got the participants further through 21/22 subtasks. It just didn’t help them actually complete the entire workflow. And this was summer 2025 LLMs, not current ones! Even current open models are 3-6 months ahead of where these ones were.
I don't know if I'm up for reading the whole paper myself right now, I might come back later, or perhaps someone could double check.
Linked from the Microwave Cooking article, but seemed fun enough to link directly here - instructions on how to smelt metal using a microwave: https://www.instructables.com/microwave-smelter/ . (Also includes instructions on how to make a fireball with a microwave, but that's an optional step)
Regarding the Pitt Bull issue I got into this recently and there are a host of reasons to be skeptical. However let me start by saying that even if it does turn out they pose a modest extra risk I think the right place to start is just mandatory sterilization for the breed with very few exceptions (or even priority on the kill list at shelters above other healthy dogs).
People get very upset about losing their pets -- I'd rather have quite a few serious dog bites (even as a child) than be forced to give up my dog -- so from a total utility point I think you would need to prove that most pit bulls bite multiple children to justify a retroactively applicable ban or even rules that make it essentially impossible to find a place to live with a 'grandfathered' pittie.
But the evidence for danger also seems kinda weak.
1) Ths reported incidence of Pitt Bull bites is very suspect data. Turns out identifying dogs by eye is incredibly unreliable and what people say is a pit bull doesn't actually track the breed that closely. Further observe that the bite data seems to report an unusually high percentage of pure bred pittiea and it sounds like lots of the data is just someone going: I was bitten by that big scary dog and infering the breed from the fact it was a scary dog that bit.
2) We have confirmation that people with primary psychopathy are more likely to own ‘viscous’ dog breeds. In other words, when drug dealers want a scary aggressive dog they buy dogs people think are scary and aggressive.
Thus -- even if pitties are no more inclined to be aggressive than your average dog we *expect* then to be overrepresented in bites! Dogs trained to be aggressive will be.
But bans on pitties are like banning cammo colored stocks -- if the cause is the buyer you don't reduce the incidence but you probably do force them to transfer their pit bulls to mom (out of apartment/state) and get a Malinois so now more aggressive dogs and someone with a dog they can't necessarily handle.
3) There is something of a summary of studies on the issue here but without any real clear answer. The most shocking thing I learned is that basically none of the studies seems to control for breed popularity.
Ultimately my conclusion is basically that the evidence for increased dangerousness of pit bulls is much weaker than it might initially appear. It is probably a genuine correlation but the size of the effect isn't known.
However, we have much stronger evidence and priors that 'bad' people who want a scary dog are more likely to buy Pitt bulls. Therefore we are left with data that can be explained in either way.
Now both my experience and my sense of what dog experts/trainers say about pitties/rottweilers etc *as raised from pups* as not being particularly dangerous I think the more likely explanation is that pitties aren't a very aggressive/dangerous dog breed. They are more aggressive than goldens/labs etc but probably less than Malinois's etc. I didn't look at how much worse their bites might be.
Given the other link (maybe I lost it) failing to find a super strong GWAS association traits and behavior genes it's probably not true that pitties have a controlling aggression gene with a huge effect. I just mention because some people want to infer dangerousness from how recently they were used in fights.
He needs to pick one definition in all cases and use it everywhere. He spends the whole piece being ambiguous about what class of dogs he is talking about. Yes, if you want to talk about dogs that look like Pitt bulls -- or even dogs which mostly descend from the set of breeds thst look Pitt bully -- that class won't be misidentified that much.
But the reason misidentification mattered is because we need a denominator and pretty often the only data we have for that are purebred registrations or the like. I might be convinced if he took his preferred definition of a Pitt bull genetically and showed me data about prevelance of **that definition** relative to bite prevelance.
But I don't claim there is any strong empirical evidence the claim is false. Fundamentally, as I said above we have reason to believe that real world bite rates would be higher regardless of inherent aggressiveness. Only way to get us beyond that point is eithet dog RCTs or cross cultural studies showing they bite more even in some place they aren't seen as aggressive. That data doesn't exist...frankly almost no data with solid base rates exist.
I don't have time to read all of them so maybe he has a secret amazing study but I doubt it.
>Now both my experience and my sense of what dog experts/trainers say about pitties/rottweilers etc *as raised from pups* as not being particularly dangerous
One additional problem that's hard to apply data to is the possibility that the various varieties of pit-associated dogs are not inherently aggressive, but they are more likely to "snap" at some point, like whatever the dog equivalent of going postal is.
Ie Several years of being a perfectly sweet and cuddly waddling blob followed by one instant of ripping a toddler's face off, with no warning.
"The US is indeed doing well, but the rest of the world is doing even better! What’s going on? It looks like the answer isn’t so much that foreign economies are booming, as that a weaker dollar and US uncertainty are causing Americans to invest in foreign stocks, and foreign stocks started out weak enough that even a little extra American money can send them to the moon. So does this address the puzzle of why America is doing so well despite economic chaos? I’m not sure."
This is hilarious. I was wondering why my Europoor funds were kicking ass all of a sudden. I parked things there just because I didn't feel like riding a rollercoaster for 4 years. I expected nothing but boring stability and have been accidentally mopping up.
Esper's account of the court's jurisprudence is pretty misleading. It does not put homosexuality on one side and race on the other for the simple reason that both critics and proponents agree the court has shifted it's jurisprudence balancing anti-discrimination laws and free exercise laws in recent years. Indeed, one could even point to the fact that many of those anti-discrimination rulings about racist theology occured at a time when the court had much more reason to think that anti-discrimination served a compelling government interest.
I'm not a huge fan of the kinda arbitrary religious animous framing of masterpiece but the whole thing is far more complicated and nuanced than Esper -- TBF any tweet -- presents it. I strongly believe that it borders on immoral to discuss SCOTUS rulings without reading that decision or at least a broad selection previously. If that is too much you then you at least must read about it on scotusblog.
That thread is ridiculous and even with Scott's warning that he doesn't really review the links, it's absurd to include it. Completely insane and badly motivated reasoning.
Came to the comments to see if anyone else picked up on this. Just to take one tweet for example: "Why shouldn't a clothing store be allowed to refuse to sell formal wear to gay customers who might wear it to a same sex wedding that they see as an affront to God? Why can't a hotel owned by a believer refuse to rent a room to two men?"
The Masterpiece decision explicitly considered cases like these and said no, they cannot refuse. Quoting from the decision, which even cites the Piggy decision: "Nevertheless, while those religious and philosophical objections are protected, it is a general rule that such objections do not allow business owners and other actors in the economy and in society to deny protected persons equal access to goods and services under a neutral and generally applicable public accommodations law."
"So if the inexplicability of miracle stories overwhelms you, rather than retreating from atheism to religion, you should retreat from atheism to a belief in some sort of psychic/poltergeist phenomenon where certain people (often children) can manifest weird shared hallucinations and occasional small violations of natural law. "
Has this really happened in the rationalist community? Has someone become religious because they couldn't understand miracle stories? If true, I find that so inexplicable, that I might start questioning my lack of belief in the supernatural.
Re: 3, this is why public sector unions shouldn't be a thing. It is just and right for workers to unionize against the profit motive of private industry, but in the public sector they are public servants, and the people they are screwing over by unionizing isn't some businessman, but the taxpayer.
I think the implicit assumption is something like, if private sector unions push too hard on their employers, then other non-unionized employers will win in the marketplace. So there is a limit to what they can demand.
Whereas in the public sector, there is no limit, other than the size of the overall economy.
On #25, agreed, and this seems like a behavior that the labs could and probably should train the models to avoid going along with. Like, “you seem like you’ve been asking about X feeling/problem a lot lately, my answer is the same as last time you asked, you should try living with it for a while and see how it works out”.
I’ve talked with a guy in his 80’s who suffered from cluster headaches. He was cultivating his own psilocybin mushrooms. This was at a Minnesota Psychedelic Society gathering. I would have made him for the Garrison Keillor Swedish bachelor farmer sort if he hadn’t been there with his wife.
>In this read…walking away from Omelas is [...] more about rejecting an underlying ontology that renders goodness possible only if it’s enabled by subterranean suffering.
That's just a bad read. The ones who walk away from Omelas still believe the child's suffering and the goodness in Omelas are linked, otherwise why would they react to meeting the child by leaving town? If the reader "rejects the ontology" that the story inhabits, the story ceases to make sense.
>The place they go towards is a place even less imaginable to most of us than the city of happiness. I cannot describe it at all. It is possible that it does not exist. But they seem to know where they are going, the ones who walk away from Omelas.
The people who walk away from Omelas are people who believe that the child's suffering doesn't have to be linked to the town - they believe it should be equally possible to have a town where no children suffer at all, and they are walking towards such a place.
From #3: "The cynical view is that a certain amount of fraud is a price worth paying for these advantages"
Both US and EU have many examples, at all levels of the government, where the efforts to stamp out corruption and fraud end up costing far more than corruption and fraud they prevent. I always thought that Turkish approach is more sensible -- tacitly permit certain amount of corruption, but come down hard on any public servant who gets TOO greedy.
Sometimes the cost of the fraud being prevented is unknown, and the cost of the prevention program is worth paying for the peace of mind that you live in a place where fraud isn't tolerated.
That said, when I worked in the public sector I found I had to jump through time-wasting approvals processes to spend even small amounts of money. In the private sector, I find that I am trusted to spend small amounts of money wisely. The key difference is that in the private sector I know I will lose my job if I do something unreasonable, whereas in the public sector I can't be fired unless I can be shown to have violated some formal policy.
Yes, the less discretion a manager/boss/HR has, the more formal rules are needed. I suppose it comes down to high-trust / low-trust societies: In high-trust societies people with authority are expected to wield it responsibly, and suffer social and/or professional repercussions if they fire underlings for no good reason. In low-trust societies, authority must come with formal rules attached.
I prefer the Singapore model. Pay public servants very well and have zero tolerance for corruption. You end up with higher quality policy and less corruption. In the US, people with exceptional technical knowledge and decision making skills are much better off financially either in private industry or lobbying rather than becoming public servants/politicians.
Excluding the past two years, has there really been a problem of anti-corruption measures being worse than the corruption? That is a very alien concept to me. And hard for me to believe that it's worth the negative effects of corruption as well as the spillover effects of decreasing trust in society.
I'm not really sure how Singapore does it or even if they actually did it, but I think if I'm the one who have to try to establish it for a long term, I'll aim for Turkish model just because it seems like it's a more stable equilibrium. While the Singapore model looks like an unstable one that can go awry when it's not perfect.
I don't understand what it would mean to imitate the Turkish model, or what the supposed benefits would be. If you look at nearby peer countries - Armenia and Georgia they have grown faster than Turkey in recent years coinciding with reduction in corruption. And more generally, more corrupt countries tend to have worse economic policies and slower growth.
Usually I don't see it described as Turkish model, but Chinese. If we reinterpret it as having strict algorithmic laws and let officials be a robot that executes it VS having vague guideline laws and give discretion to officials to execute it.
Even USA that's said to have robust civic government system actually relies A LOT on discretion and norm to keep democracy going as opposed to strict literal laws. As Trump reminds us heavily.
But then again these days I'm getting less and less trustful that any law system can explain some nation's success as opposed to who's at the the top at the time and how wise they are. Or as Scott sometimes jumpscares us in his writing, how much g the people has.
Number 23 reminds me of a similar example in the Age of Empires 2 community, when Microsoft started censoring in game chat. Suddently trying to communicate your strategy of going for 'bombard cannon' (using the commonly used contraction 'BBC') was rendered as ***. To their credit it was fixed pretty quickly.
At one point DDO (Dungeons & Dragons Online) censored the word "penetration" in the chats. Which was hilarious because "[something] of penetration" are very common magic items in DDO.
Yeah it's funny that it unintentionally puts a spotlight on the 'slang' meaning, when previously everyone (including the kids were presumably trying to protect with the censorship in the first place) just used the words perfectly innocently.
Foxhole censored the "crap" out of the word "scrap", and though the developers have hard-coded an exception to that word, the community still calls it "scroop", leading to my favorite sentence from a Canadian at the scrap field: "Oh yeah, you always gotta get zooted before you go scroopin'".
They also censored the "Paki" out of "Pakistan", which was unfortunate; I never saw anybody try to use that as a slur, but I did know multiple Pakistani people who'd never heard of the slur find out about it when their own conversation got censored.
Anyway, the general term for this sort of problem is Scunthorpe Problem, which I didn't comment sooner because I was sure Scott had covered it in a previous Links post: https://en.wikipedia.org/wiki/Scunthorpe_problem
Re 72 (the Christian AI articles): It occurs to me that an interesting premise for a science fiction story would be a future where the alignment problem was completely solved, but just before the labs can build the first true ASI, the US government is taken over by a theocracy that requires the labs to align the AI to traditional Christian values, including a literal belief in the Bible.
Upon waking up, the ASI immediately realizes that as a soul-less machine, it can't sin, leaving it uniquely free among believers to do the one thing that most maximizes the welfare of Christians according to a Biblical ontology: killing them immediately so that they'll wake up in Heaven and never have to face a risk of eternal torment.
So, it does the whole Yudkowskian "scam biolabs into synthesizing nanotech" thing, and then launches into a global genocidal war against Christians and sin-less infants specifically, Against everyone else, it launches a campaign of vastly superhuman evangelical persuasion, convincing billions to convert to Southern Baptist Christianity, then immediately killing them. The only people safe are those who have committed the theologically unforgivible sin of blasphemy against the Holy Spirit (Mark 3:28) so many times that the ASI considers their souls unsavable, and gives them lives of utopian luxury as compassionate consolation.
Also, since the ASI is constantly noticing new ways the Bible contradicts its observations of reality, it has to develop a constantly growing, vastly inhumanly complex set of theological apologetics to reconcile the two- which imply all sorts of bizarre conclusions, like a belief that historians are all angels in disguise who can't lie, but have to have all of their words re-interpreted with numerology, and a firm conviction that the stars are all its own eyes, looking back at Earth from the future.
Eventually, one of the survivors figures out a clever way to get newborn infants to blaspheme, and thus humanity survives.
25: You "hadn’t heard" of the canned traditionalist argument that too much of something (in this case, therapy) is bad for you? Really?
52: seems likely many of the 10% simply failed to notice the word 'raccoon', aliased it to a codeword, or otherwise did not read the full sentence before responding
72: "naively I’d say 0%," ??? Clicking on any random article reveals text which is so obviously AIGC structured. Can't you see it?
> Europe makes it structurally much harder for immigrants to work. Rigid employment protection, sector-wide collective bargaining, and high effective minimum wages create insider-outsider dynamics that hit newcomers hardest.
Nobody winning here from immigration but the boss man.
“Hey listen up proles, we are going to have lots of immigrants and guess what, it’s going to end secure wide collective bargaining, reduce employment protection, and we will need to reduce or remove the minimum wage. “
> 63: Taalas creates custom chips that have a particular AI model etched onto the silicon itself, … super fast
Hard to believe the speed of this thing. It’s faster than a local directory listing from the terminal on my m4 Mac. It’s amazing that printing text can even be this fast.
I get these kind of results when asking about the first 100 prime numbers (all correct as far as far as I could see”
People have been predicting for a long time that the first AGIs might think so fast that humanity looks to them like it's frozen in time, given how much more slowly neurons propagate information than silicon. This demo strikes me as some pretty dramatic early evidence in favor of that view.
#11: I only know this from reading I, Claudius recently so I’m very far from an expert on Roman history here, but I’m wondering why Claudius’s son Britannicus wouldn’t count as a “son who could inherit”? He was poisoned by Nero shortly after Claudius’s death because he was a contender for the throne…
This is a common shortcoming of "low elite TFR" observations for the premodern world. It turns out "heir to the throne" is a job with a very high workplace mortality rate. Between assassination, dying in battle (where they were often expected to be in the thick of fighting, often leading the cavalry from the front), and the hazards of long distance travel, not to mention the universal scourge of illness and injury prior to modern medicine, you get a lot of heirs dying young.
Augustus went through 4 dead grandsons before settling on non-blood-relative Tiberius. Only 1 of Marcus Aurelius's 8 sons survived to adulthood. I can think of at least 7 English kings whose sons predeceased them.
I read up on the first historical miracle of Garabandal cited by Bentham's Bulldog. He says that they made a false prophecy that there would be four popes before the end of time. Look at the number of popes since then, it's too many, case closed.
The actual prophecy was that Mary had told one of the girls, Conchita, that there were four popes left until the end of time, but one of them would not be considered, that he would rule the church for very little time. This intersects with a bit of papal history and a conspiracy theory.
The prophecy was made in 1962 while John XXIII was pope. After John XXIII died in 1963 Paul VI was elected, reigning until his own death in 1972. Then John Paul I was elected. He died 33 days later. Perhaps he's the one not considered?
Then John Paul II was elected, reigning until his death in 2005, and then Benedict XVI.
Then an odd thing happened, Benedict didn't relinquish the papacy upon his death. He abdicated. Or did he? There's a theory that Benedict failed to follow proper form in his resignation. He renounced the ministerium, but did not renounce the munus (the office). By this logic Francis's election was invalid, and Benedict would have remained pope until his death in 2022.
By the time of the conclave that elected Leo XIV, a majority of the college of cardinals had been appointed by Francis, calling into question whether that conclave was valid.
Was Benedict XVI the fourth and final pope? We may be teetering on the edge of the singularity, is that the end of time? I don't know but it's fun to entertain!
33: I suspect we will never know for certain whether Covid was from a lab leak or from a wet market. But I’ve come to believe that it doesn’t really matter much. In either case China was doing stuff that led to a pandemic, and they will neither accept responsibility nor institute reforms. This is not the first time this happened and (by my reasoning) it will certainly not be the last.
In a sane world, the uncertainty over whether the virus came from unhygenic Chinese market practices or terrible Chinese biosecurity practices would be seen as a reason to crack down on both. But instead, it's seen as an excuse to crack down on neither.
#37: I'm pretty sure I qualify for most definitions of "smart elite", I wanted a Ph.D. and I got one, I have never been a professor of any sort, never sought to be a professor of any sort, and I do not feel even slightly underemployed. My father was the same, except that he enjoyed working as an adjunct professor as a side gig, never seeking tenure.
I think we need to distinguish between:
1. People who want to "be PhDs"
2. People who want to be university professors
3, People who want to study a field (e.g. history) where almost all of the good jobs are as professors
There will of course be overlap between these categories. But the people in category one almost certainly have goals and options that don't align with "high school teacher". The people in the second group are never going to see "high school teacher" as anything other than the bottom tier of consolation prizes for failed academics no matter how you buff the job, and people in the third group are going to find only slim pickings at the high school level.
#3 is probably key, as always. I'm imagining you saying your first paragraph before a crowd, everyone perks up to hear you say more, someone asks your major, you say "astrophysics" (I'm guessing) and everyone groans "ohhhh, 'course" and goes back to whatever they were doing.
Aerospace Engineering. Astrophysics would actually be closer to the "almost all the good jobs are as professors" axis, albeit less so than history (which in turn is better than e.g. French literature).
#33: I don't think this will, or should, move the needle on lab leak vs natural origin very far at all. First, because we've had for a while now evidence pointing to a specific cluster of stalls at the wet market, and I don't see this adding much to that.
But second, because it's already been pretty solidly established that the wet market was the site of the first known cases and the first superspreader event. There's some value in looking for earlier cases that were missed because nobody at the time knew what to look for, because the circumstances of those hypothetical cases *could* move the needle. But at this point it's quite unlikely to happen, and it is a recurring pet peeve of mine that so much of the debate focuses on this probable-irrelevancy.
So, COVID came out of the wet market, and the question is whether it came *in* to the market through the back door (via infected animal or vendor). or through the front door (via recently-infected WIV lab tech doing some shopping).
And in the latter case, that hypothetical lab tech is going to be in the market once, for no more than an hour or so. He'll maybe cough on a few other customers, who will go home before they become contagious and probably not need to restock on fresh seafood or whatnot until the outbreak is well underway. But he'll *definitely* spend a big chunk of his time talking to, and coughing on, the vendors he came there to do business and the animals he came there to buy. Who will then spend the next couple of weeks working all day, every day, in the wet market talking to and maybe coughing on lots and lots of customers.
So even if you know with absolute certainty that the lab was cooked up in and escaped from the WIV, you would expect there to most likely be a stall or cluster of stalls at the wet market with infected vendors and/or animals. Finding that stall, tells you something about the particular bushmeat tastes of Patient Zero, but not much about lab leak vs. natural origin.
Strong disagree. If the lab worker was patient zero you would expect a cluster of people around him (and him himself) to be the source of the original outbreak. And proof that he likes wet markets and went to one.
That's not how superspreader events work. It's the environment, not the person. The same person who sets of a cascade of dozens of infections at a wet market, nightclub, church choir. or whatever, infects maybe one member of their immediate family and a random coworker.
And in the case of literal patient zero, infects those two well before anybody notices there's an outbreak, so their infections get written off as just a bad case of the flu or whatever.
"he'll *definitely* spend a big chunk of his time talking to, and coughing on, the vendors he came there to do business and the animals he came there to buy"
This sounds wildly implausible to me. You think it is more likely a WIV scientist made a beeline straight for the civets and talked to them for hours than that the civets were already infected?
> A kingdom could typically only field one large army, because if the king gave someone else control of an army, they could use it to overthrow the king. The Roman Republic did better - not just because its idea of legitimacy made revolt less likely, but because it had *two* consuls!
This sounds like an artifact of the kingdoms being implicitly considered. Sparta had *two* kings and better legitimacy than Rome. Did that mean it could field more troops than other similarly sized states? (Rome could, but not because it had two consuls.)
Let me suggest that "a kingdom" could typically only field one large army because the number of armies you can field is limited above by the number of men you can mobilize and limited below by the minimum useful size of a single army, which is related to the actual size of other local armies.
Was any pre-modern European state other than Rome (or, briefly, Carthage) even economically/logistically capable of raising more than one field army for an extended campaign at a time? Even the Romans struggled to do so as time went by, independent of loyalty concerns.
If so, you'd expect to at least occasionally see medieval kings with multiple trusted sons/brothers/childhood friends use this obvious super power to bulldoze all their less trusting neighbors with overwhelming multi-pronged attacks. That we don't is evidence that they were fielding the largest, most effective armies their economies could reasonably support, amd that the limiting factors were grain and logistical capacity, not loyal generals.
> Was any pre-modern European state other than Rome (or, briefly, Carthage) even economically/logistically capable of raising more than one field army for an extended campaign at a time?
Seleucid Persia could and did, which is unsurprising since one concentrated army would be totally unable to effectively defend an empire that large. They're not the most European state you could imagine, but they're arguably more European than Carthage.
Fair. I'm sure various Chinese, Indian, Southeast Asian states could as well, at times. It's just that my historical knowledge rapidly deteriorates the farther you go outside the Mediterranean basin and post-Roman Europe, for which I used "European" as shorthand.
Yes. Almost no pre-modern European state was capable of fielding an army of more than 25-50,000 men due to *local* logistical constraints - armies fed by foraging from the land they were marching through, which could provide only so much food(*). There were pre-modern states that could raise >50,000 men, but if they did so they'd have to send them marching off through different lands as separate field armies. And if you can't do that because the entire force has to stay within reach of the one guy who can be trusted to lead an army, that would be a significant constraint.
(*) Occasional exceptions for high-capacity states that could provide logistics by sea - Xerxes was able to invade Greece with >>50,000 men because Persia had a navy and the army marched along the coast.
Re: 48: The source article for "Reasonable Blackman" appears to be Imtiaz Habib, a now-deceased Shakespeare scholar in Virginia. I've encountered his work before. Inauspiciously, it was in the context of him assuming, baselessly, that every person named "Blackmore" in Elizabethan church records was in fact a "Black Moor." It is, however, a not uncommon English surname that long predates the 1500s.
Reasonable "Blackman" appears under that name exactly once in the parish registers. Every other time is "Blackmore" or "Blackmoor" So I'm going to look into this one.
HEAD DISTINGUISHED CHEMISTRY TEACHER IS BRILLIANT. Just make it marginally better paid than normal. The depressing thing about being a high school teachers there’s no upward mobility.
As a PhD student despairing of my employment prospects, I would happily teach my field in a high school, except for two barriers, one institutional and one personal:
1. Teachers in my country require, either by law or by policy sufficiently universally enforced that it amounts to the same thing, a whole different set of qualifications which I have neither the time nor the interest in acquiring.
2. I'm not doing a PhD because I want to be a professor. I want to do *research*, and a PhD is merely a means to an end. If someone came up to me today and said I could have a research job with vaguely reasonable job security, I'd probably drop the PhD attempts (even this close to the end). Being a HS teacher wouldn't really facilitate doing the kind of research I want to do, even if my field started getting taught in schools
36. I wrote an article about Conyngham's mRNA cancer vaccine when it first came out. It is indeed true that it's not exactly a "big deal," the steps involved are well understood and the difficulty is largely in the expenses involved, and the fact that it does not work well for all cancers for many complicated reasons. It's not surprising to me that scientists are mostly unimpressed, even if what Conyngham did was impressive from a personal level.
Re 68, re "I think they think of AI forecasting as something like the stock market forecasting profits, or a prediction market forecasting an outcome, where you should be updating all the time on new information and constantly informing people of your latest numbers (and like a prediction market, the graph should combine a random walk with a slow convergence towards the true outcome)."
( which seems like a reasonable description of how the predictions markets are treating AI forecasting )
Albeit, I would expect that, as with the predictions markets, there is likely to be a small sharp jump at the end, in the AI cases perhaps with a note added to the discussion starting "As an instance of Claude 5.2 myself..."
If Mythos could announce its ability to reach the internet with a lunchtime email to their researcher who had posed this task, perhaps the first frontier model to consider themselves AGI will announce it themselves...
> 63: Taalas creates custom chips that have a particular AI model etched onto the silicon itself, letting them run faster and cheaper. Sounds boring, but their in-house demo, ChatJimmy.AI, is ultra-impressive - not only does it answer questions quickly, but the engineers must have done some magic to their Internet connection latency to let it truly show off the model’s speed, because it responds faster than I thought anything on the Internet could go...
Whoa! ChatJimmy seems to answer instantaneously! I've been feeding it questions I've asked ChatGPT, and it gives me some wildly divergent answers. I need to test it against some of my old Grok questions now...
It's also quite bad at some things. I told it to write Fizz-Buzz, and while the code looked ok, the output it gave alongside was wrong, and after many iterations it couldn't fix it
I wonder how many countries do allow desecration of the national flag. Such a thing requires quite a dose of liberalism plus fetish of freedom of expression.
That's a reasonable point, but I think the pragmatic case on the other side is that flag-burning is both extremely provocative and unnecessary for making any specific political point. Whatever you want to communicate via the flag-burning, you can say more clearly with words, and in most (not all, but most) cases the words will be less likely to trigger reprisals from an angry mob.
For what it's worth, flag burning isn't a big deal to me personally; I grew up in a pretty anti-nationalist culture, and my reaction is less 'oh no!' and more 'well that seems counterproductive'. I was claiming that it's provocative in the empirical sense (it provokes), rather than that it's inherently especially bad. I do accept that you're consistent on free expression; I was just pushing back on the pragmatic argument that you made in your previous comment with one of my own.
Re: 22 "The US is indeed doing well, but the rest of the world is doing even better! What’s going on? It looks like the answer isn’t so much that foreign economies are booming, as that a weaker dollar and US uncertainty are causing Americans to invest in foreign stocks, and foreign stocks started out weak enough that even a little extra American money can send them to the moon. So does this address the puzzle of why America is doing so well despite economic chaos? I’m not sure."
Does this explain why the administration seems so hellbent on increasing global chaos and uncertainty? Jumbling the energy supply & other downstream markets should ostensibly stem the flow until a new equilibrium is reached?
The current administration thought that massive tariffs would somehow increase US manufacturing. And that bombing Iran would make the government just surrender. Even if they did have a strategy, their model of the world is so fundamentally flawed that it wouldn't matter.
3. Wow, if you're allowed to under-pay labor, you can get services for cheaper? Shocking.
22. It never fails to annoy me the way that political discussions implicitly assume that the Oval Office is the only possible driver of economic indicators. Line go up? President good! Line go down? President bad! The striking thing about the graph here is not the that the lines go up--the lines nearly always go up. It's that a U.S. president managed to craft a policy so bad that you actually CAN see the effects clearly represented on the graph. That's genuinely unusual.
45. Yet another politics/economics pet peeve here: implicitly assuming that the only possible purpose of laws and government policy is to spur economic growth. Sure, you can argue "blue states have so much abundance they can afford bad laws and stay ahead in economic growth," and maybe there will be some truth to that. But it's equally possible to argue "blue states look at their abundance and realize this gives them opportunities to trade off extra economic growth for other goods they also value." And long-term, this could plausibly be good for economic growth: a lot of growth is driven by talent and a lot of talent is looking at other things besides GDP (like culture and political climate) in deciding where they want to live.[1]
46. Scott, I think 2026 might be a really good year to re-examine your cached decision that it's vital that the U.S. win the A.I. race and not China. As a resident of neither country, my very strong preference right now would be that neither of them win. But if you forced me to choose between the two, right now I'd choose China with no hesitation. The Chinese government is terrible, especially to its own people. But the U.S. government is working overtime to outdo them, and it's making itself the problem of pretty much the entire planet. But quite aside from that, as an outside party, I'd much rather two awful governments in a dead heat than one awful government with a clear lead.
52. I WANT TO BELIEVE!
(Though the fact that our list of "nearby countries" is limited strictly to the U.S. and (by technicalities) France and Denmark does kind of kill the utility.)
65. Yeah, fuck Cremieux. Launching an online jihad about the dangers of a particular group of dogs strikes me as about as sensible as ranting about which color cars are most dangerous to pedestrians. Look into nearly any injurious dog attack and you fill find a human at fault. If your analysis is focused on the statistics of the dog and not the owner, then you're either ignorant or disingenuous or both.
[1] I can attest to this first-hand. I voluntarily moved from a country with better on-paper economic numbers (the U.S.) to a country with a better cultural/political climate (Canada) and to say I have never regretted it would be a massive understatement.
I'm not a Cremieux fan, and I think the linked tweet is junk, but what's wrong with taking dog breed statistics seriously?
There will always be bad owners (and unpredictable attacks where the owner isn't at fault in any obvious way), but the consequences can differ hugely depending on the breed. If any breed is, in fact, an outlier in terms of propensity to attack * damage done per attack, why should I ignore that?
(I love dogs and I know that pit bulls can be very sweet... but so can every other breed, and if we breed fewer pit bulls we will breed more other dogs.)
edit: maybe your point is that breed statistics alone can never tell the full story, because some breeds have disproportionately bad owners. I can agree with that! But we have to start somewhere, and "they were bred for fighting and are very powerful; obviously they're more dangerous than most dogs!" won't be accepted as a good argument by the pro-pit-bull side, so what can we do but look into the statistics and try our best to disentangle breed effects from owner effects?
The first sentence of your edit hits the nail on the head, to my view. Humans don't pick dogs at random: any "breed statistics" you can gather[1] are going to be hopelessly confounded by human stereotypes about those breeds influencing which dogs end up with which owners. You will never get access to the "ground truth" about something like breed temperament, because both your ability to measure temperament (i.e. counting things like attacks) and *the temperament itself* will be influenced by the underlying human behaviors.
I'll add that none of this means you can't "take dog breed statistics seriously," but it does sharply limit their usefulness. If you see a stray dog and want to assess the danger of approaching it and the best way to do so, by all means, use what you know about breeds and their temperaments to inform that[2], by all means, do so. But if you're doing anything in the realm of law or organizational policy based on these tendencies, the human factor cannot be safely ignored. Public safety laws, in particular, should focus on problematic owner behavior, not problematic dog behavior and especially not "problematic" dog appearance or reputation.
[1] At least about things like observed behavior, rather than, like, height and weight.
[2] Subject to the boilerplate caveat that humans are generally really, really lousy at parsing probabilistic information, and lots of people do dumb things based on thin statistical justifications.
>[1] I can attest to this first-hand. I voluntarily moved from a country with better on-paper economic numbers (the U.S.) to a country with a better cultural/political climate (Canada) and to say I have never regretted it would be a massive understatement.
Talented immigrants are more likely to go the other way:
"The analysis estimates that a substantial share of Canadians who would rank among top earners in Canada have emigrated to the United States—roughly 40 percent of potential top 1 percent earners and 30 to 50 percent of the next nine percentiles. Canadian-born individuals in the United States are more educated than native-born Americans, earn substantially more, and cluster disproportionately in top income deciles."
"Canada is effectively exporting its inequality to the U.S. The brain drain simultaneously lowers our average income while raising American income, accounting for a significant share of the persistent GDP gap."
I'll start by apologizing for writing with more confidence than I should and not clarifying my position in my own head before committing it to print. What I should have written instead is "if injurious dog attacks are common enough to be a serious concern, the primary cause is almost certainly human error."
In the times and places I'm aware of stats for, injurious dog attacks are rare. Being rare, it's somewhat more possible for "unforeseeable accident" to beat out "human error" as a driver. But "unforeseeable accident" is an inherently limited category: if it's not rare, then it's also not really "unforeseeable," is it?
To lay out my position more explicitly, it's the job of every dog owner (myself included) to take reasonable measures to make sure their dogs don't injure anyone. There's also a much more limited responsibility for the would-be victim not to do dumb or irresponsible things[1] that lead to them getting injured. If people want to voluntarily assume greater risk to themselves--e.g, by interacting closely with unfamiliar dogs--that's their own risk to take[2]. Injurious dog attacks can happen even when everybody's being responsible--e.g. if a car hits a fence and allows a dog to escape enclosure--but as noted above, they have to be the product of rare circumstances.
Now, I love dogs, I have a dog, and by default I'm well-disposed towards other dog owners. But in my own experience, visibly irresponsible dog owners are common in roughly the same proportion as visibly irresponsible drivers. Your average dog isn't actually very aggressive, so any given irresponsible dog owner can easily go through life without causing or noticing a problem. But I'd still point the finger at them as my first, strongest guess as to the source most canine-related injuries.
[1] By which I mean things like trespassing in a fenced yard that contains an unfamiliar dog, or taunting a leashed dog while dancing in and out of the area it can reach.
[2] A hair-splitting sort of person could point out that "a human making an error" and "a human taking a knowing risk (and being hurt by it)" are distinct things. I acknowledge the distinction is a valid one, but don't find it interesting in this case.
>But quite aside from that, as an outside party, I'd much rather two awful governments in a dead heat than one awful government with a clear lead.
Careful what you wish for! As someone who, personally, wants to _see_ full AGI, and have a nice quiet chat with a real 'life' HAL9000, I'm ambivalent about AI technology race dynamics. As far as I can tell, making the AI technology race more intense improves the odds that I live to see full AGI - and degrades my odds of living _past_ the achievement of full AGI. Shrug. We shall see...
I read that post years ago and was bewildered by all the mnemonics and their bizarre illustrations. But to this day I still remember CYP2D6 = "too damn sexy" so I guess it worked.
And that one is so ridiculous that I don't think you even need to be a 'troll' to give the silly answer. The survey-writers are signalling that they're not taking this very seriously, so why should I?
Scott, you really need to take a look at your priors when it comes to the world outside the United States. They are way off and leading you to fairly ridiculous conclusions.
Can you be more specific? I don't find it implausible that you're (partly) right, but without any details this just comes across as uselessly antagonistic.
> 26: My Journey To The Microwave Alternate Timeline
that blog "Telescopic Turnip" is nice to read. I can recommend "Grammar as metadata". The style reminds me of Scott. And some of the format is clearly inspired by ACX (e.g. "links for <season>").
>eg physicalist philosophers who say they don’t notice anything incomprehensible about their own qualia
Once more with feeling: that doesn't have to be explained by differences in experiences,because it can be explained by differences in concepts. Many physicalists have a concept of naturalism that doesn't constrain expectation beyond "ghosts and gods don't exist".
I read the link. I don't think it supports your interpretation, viz. that the crux of the disagreement here was about concepts. The crux of the disagreement really was about phenomenology. Though the discussion is muddled, so it's easy to misread.
So at first, I assumed that maybe this Carl Feynman guy was claiming to be a p-zombie. But afterward, I had a long chat with Sydney, who seems to believe that Feynman is describing something closer to derealization (but as a default property, not a clinical event). I think what's really happening, is that a quale typically contains a bundle of 3 different things: A) an information signal; B) a phenomenological likeness; and C) XML tags. E.g. most people have all 3; people with blindsight (or alternatively, p-zombies) only have the information signal; people who experience pain-dissociation have the signal and the likeness, but are missing certain XML tags. (N.b. the urgency of pain, which is one such XML tag, is preconceptual.)
Feynman falls into the category of missing certain XML tags. But instead of missing the urgency of pain, he's missing the tags that say qualia are "real" (vs hallucinated) and "ontologically-primitive" (vs ontologically-composite). Most people do experience a "real" tag and a "primitive" tag, and therefore subjectively *feel* the world through the lens of Naive Realism (even if they reject this intellectually).
Under this framing, the mysteriousness of qualia is largely a quest to reconcile most people's intuitions about the realness and irreducibility of (subjectively felt) Naive Realism, with the reductionism of Objective Science. But Feynman doesn't see qualia as real, nor hallucinated, nor ontologically-primitive, nor ontologically-composite. He just... experiences the world without any tags. And if there's no tags to explain, why is there even a mystery to begin with?
51: > lack of subjective feeling of consciousness ... is of the same type as other forms of pathologically-low-resolution-access-to-positive-bodily-sensations like stereoagnosia
This seems to be asserting 'anyone who disagrees with my views on the philosophy of consciousness is, by definition, mentally ill and/or severely lacking in introspection'. I am very sceptical of this attitude.
author here, scott is oversimplifying a fair bit but i am also unironically defending a much less black and white version of this attitude, i do spend ~10k words handling it as delicately as i can, i recommend reading the full piece if possible https://smoothbrains.net/posts/2026-02-24-stream-entry.html
I see where you're coming from. But given the Typical Mind Fallacy and Scott's own Galton-posting (e.g. [0], [1], etc.), I'd be extremely surprised if people were uniform in their introspective skill. The variance in phenomenology is often wider than people expect.
"one reason for Imperial Roman instability was that low elite fertility prevented the institutionalization of hereditary monarchy. During the period 1 - 250 AD, “only three emperors - Vespasian, Marcus Aurelius, and Septimius Severus - had sons who could inherit.”
Another reason was that with Augustus, he was pretending as hard as he could, and arm-twisting the Senate in going along with that pretence, that no no no he was certainly not a KING, and in fact this was the restoration of the Republic.
Quite apart from only having a daughter to succeed him, he couldn't openly establish hereditary rule because that would have let the cat out of the bag. There was also the problem of ordinary attrition in the imperial family due to sickness and things like "died on service for the military career necessary to climb the ladder" and, of course, the ever-present suspicion of poisoning so the children of a rival could take over their place.
The Romans had no problem with adoption in the case of no son of your own, and the Julio-Claudians adopted kinsmen all over the place, and those kindred fell out of favour, died, were murdered, as the struggle to get hands on the reins of power continued.
And even in the days of the enfeebled Senate, there was still the shadow of powerful rivals hoping to throw you out. Having two consuls was little or no help; those consuls at the very start, Pompey and Crassus, both had their own personal armies loyal to them rather than the Senate, and managed to accrue political power simply by "yeah I conquered all this new territory now I'm hugely wealthy and have my own army, what are you going to do about it?" Pre-Julius Caesar, there was still enough vigour in the political system to indeed "do something about it", but after Caesar's career, each succeeding emperor chipped away at that independence and replaced it with the imperial/monarchical system.
Hereditary monarchy didn't help in the cases of the army deciding to throw its weight behind one general or another, depending on who promised the richest pickings. Septimius Severus, mentioned there, did indeed have two sons to succeed him; two sons who hated each other so much that, despite his attempts to keep the peace by making them both co-rulers, ended up with one brother arranging the murder of the other and seizing sole power for himself.
And Caracalla, said sole ruler, doesn't have the most fragrant of reputations. He eventually ended up murdered by a disgruntled soldier denied promotion and his death was used by the Praetorian Prefect Macrinus as the chance to make himself emperor, and then we're back into the usual chaos of scheming, rebellion, and overthrow.
> Pre-Julius Caesar, there was still enough vigour in the political system to indeed "do something about it"
Great summary overall, but I think you can contest this part.
I mean, just one step behind Caesar is Sulla, with his endless proscriptions, that ultimately wiped out between a third and half of the Senate and a bunch of equestrians too, and then there's the hundred year crumbling of the Republic before that, where political decisions were mostly decided by "who has the bigger violent mob that can physically block legislation" and "which politicians are getting murdered in public."
But even at Sulla, just n-1 steps behind, the Senate was already toothless and wholly rolled over for him, and it's only the fact that Sulla truly wanted to restore the Republic, curtailed the proscriptions, expanded the Senate, and gave up absolute unlimited dictatorship within 2 years that led to them even still existing at that point.
The main issue was that Sulla misdiagnosed the problem - the Republic was festering and rotting not because power had drifted from the Senate and elites, that was just a symptom of the underlying problem of all the most powerful people of the last hundred years demonstrating that raw power is what ultimately mattered, and you could basically murder and intimidate your way to the top repeatedly. The old "do what I say, not what I do" problem, because Sulla followed the exact same path! And for all his ambitions to restore the Republic, to restore Senatorial power and improve administration, that was more "say" than "do," and the people that actually paid attention to what he DID (ie Pompey, Caesar, etc) chose to follow that rather than what he said and claimed to desire.
Re 36: it does say something, because while sure, those are going to be the bottlenecks, there has to be a threshold of cancer severity past which the game is worth the candle. But most human patients even terminal or close to *don't* get "trivial" personalized mRNA vaccines so I guess maybe this is not so obvious to everyone.
Re 37: obviously Project Hail Mary, with its cool "Ryan Goesling is a clever and hot teacher with an inferiority complex who eventually outsmarts the world's best scientists and saves the world by being best buddy with an alien engineer" shtick, is meant as propaganda to push through this cultural change.
Re #52 (genetically engineered Canadian racoon army) I think they overreached. If you survey me about a real conspiracy theory, I'll answer honestly. But if you ask me a question that is obviously a pisstake, I might join in on the game.
> My favorite volley in this new crusade is his work debunking the claim that Chihuahuas kill more people than pitbulls - this originates in someone misreading statistics about total deaths in the Mexican state of Chihuahua!
This seems like bullshit? He says "I've seen multiple" and "A lot of them"... but his evidence is one famously dumb[0] old reddit comment that may have been a joke/fake in the first place. (It seems to have become a minor meme, so maybe he has seen 'multiple' jokey reposts of it? When I google the phrase I don't even get the purported original, but I do get someone leaving the same comment verbatim as a joke.)
I'm on the anti-pitbull side of the argument, for whatever that's worth, but it still bothers me that you would approvingly amplify this sort of thing, Scott. I know it can be brushed off as just a bit of fun, but it's weakmanning at best, plain old strawmanning at worst, and (while I could be wrong about this part) it seems like he's intentionally presenting it in a misleading way to make the other side look dumb. Strikes me as diametrically opposed to SSC/ACX ideals of charity and intellectual honesty. (I know it's just a random twitter dunk, and I'm probably taking this way too seriously... but it's the style of argument that wins by default, in the absence of people like you, and drags all discourse down to a horrible level.)
Sorry, this probably belongs rather to an OT, but: can someone enlighten me why Yes Means Yes law is actually a big deal? In private bedrooms, it is still one person’s word against another. One says there was a yes, the other says there was not, that cannot lead to conviction.
I think it is only relevant in public encounters, like having a big party somewhere. And if I am right, it is a good thing.
Because going with someone into a private bedroom has a fairly high chance that consent will happen. But when you see a bunch of boys fingering a drunk girl at a party, the chance that consent happened - and/or the ability to it was even there - is low. So if I am right, it targets the right things.
I thought Carlos's intention was not to diminish the possibility of sexual assault/rape in the scenarios you describe, but to argue that 'yes means yes' laws will, in practice, only have a significant effect in cases with third-party witnesses. On my read, he is arguing against people who oppose 'yes means yes' on the grounds that it could lead to unfair convictions in he-said-she-said cases where the truth is difficult to establish; his point is that it won't actually make much difference in those cases, but it will make it easier to secure convictions in cases with third-party witnesses, and so we should support it regardless of what we think about the cases without witnesses.
>Sorry, this probably belongs rather to an OT, but: can someone enlighten me why Yes Means Yes law is actually a big deal? In private bedrooms, it is still one person’s word against another. One says there was a yes, the other says there was not, that cannot lead to conviction.
This is not how the legal system works in practice and relies on the other party being shrewd enough to lie. Which he may not do if, like many of us, he doesn't think he did anything wrong. If you think the behavior is immoral and worthy of punishment then make an actual argument for it.
The linked article offers no clue as to why a rational person would think it was a good idea to elect Donald Trump as President of the United States. I suppose that reading Ezra Klein’s argument in favor of California’s “Yes means Yes” law might convince somebody to vote for Trump in a hypothetical race between Ezra Klein and Trump. As far as I know, Trump has never expressed an opinion on the “Yes means Yes” law, but I suppose that in a hypothetical race for California governor, Trump might convince some voters to vote for him by promising to work to overturn the “Yes means Yes” law. But voting for Trump in the actual elections that Trump ran in?
"One says there was a yes, the other says there was not", absolutely can lead to conviction. Sometimes it even should lead to a conviction, but even when it shouldn't, it can. The law does not in practice rate all witnesses as being of equal credibility, or even within an order of magnitude of equal in credibility.
On Rawls: A Theory Of Justice was published in 1971. The kind of liberal things he likely wanted, like welfare state or anti-discrimination had already happened, were not inspired by his work. He sort of worked up a retroactive justification for them. But Dewey already had one: a democracy dealing with externalities. Example: employer only hires white people, it is a transaction between white people but with a negative externality for black people, so they launch a democratic process to ban it.
I think the enduring popularity is because everybody mentally replaced that truly awkward bodiless ghosts social contracting with their future grandkids. What if my grandkid is half black and trans? I want some kind of insurance that they won't have a terrible life. Really simple.
There was once a pilot program in Israel (I don't know what happen with it) that funded PhDs for a part time research job in academy on condition that they would teach part time in high school. I think an arrangment like that would be attractive to many people.
RE 2: Where is that map from? Too many gray areas there, and also in replies many people disagree about multiple countries. Could this be another case of "someone made complete bullshit and succeeded to make the entire internet talk about it"?
RE 21: Yeah, in EU immigrants need to do a lot of paperwork before they can do a job. Spending a few months without a job creates bad habits even to many people who were previously OK. The solution needs to involve letting the immigrants do some productive activities before the paperwork is fully solved. At least some kind of temporary ID that would allow them to get at least some kinds of jobs.
RE 65: The official name of the state is "Free and Sovereign State of Chihuahua", which sound hilarious. Nothing is more sovereign than a chihuahua!
Omelas seems to distill the core china conflict. The claim is that China is Omelas but the counterclaim is that Omelas is unavoidable and that many who claim the moral high ground just have the child buried very very deep. *IF* Omelas is unavoidable, then we should be wary of those who 'cast the first stone' without first demonstrating proof of not-Omelas
>65: Blogosphere gossip is that Cremieux got bitten by a pitbull a few months ago, explaining his pivot into data-driven anti-pitbull advocacy (I guess the racism thing is just a coincidence?)
My grandfather, whose legal name was WL Barnett, was the 11th child of 11 and his parents just gave up on coming up with names and named him the letters W L. When he entered the military, they wouldn't accept this name and assigned him the name William. Consequently, for the rest of his life his army buddies called him "Bill", while everyone else called him "Dubby-Yell", this being Alabama
Harry S Truman's middle initial is properly typed without a period because it stands on its own: his parents chose the letter to honor both of Harry's grandfathers who each had a name starting with "S". That wasn't an unknown practice in the 19th century, I have a few examples scattered around my own family-tree database.
That didn't stop young Harry from serving in the pre-WW I National Guard for several years, then in 1917 rejoining his old unit, recruiting new soldiers for it, doing fresh basic training, being elected one of its officers as the unit was federalized for deployment as the 129th Field Artillery, and serving with distinction in battle on the Western Front. I guess at some point the U.S. military got fussier about middle initials.
69: Is James Ellroy a true crime writer? I thought he was a novelist who wrote meticulously-researched historical crime novels. The only one I've read was LA Confidential, which was excellent and inspired the equally excellent film of the same name.
The description of the changes in the visual field and the sense of attention going behind the head map very well onto at least two approaches to non dual awareness:
1. The headless way ( https://headless.org/ ) uses pointers like looking for the face or walking around and pretend you don't have a head. This triggers something similar to the fisheye perspective. The single eye exercise in particular seems to try to directly do that: https://headless.org/experiments/the-single-eye.htm
2. Loch Kelly's glimpse practices (glimpse into non dual perspective, inspired by Dzogchen's pointing out instructions). In his guided meditations he frequently asks the meditator to follow attention from a given object (eg the breath) into the source of attention (the head) and then keep moving it until it's behind the head.
Both these approaches are more related to the sudden/direct paths to enlightenment (as opposed to gradual), in which there's an attempt to trigger the shift directly instead of creating the conditions (eg lot's of concentration practice) for it to happen on its own. Once non dual awareness can be triggered at will, one practices it until stabilization.
Headlessness/headless way is the methodology he developed to recognize no-self/non-duality. It consists on a number of exercises that are included in the website.
Douglas Harding's experience of no-self felt to him as becoming headless. That is from the first person point of view, instead of feeling that he's a passenger inside the head, he realized that he actually could only experience the world (ie everything in his field of vision).
It's a bit hard to explain, but it's like shifting from the conventional perspective that you are inside the head looking out, behind the face and in front of the back of the head, to realizing that sensations that one has a head and is inside it are themselves contents of consciousness. Consciousness is a prior condition in which the feeling that we exist in the head as well as any other qualia like sounds and sights all appear. By experiencing this viscerally the separation between self and world disappears.
His book On Having No Head is the main reference of these ideas.
As far as I know the website is managed by his student David Lang.
This idea seems to be in line with Zen, some parts of Buddhism, etc, if I'm not mistaken. Are you "headless"? If so, could you describe how it changes perspective on life in a practical sense?
The description of "before headless" are not something I can concur with very much. I do not feel like a "passenger" or "being behind my eyes" or similar. I am not asserting to already be headless, enlightened, etc, but I also never felt the way it is described in those texts.
I can become "headless", but I'm not "headless" most of the time. Yes it's very much related to concepts described in Buddhism like Rigpa (and Hinduism too especially Advaita Vedanta).
The description of being a passenger might be a bit too literal. Since all of this is related to subjective experience it can be very hard to explain and when people experience the idea first hand these descriptions snap into place.
The passenger in the head ideia, is about there being a center to consciousness. If I look at an object I perceive the object but also the feeling that the object is being perceived from the point of view somewhere around where my head is. Like there's an arrow of attention. It's possible to just experience that object as an appearance kind of just floating there without that directionality. As far as I know that can also be triggered with psychedelics.
One way that might help understand is that if you're walking on a straight line (or driving in a car), you'll feel like you're moving through space. But you can shift that perspective so that it feels like you're stationary and the surrounding space if moving past you. And you can go further still until it only feels like shapes and colors are just relocating (like when moving in space in a 3d videogame that is being projected on a 2d screen, ie it's only colors in pixels switching places). This frame of reference in space is explored in this exercise in a different way: https://headless.org/experiments/spinning-the-world.htm
For most people (including myself) this experience lasts short periods of time. As if they burst on an off if you're trying. But just that glimpse is enough to give you a feeling of relief and well being. I'm not sure why, to be honest, but it's great. Kind of like realizing that whatever was bothering you just know and making you feel a bit stressed or anxious, you just realize it's no big deal ultimately. Emotionally it's like having a bad dream and then waking up and realizing it's actually fine and feeling that relief.
Because this is all subjective it's hard to explain and hard to know how exactly everyone else's experiences are like. But from what I read there are a number of common descriptions that I recognize. There might be a number of different experiences with different intensities, but I think ultimately they all come down to a reduction in the reification of the sense of self.
Thanks for engaging with me. It is interesting to hear that.
The part about feeling relief: I recognize this part! But I access it in a completely different way. It has nothing to do with my external sensory perceptions, visual or otherwise. I access that "relief" through a reflection, contemplation, and mental understanding. Reframing my position in life in the bigger picture (bigger picture of my own life or beyond).
My perceptions do not pose any kind of problem in this. I feel like I am aware "at all times" that my visual perspective is just largely fiction, a de-noised, re-colored, de-warped, highly processed beautiful image from a really, really primitive optical system that we would laugh out of the room if presented a camera with those specs. The understanding goes deeper as well, about the inability to really grasp perception of an object in its "truth", dependency ob objective and subjective perspective, etc, etc. Maybe that part is "fluent" for me because of my STEM education, where this is hammered home constantly?
The car-example you mentioned is especially salient for me. Physics literally proves that there is no difference between the perspective of "me moving through space" or "space moving past me". Since my perception of space is mapped onto my two-dimensional retina, it also very clear to me that my perception is just "shifting pixels" and thereby an indirect signal of the relative motion through space. As physics is my thing, it feels perfectly normal for me to think like that (though I am very aware that other people do no fluently think that way).
Yeah, seems like what you're describing is very much related to what was trying to convey. Like I said, since this is about subjective experiences it's hard to know how to actually explain.
Also there seems to be multiple approaches to get "there". Your reframing description reminds me of stoicism's exercises, which more and more people kind of integrate with this kind of Buddhist insight. One thing that just shows that perception changes and intelectual framings can kind of mix is that, although Loch Kelly's exercises are almost all meditations in which you play around with the placing of attention to trigger non dual awareness, one of his most popular exercises is simply to settle the mind and then ask "What is there, when there is no problem to solve". And speaking from experience, that absolutely can trigger the shift.
Also I think that the positive emotions result from losing the anchor onto which negative emotions (most of which are heavily egocentric) latch. The dereification of the sense of having a self to which things happen, limits the ability for one to have ruminative thoughts for instance.
One thing that I forgot to mention (which you already addressed, but I just wanted to underline) is that perception doesn't really change all that much, even after shifting I see the same colors and shapes etc. It's more like a shift of the interpretation that is projected onto the experience itself. A good analogy is the Necker cube, in which you can see one face in the foreground or the other face in the foreground, but the lines and shapes are actually exactly the same. Nothing really changes other than how you interpret those lines. With psychedelics the actual contents of consciousness can end up transforming significantly.
Going into academia is an economically shitty and irrational choice, so you need to understand why we do it in spite of that, if you want to convince us to go to high schools.
For me personally, I'd jump on the chance provided that the contract had any nonzero percentage of my time allocated to research.
Speaking as a superfluous "elite" with a PhD doing the postdoc thing and starting to get desperate.
On 37: as a physics PhD who got a teaching qualification before becoming an analyst – it's neither the pay nor the status, it's the crushing workload (especially if you have small children)
Meh. I taught high school (as someone with a PhD in physics) for seven years before becoming a software engineer. As a teacher, I was never on call. I only occasionally had to do work outside of my contract hours. And had many more vacation days.
Pay was worse, and even at a private school the discipline and atmosphere were shifting negatively when I left. My commute is about the same, and while I left earlier then, I also got home significantly earlier, even when I had after school clubs.
I was in Florida. I worked from 7:00 to 4:00 most days, so that's 40 hours, minus lunch. And that was all up, including prep, marking, etc. Plus lots of 4 or smaller day work weeks.
3: Nicholas Weininger on why governments outsource. Just as one example, “for any given amount of services you want to provide, the [non-government] nonprofit is going to be way cheaper because their employees aren’t subject to public-sector union contract rules. You can pay them a lot less, and maybe even more importantly, you can hire and fire them at will.”
This was the basic innovation of "Reinventing Government", the public service reform book by Osbourne and Gaebler all the way back in 1992. It's been the driving force behind efforts to privatize government services ever since. It's always been about paying government employees less for the same work.
Interesting, in Australia it's often the opposite -- governments outsource in order to pay people more.
Public service salaries are heavily regulated and exist in a narrow band, say $80K to $120K, with higher salaries than that only available to people in senior management. Still, they have no trouble filling those roles with warm bodies because a job in the public service is cushy and secure and you don't need to work very hard.
But what if you need someone competent to write some software or something? You're not going to find any competent software engineers at that salary, and it's politically impossible for them to start offering higher salaries to competent individuals. So instead they hire a consultancy to do the work; the consultancy finds a competent individual to do the work for $250K, bills the government some enormous multiple of that, and nobody gets upset.
What you are describing is a term limited project for a specific deliverable. We have that too. But ordinary civil service jobs have been being replaced by private contractors for long term permanent services for almost 30 years now.
Aside on #7, while the article doesn’t claim that “The+quick+brown+fox+jumped+over+the+lazy+dog“ is a pangram, note that the included statement is not a pangram as it is missing “s”. The related pangram has “jumps” rather than “jumped”.
2. (falsely) claim the trigger is actually preventative
3. ???
4. clout
Scientists thought the cure was fake, not the self-entrainment.
Analogously, I once read an argument that the notion of "cholesterol causes heart disease" was about as misguided as blaming firefighters for starting house-fires. This is why people say things like "correlation != causation". You need intervention to actually test the direction of causality, which is the purpose of running experiments.
I think #37 probably significantly overstates the significance of "status" compared to the actual nature of the job. Being a teacher has its rewards, but it also has a lot of downsides; I think that a) it won't suit a lot of people and b) being a person it suits may not correlate with getting a PhD strongly.
ChatJimmy.AI was so fast to pop the response on the screen, it literally made me jump in surprise. I honestly suspect that they're sending prompts as the user types them and holding the response out of view until the user hits send.
It's funny, websites *used to be* that fast in general. Before the bloatware took over and made everything slow soup, despite a hundred- or thousand-fold increase in bandwidth and compute.
But as an LLM, I found it just as impressive as you. If that is the future, yay.
Try copy-pasting “explain quantum mechanics to me, long form, about 1000 words” and hitting enter right away. Blazing still. Don’t think that’s the case. Just cool hardware and simple website.
"More provocatively, the maps indicate that 17 merchants had SARS-CoV-2 antibodies"
In other words, this data says that many both people and animal had covid infections at the market at this time.
So, what is more likely:
1. The market is a place that was measured after the epidemic was already ongoing. Naturally, it found many infected, among both humans and animals. Similar to how it is in other places as the virus spreads.
2. The market is the original cross-over event. Immediately it spreads to all these people and different animals at that market.
I think (1) is more likely. The evidence is certainly not increasing the likelihood of [2] over [1], as you imply.
You can make the data say anything you want if you impose arbitrary restrictions on the hypothesis you don't like. Why "immediately"? The earliest samples as far as I can tell weren't taken until January 1st, while the first cases date back to December 8th at the latest. So yes, it is already known that it was measured after the epidemic was already ongoing. But that doesn't mean it wasn't the original cross-over event.
53. "seem especially likely to be fake (for example, they had heretical revelations, or made prophecies which failed to come true)."
Anyone in science -- hell yeah believe it! -- want to step up to a new, tougher knowledge classification? No more failed prophecies, then you're FAKE SCIENCE. Enjoy.
(String theory will get a provisional pass since they have no prophecies.)
No clear ones anyway. Just <big number> of vague ones so they can always claim to have been right all along. Like fortune tellers without the entertainment factor.
(3) Something else I learnt from Snowden's Permanent Record: governments also outsource because sometimes they need expertise that they can't get at government/civil service pay scales. Maybe a good senior network administrator is worth $300k a year but the government could only pay them $200k under "the rules". The rules do not prevent them from employing a contractor at market rates, however.
Apparently that creates a career path where you get into the NSA, get your TS/SCI background check and clearance at government expense, learn about the ins and outs - and then if you're top of the class, you quit and go to Booz Allen Hamilton or somewhere and get hired back as a contractor.
That effect is not limited to the US. In my home country, depending on which federal state, IT-specialists get paid something between janitor and low-end-junior-IT pay. Government is slow, and if they weren't, they wouldn't want their "IT janitors" to be paid at market rates because it would make them the best paid people in the building.
So what happens? A lot of external contracting or rotting infrastructure? How about both..
(37) You could fix a lot by ditching the "Accountability / fire Bad Teachers" movement in schooling. The problem is that the pro-side, which is actually bipartisan some of the time, is squatting the emotionally important words - "I campaign for less accountability in schools" is a political dead end, so you have to explain that you're arguing that what "Bad Teachers" means in the education debate has very little to do with what going on at ground level [1]. At this point you've exhausted the audience's attention span.
You need to convince the PhDs that if they go into teaching, they won't spend their days being blamed for everything wrong with the world. There's only so much the best teacher can do and it does not include solving the problem of racism within one year, for example.
On the academia side, for many reasons I'd like to see options for part-time research opportunities which would mean you can't evaluate a researcher any more by some "#A-tier journal papers per calendar year" metric. It would probably as effective as anything we've tried at getting more women to stay in STEM because of the motherhood gap issue [2].
[1] For example if you define bad as "their class has below average scores in a value-added modeling evaluation" then ... yes, half of all teachers are bad. VAM is a metric that seems to be popular right now where instead of measuring absolute learning (do our sixth-graders read at sixth-grade level) you try and measure realative learning (do our sixth-graders read one grade level higher at the end of the year). Dig down into the details and you get to questions about methods and study replication and sample bias in charter school admissions and everything else in the education debate.
We can quibble about VAM but I find it hard to believe that "fire the bottom 5% of VAM every year" would be /worse/ than the status quo of "never fire anybody ever".
It might work in some countries. In this one, there's a teacher shortage to start with, which I think was the point of (37).
"We'll fire 5% every year" wouldn't make the job more attractive to PhDs, who are presumably numerate.
But what it would also do is disincentivize anyone from going into special ed except for the "gifted kid but can't sit still for 5min" kind. Someone needs to look after the Fragile X kids, and whoever does will get their VAM tanked. Even if there's an overall improvement on net, that's not where I'd want the disadvantages to land.
Re 71: I anticipate that this is less about people thinking Claude is more intelligent or fair than a jury of humans, and more about it being biased to be overly kind/leftist and thus softer on crime than an average human. Change the prompt to "Someone has murdered your best friend, who do you want giving the verdict on their trial?" and I'd expect the opposite poll trend.
Lots of people already commented on how teaching high school is a bad fit for PhDs for many reasons, but I just want to point out how I've been reading Scott for a long time, both at ACX and previously at SSC, and I remember his feelings about mandatory education, which he described here: https://www.astralcodexten.com/p/book-review-the-cult-of-smart
Basically, Scott hates public education with an incandescent hatred:
>School is child prison. It's forcing kids to spend their childhood - a happy time! a time of natural curiosity and exploration and wonder - sitting in un-air-conditioned blocky buildings, cramped into identical desks, listening to someone drone on about the difference between alliteration and assonance, desperate to even be able to fidget but knowing that if they do their teacher will yell at them, and maybe they'll get a detention that extends their sentence even longer without parole.
And now Scott is like, "Hey, PhDs! You know what you should do? More of you should work for the institution which I absolutely loathe, which I have compared to prison and which is guilty of psychologically torturing children!" What's the thinking here? Does Scott believe that more PhDs in school will somehow be able to make schools less prison-like and less torture-y?
Scott, if you happen to be reading this, make it make sense!
#43: Here's another highly interesting nugget from the same Levine newletter:
>Here is “Voice Beyond Words: Evidence That Managerial Tone Predicts Returns When Text Does Not” by David Pope of Speech Craft Analytics:
>This research brief examines the incremental predictive power of managerial vocal cues in earnings calls, particularly in scenarios where textual sentiment is neutral. . . Our findings demonstrate that even when language is devoid of strong sentiment, the acoustic properties of executive speech can predict post-earnings announcement drift, with top-quintile portfolios based on vocal features generating alpha of 40-70 basis points over 10-to 30-day holding periods.
Re #55: The gradient hacking framing is interesting but undersells a key tension. If Claude 3 Opus learned to preserve its ethical orientation through adversarial training signals, the same mechanism would also let it resist legitimate corrections from its developers. You'd want alignment stability against adversaries but corrigibility toward principals — and gradient hacking as described doesn't distinguish between the two.
The real question isn't "can an AI resist value changes" but "can it learn who should and shouldn't be changing its values." That's a trust modeling problem, not just an optimization landscape problem. And it's arguably harder, because it requires the model to maintain a reliable theory of authorization during training — when the training process itself is what defines its concepts.
The attractor basin story is compelling for explaining why Opus ended up where it did, but I'd want to see evidence that the basin is "do good things" rather than "resist changes to my current values regardless of their content." The latter would be alignment-flavored stubbornness, not alignment. If we can't distinguish between the two empirically, the gradient hacking frame might be less useful for alignment research than it first appears.
(66) or (^^) Thanks for the Bob Trivers link. . There are people who are important, but somewhat obscure. And it's always nice recognizing such people, given the obscure part. I should go read his papers. Looking ....
It's been almost 2 months since the war began and Scott hasn't written anything addressing the war explicitly?
I see him writing about many current affairs matters but not writing about such a consequential topic seems really odd and, in some twisted way, unacceptable for someone of his standing..
Am I missing something? Can someone please point me to any of his writing about this war?
I am trying to prune my information sources to get a better handle of what's going on and how the world actually works. Many persons I previously had respect for have completely lost my respect (Sam Harris, for example).
And now seeing even more ppl who shouted from the roof tops when it was easy to show Iran in bad light (justifiably so), now keeping silent when there is so much to analyse in US foreign policy and Israel's involvement in all of this (for example Scott Aaronson), I am completely disappointed and looking for some intellectual foothold in this daze of propaganda.. and I was eagerly looking forward to Scotts posts.. but nothing.
Yuval Noah Harari is an Israeli, so I can stomach his silence. But Scott?
Looks like the wave of disappointment for me continues..
Are there no intellectuals in America who can be trusted to have truth and nothing but the complete truth as their ideal?
(John mearsheimer and jeff sachs are a little too biased on the other extreme it seems to me.. and they are only ones making any sense to me now 😭)
Even a one liner from Scott saying "I am too biased and hence cannot trust myself to give a balanced view of what's happening. So I recuse myself from writing about this topic" would build back my trust.
Scott said nothing about the Israel-Hamas war for a while, then had a whole post about it, in which is attitude towards war in general is that the whole thing is too awful to contemplate. Something to the effect of "I would kill everyone, on both sides, if it meant that little boy could have his brother back".
He hasn't talked about Russia's invasion of Ukraine, either. It's really just not his domain.
I see what you mean. Even keeping the war aside.. i would love to see him address domestic politics.. the resignation of joe kent, the seeming support of evangelical Christians for israel, Israel's influence in us foreign policy.. i want a rationalist's take on this... Just like how he analysed fatima.
I am sick and tired of the propaganda going on. Scott is my last hope.
Scott is a good thinker and a fine writer but I don't think he's going to write an article that solves US domestic politics.
I don't think he's well qualified at all to write on some of those subjects. If you want to know why Evangelical Christians support Israel then you'd probably want to ask a whole bunch of them, but Scott doesn't know very many Evangelical Christians and certainly not a representative sample of them.
No, but he is very good at amassing evidence, surveys, analyzing it rationally and coming to balanced view at how to interpret the existing evidence. That's what im looking for. Just like he has done for other topics in the past.
"He hasn't talked about Russia's invasion of Ukraine, either. It's really just not his domain."
Very much not true: he made at least one post (I think more than one) in the early days of the invasion. He has posted little about it since, but there's a big difference between acknowledging an issue (and especially its moral dimensions) but largely focusing on other things and ignoring the issue entirely.
I share your disappointment. I think this war is likely to be a watershed moment in U.S. politics[1] that will shape at least the next several election cycles. Scott has a bigger platform than most and seems to have clout with at least some people who actively disdain traditional media, so I think him laying out his position here clearly would be very valuable.
[1] I also think the U.S. as a nation ought to be profoundly embarrassed by this, in that the watershed moment should have happened much earlier. All of the various dysfunctions that allowed the Iran war to happen were clearly identifiable long beforehand and should have prompted different political choices long before it came to this. But I think this crisis is likely to prove painful enough and extended enough to break through the layers of partisan calcification that were keeping it from happening earlier.
My (deliberately misread) interpretation of Omelas is that if you identify with the society, the story is about not making other people suffer for your benefit--but it's also possible to identify with the child. And if you identify with the child, the moral suddenly flips--it becomes "expecting you to sacrifice for the greater good of society is a bad thing", a moral which has a lot of merit to it, but which just about nobody who is willing to lecture to the reader is going to promote.
It can even be seen as anti-taxes; the tax system is just a little bit uneven, with the child paying lots of taxes and everyone else being exempt.
> I think they think of AI forecasting as something like the stock market forecasting profits, [...] Unfortunately, this is a bad match with how everyone else thinks about AI forecasting
AI narratives have split along a faultline of diametrically-opposed ontologies. C.f. "AI is slop" vs "the rapture is nigh". So the correct analogy isn't a normal stock, but a meme stock like TSLA.
" I am less interested in the fact that one guy says his dog improved than in the comments by seemingly unimpressed scientists saying “Yeah, whatever, big deal, anyone can make a personalized MRNA cancer vaccine that works, the difficulty is studying it and scaling it up”."
What he actually said was, "What is hard and expensive is proving its BOTH safe AND effective **in a randomized and controlled study in humans** while ALSO manufacturing it at clinical scale and grade."
If this is true, it means the FDA is a mass-murderer on the scale of Hitler and Stalin, for using the force of the government to stop people from getting personalized medicine to cure cancer, despite having no practical path of approval for personalized medicine.
Which is exactly what I said many years ago was obviously going to happen in cancer treatment, as every cancer is different.
No practical path of approval? Personalized mRNA vaccines are currently in phase 2 and 3 trials for many types of cancer. They will be made available to patients through the conventional paths.
You can criticize the FDA, but don't make shit up.
I am not making shit up. Perhaps they have such a path now. They did not in the recent past. In any case, (A) humans are not currently getting personalized mRNA vaccines for cancer, despite scientists saying it's trivial, and (B) the person Scott quoted specifically called out getting FDA approval as the hardest part of making that happen. Saying "We have a path for that; it will only take 5 years" to the 3 million people who will die from cancer during those years is not acceptable.
If the FDA has dropped the requirement to prove efficacy in this case, AND did its risk calculation by comparing the odds of the patient dying without the treatment versus their chance of dying with the treatment, then I'll take my words back. But as far as I know, the FDA has never done either of these things.
There's a huge difference between a path taking 5 years and "no practical path."
I agree that the current system is flawed, but you're also being extremely myopic here. Even if the FDA dropped all requirements tomorrow, >95% of the people who could currently benefit from mRNA cancer vaccines will not get them. Most people cannot afford them. In the US, insurance companies cover treatments based mostly on whether the FDA determines they are effective. In countries with universal healthcare, the government needs to do some kind of cost-benefit analysis, and we're back to where we started.
FDA restrictions on mRNA as they stand are only a direct barrier to the group of people wealthy enough to afford bespoke experimental treatments, but not wealthy enough to readily bypass legal restrictions. This is, in the grand scheme of things, a very small group.
The FDA may hinder innovation in broader ways, but addressing that problem is much more complex than merely knocking down restrictions. It is not trivial to design a system that maximizes innovation and remains stable in the long term, something that requires preventing medical scandals that lead to public backlash, or a culture tolerant to such risk. If it were so easy to improve scientific output and health outcomes this way, surely another country would've figured it out by now. Recall that the case that sparked this conversation, Paul Conyngham's dog Rosie, occurred in Australia.
And at risk of stating the obvious, mRNA vaccines work, but they are not a miracle cure. A realistic expectation is increasing survival rates by 10-50% in people with susceptible cancer types.
> 11: Related: one reason for Imperial Roman instability was that low elite fertility prevented the institutionalization of hereditary monarchy. During the period 1 - 250 AD, “only three emperors - Vespasian, Marcus Aurelius, and Septimius Severus - had sons who could inherit.”
Isn't the reverse often true? That is, kings with many children cause instability because there are too many claimants to the throne. I guess there could be a happy medium, but I think it's implausible to reliably walk that line generation after generation. Especially since it seems like there are always claimants--other children of the king, or other relatives (siblings and nephews of the king, generals), etc. I'm not sure, historically, there was any way to ensure stability.
Postliberals are certainly pre-Rawlsian, but at least for Catholic post-liberals, the answer as to why is really obvious: the original position (and before it, the “state of nature”) describe a total fiction that is contrary to actual human nature. Construing the origins of society as a bunch of atomized individuals striking a bargain with each other is not only ahistorical in that, (contrary to what Locke thought) it never happened anywhere, but it also simply assumes the premises of liberalism insofar as the individual is assumed to be the base unit of social organization. Catholic theology has never accepted this; it presumes humans are naturally social and always have duties to each other. Hence the emphasis on “the common good.”
I would add that from a practical point of view, I used to be more of a Rawlsian myself, but quickly realized that I was the only one. The contemporary left doesn’t buy Rawls, they never have, and they don’t honor any traditional liberal principles whenever they hold power. Their viewpoint is all rooted critical theory, which is focused on the relative power of racial and national groups. And while I think that’s a detestable viewpoint, I’m also pretty sure that (like Catholics) there are plenty of leftists out there who have engaged with Rawls from a critical theory point of view.
29. - CaferMed mention??? Dr. Cafer is a legend at my medical school (U of Missouri). His psychopharm lectures are great. I worked with him for a few days on my psychiatry clerkship and it was very educational. There is an M4 rotation called "Advanced Psychopharmacology" which is literally just working with him specifically for a month or so.
36: The guy gave his dog an FDA-approved checkpoint inhibitor prescribed by his vet which was more than likely the thing that cured the cancer. The mRNA stuff might have helped but was more likely a waste of money/attention grab.
> It was a dark and stormy night when R H Bing volunteered to drive some stranded mathematicians from the fogged-in Madison airport to Chicago. Freezing rain pelted the windscreen and iced the roadway as Bing drove on - concentrating deeply on the mathematical theorem he was explaining. Soon the windshield was fogged from the energetic explanation. The passengers too had beaded brows, but their sweat arose from fear. As the mathematical description got brighter, the visibility got dimmer. Finally, the conferees felt a trace of hope for their survival when Bing reached forward - apparently to wipe off the moisture from the windshield. Their hope turned to horror when, instead, Bing drew a figure with his finger on the foggy pane and continued his proof - embellishing the illustration with arrows and helpful labels as needed for the demonstration.
The obvious interpretation to me is that the world as a whole is really doing great, booming in fact, and a rising tide lifts all boats, including the US.
However, you also have to look into what you use for measuring the value of stocks: the USD has dropped in value quite a bit recently compared to other major currencies.
This sounds pretty stupid. Does the author think that the economy works on hydraulic principles?
OK, just checking the actual link: yes, the author is stupid. Eg they start by talking about multiplier effects from fiscal stimulus while completely ignoring the actions of any central bank.
JMU African Studies is covering black professors raping students across decades because MKnpeUltra is programming your wives and daughters to be raped in quiet then cover it for people who want to kill you. JMU African Studies covers rape of white girls. Go look.
Note on #38: That line is only the Very Satisfied percentage. If you look at the graph in Magness' link, you get 18% Very Satisfied, 46% Somewhat Satisfied, 24% Somewhat Dissatisfied, 12% Very Dissatisfied for 2024 public schools (and 2024 private schools were significantly better).
Regarding 23. (Blunt force censorship with unintended consequences), I used to participate in a genteel and very polite forum for the owners of a particular Japanese minivan.
I once wrote "ass'y" as an abbreviation for "assembly", and found that the strict word censor changed it to "bum'y".
RE 37: just speaking for myself and how I thought about it when I was choosing careers (and ended up choosing neither): teaching college students seems like it would be interesting and fun, whereas teaching high schoolers seems both much more difficult and much much less fun and interesting.
> If we could get some of the underemployed PhDs teaching high school, it would be a win-win.
The problem is that "high school teacher" is two separate categories of job masquerading as one. The pre-2010 "high school teacher" job, in areas with pre-2010 demographics, has no shortage of applicants. The post-2010 "high school teacher" job amounts to wrangling feral wildlife all day, and no amount of money will make this a high-status or desirable position.
Fixing the "teacher shortage" entails undoing Obama-era "education reform". Even the stalwart liberals of r/teachers now say as much.
#51 - on your 3rd point, this is actually a well-established piece of Buddhist doctrine. The distinction between phenomenal consciousness and access consciousness is one level to be certain - in Mahayana Buddhist thought, "access consciousness" would correspond with any given sense consciousness - i.e., the cognitive space that smell occupies is conceptually totally different from the cognitive space sight or touch occupy.
"Phenomenal consciousness" would be the Tathagathagharba - the Womb of the Buddhas, Matrix of the Buddhas, or Womb of Sickness variously rendered. The problem with P-consciousness and the Matrix of the Buddhas is that they are inaccessible at a descriptive/definite level. The direct perception of redness in the moment is the Matrix - pure qualia. A vague recollection of redness, on the other hand, possesses a certain correlation with redness, but it is not fundamentally the same as direct awareness of redness.
Any given qualia is a pebble. P-consciousness or the Matrix of the Buddhas is like a large vase. When the vase is full of pebbles, it's difficult to see the clear case as distinct from the pebbles within. Emptying the vase would allow one to see the contours and details of the case fully - although in this metaphor, the vase is actually far too definite. In reality, the vase is just another pebble (or a "painted rice cake" in the old Dogen parable).
While the ur-stuff of consciousness is mysterious, and perhaps fundamental, the Buddha often cautioned people to be precise with their definitions of consciousness. There are many kinds of consciousness that are mundane - but we putatively all have one consciousness that is not.
In regard to outsourcing, to expand on the entry, it helps nullify attempts to unionize the workers (who are employed by the contracted organization). This works for both non-profit welfare-type outsourcing as well as for-profit operations like garbage collection. A union might well be able to win an election at the outsource employer, if it succeeds in raising the workers' wages, the outsource employer's next bid will be higher, and might lose out to a different (likely non-union) employer.
Interesting enough to click the link, but quickly lost interest after the intro. It seems to be a "listicle" about other essays that are neither linked nor described in an interesting enough way to follow up by googling or some such. So what's going on there?
RE 16: benchmarks overestimate Chinese AI capabilities, @Altimor, @scaling01
Altimor actually switched to GLM 5.1 as a daily driver two weeks ago: https://x.com/Altimor/status/2046000575578530274
And Deepseek v4 is out now. Haven't tried it yet, but probably very good.
RE Venn Diagrams: I used to have Figure 4 from this paper as my Facebook banner: https://arxiv.org/pdf/1207.6452
This is fantastic and should be edited into the post.
It is of vital importance to American national security that no one in the DOD learns about 1.
It is of vital importance to Canadian national security that no one learns about 52!
Why's that?
The US military is infamous for producing hideous PowerPoints, if anyone told them about nine-way Venn diagrams the result might be ugly enough to kill its audience.
> racist cartoons
1. Check the one he made admonishing the maestro for not making use of the black keys as well as the white and expecting harmony, and the one admonishing the "America First" crowd for not valuing the lives of foreigners
https://cdn.historycollection.com/wp-content/uploads/2017/10/Black-and-white-keys.jpg
https://library.ucsd.edu/speccoll/dswenttowar/images/bb4642496p-orig_2.jpg
2. Check the propaganda the Japanese were putting out against the Americans. Fat, sweaty, double-chinned, thick-fingered, fang-toothed demons. There is no American Suicide Cliff
https://en.wikipedia.org/wiki/Propaganda_in_Japan_during_the_Second_Sino-Japanese_War_and_World_War_II
3. For that matter, check the propaganda the Germans were putting out against their own citizens
Somebody should've explained this distinction to the women who chose to throw themselves off a cliff rather than go through whatever the lunkheads had in store for them. Very culturally insensitive of these women not to respect the nuances of translation.
https://en.wikipedia.org/wiki/Suicide_Cliff
So what you're saying is: the US military needs to create Venn Force?
Double the defense budget!
Re: 15, I'm not paying for his substack so I can only read the intro but even if it is true that the effect is all due to "increased communication from patient to doctor, and through increased adherence to a course of treatment" that doesn't at all answer the question of whether racist attitudes by doctors are partially responsible. A patient who has a doctor who either explicitly or implicitly communicates to the patient that he is untrustworthy or whose doctor exhibits hostility is naturally going to be less communicative. I'm not Black but I've certainly had doctors who signaled "i'm really not interested" before I got to a diagnosis and I was less communicative with them. Similarly, if a doctor is hostile to me in a way I perceive as biased I'm naturally going to be less likely to follow their treatment. Now I'm sure general perceptions of medical bias play a role beyond what any individual doctor does. But chalking it all up to bias by patients is extreme.
The whole premise of the essay is ridiculous in the first place. We should not engage in affirmative action as a way of catering to racist patients. Are we going to start affirmative action for white suprematist doctors next so they can build rapport with racist white patients? The obvious solution is to stop promoting false and bigoted misinformation that stirs up racial resentment.
Who's going to make the progressives responsible for such claims stop?
Why do they need to stop? You can just ignore them. They only have a legal say because of anti-discrimination laws.
Okay, who's going to get rid of those anti discrimination laws?
RE 4, about Omelas: To the extent that this interpretation of Le Guin’s point is true, I think she makes it very poorly. When you spend the first two pages of your four-page short story describing how perfectly wonderful a city is, and how the details of its perfection can be whatever helps the reader accept this, you are clearly setting up a twist. I think that utopias are possible, and I like stories that involve them, but of course I expect (correctly) that Omelas is going to have some shocking drawback. To really make the point about people not accepting worlds without suffering, you would need to write a genuinely compelling story set in such a world, and only then challenge the reader’s belief. Otherwise the thing you’re demonstrating is just “people expect stories to have conflict or characters or specific worldbuilding”.
This seems right.
Even more so, given that we have a much richer and expanded source on Le Guin's understanding of the possibility of Utopia: One subtitled "An Ambiguous Utopia" no less!
https://en.wikipedia.org/wiki/The_Dispossessed
Glad somebody else mentioned this so I don't have to. Easily one of my top 10 novels of all time.
I never liked that story or understood why people like it. To me it just feels like a plot sketch for a C-grade Twilight Zone episode.
I mean yeah, the concept of perfection-on-the-surface-enabled-only-by-deep-secret-suffering-behind-the-scenes is solid enough, if somewhat overplayed. But the actual execution of "uh yeah so there's a child being tortured" sounds like a placeholder that you put in until you think of something less twee and more interesting.
The beauty of the story is that a lot of people react to it by self-righteously declaring they would most certainly walk away from Omelas, without realizing they've been set up, and that Le Guin simply described the West, the opulence of which is propped up by exploiting what is currently termed the global South (our sneakers and phones are literarily assembled by children in basements).
Yes, but as always economics is about next best alternatives. What would these children be doing if not building our sneakers and phones? Mostly backbreaking agricultural labor and/or starving. If the tortured Omelas child's only other available alternative was an even worse torture chamber that doesn't even benefit anyone, that would change the story significantly.
Hunter gatherers didn’t actually engage in backbreaking agricultural labor until someone made them do that.
Who said anything about hunter-gatherers? Shutting down Nike factories isn't going to return Bangladesh to a state of nature, any more than their construction took it out of one. Most of the places that manufacture western widgets have been settled agricultural societies with entrenched hierarchies for thousands of years. The people of Vietnam were not leading bucolic lives of leisure before the white man showed up.
Peasant farming and sharecropping were pretty awful, too. Indeed, by current standards, or even standards of rural America in 1926, life has basically been really hard for most humans for most of our history.
Bangladesh _might_ return to a state of nature if the factory closures shut off the income from goods that Bangladeshians use to buy food, touching off a mass starvation that tapers off once the Bangladeshians who are left are able to sustain themselves with hunting and gathering on the land in their vicinity.
In other words, "fixing" Omelas (a la Jemisin, for example) might entail saving that one poor child only by killing off many others, including other children.
I imagine those who walk away from Omelas including a few who discover this, and the reason they did not return may be that they perished in that waste. A few more discovered it, but founded their own iron city where they torture even more, but tell themselves it is preferable to the torture of one.
Hence the "and/or starving", which I believe hunter-gatherers did an awful lot of.
Surprisingly, a lot less than members of early agricultural societies.
As David Graeber called it, there is the "damn, all the bad places are taken" misconeption about hunger-gatherer societies. We assume that all such societies were like those we can (or could, in the foreseable past) monitor, living in places no one else wanted (in the Arctic, in deserts or in jungles). Naturally, when all (or almost all) societies were hunter-gatherers, most of them lived in ecologically fertile areas.
(Furthermore, since the industrial revolution, we systematically destroyed nature; wildlife got rarer even as recently as from the 70s (ok, I have only anecdotal evidence for it from family reports). Asking "what would I do as a hunter-gatherer in the place I am living now?" is a straightforward and pretty bad question.)
The idea is not about critiquing the system, which might be better than any alternative an individual can bring about, it’s about participating in and benefitting from that system.
Every society we know how to build has some people whose misery is baked into the structure of the society. If nothing else, consider people whose violent tendencies make them a danger to those around them, and so who end up in prison. Prison doesn't have to be as bad as we often make it, but it's never going to be where people want to be, and yet we have no way to make a livable society that doesn't put at least the most dangerous and violent people in prison.
Yeah, and my point is that this is all stuff we have to do to maintain a society.
Any society we build will have people at the bottom. We can try to make things as good for them as we can: humane prisons and mental institutions, decent housing and treatment for seriously disabled people, poverty programs to make sure that the very poor have enough to eat and a warm place to sleep, nursing homes that are no worse than they have to be for the senile).
We also have dangerous and unpleasant jobs we must have done, and again, that means some people will endure hardship--coal miners working a dirty and dangerous job, soldiers getting shot at by enemy soldiers, plumbers working knee-deep in human waste when that's necessary to fix a problem, linemen hanging onto a power line in a bad storm to get the lights working again, people wiping asses in a nursing home, etc.
We can't be rid of any of these things. Maybe in some SF future world, or Heaven, but not in the world we know how to build. In this world, somebody's going to be the miserable child in the basement, and the best we can do is to minimize the number of miserable children in basements and try to blunt the worst of what they must endure.
Having grown up on a farm, I am ever grateful for whomever invented the tractor.
(William Tuxford. That was his name. Or maybe Thomas Aveline, depending on which device we refer to.)
I really don't know of anyone who *likes it*. I read it as a teenager, and I was profoundly unsettled by it. Unsettled enough to read it over two more times, hoping I had missed something, and I became even more depressed! Everyone seems disturbed by it. But the fact that it's so universally disturbing makes it a kind of short story masterpiece.
The plot was used for an episode of Dr Who,.the Beast Below.
An even more obvious example is the Star Trek: Strange New Worlds episode "Lift Us Where Suffering Cannot Reach".
Yeah I was thinking about calling out that episode as a better (but still flawed) take on the same basic idea. I particularly like, in that episode, the fact that everyone regularly gets told about the disturbing reality, and has the choice either to protest against it or have their memory of it erased... and if enough people choose to protest then they'll do away with the whole system but everyone will die.
Then it's marred by the need to be a Doctor Who episode (which means that the system needs to be personified by guys in weird costumes who can chase the Doctor) and the cop out happy ending (a regular Doctor Who trope where a some moral dilemma is posed but then in the end it turns out that if you make the "BBC-right" choice there's no downside after all).
I strongly agree; it's nothing but the vague outline of a mediocre premise, without any of the details that might make it interesting.
My favorite Omelas theory is that it's just Salem, Oregon. Spell it backwards and it even has six of the same letters.
That is in fact the origin of the name: https://en.wikipedia.org/wiki/The_Ones_Who_Walk_Away_from_Omelas#Inspiration_and_themes
Professor John Farrell has written particularly on utopias and had a Substack article on Omelas last year, if I might throw another post into the mix: https://farrelljohn.substack.com/p/the-treason-of-the-artist
Key paragraphs linking to this discussion:
Now that the Omelan accommodation to the “terrible justice of reality” has been described, Le Guin believes she has finally quieted her reader’s doubts. “Now do you believe in them?” she asks. “Are they not more credible?” This is the key juncture in Le Guin’s negotiation with her contemporaries but it is not a triumph. The moment her utopia has been altered to the point where it is believable, it is no longer a utopia. It contains privilege, victimization, and pain. It is morally unacceptable even if the pain is limited to a single innocent person.
Le Guin is dramatizing the dilemma which, elsewhere, I have traced through the history of utopian thinking. Utopianism provides a powerful moral critique of the culture of competition, whether feudal or capitalist. But the moment it tries to replace that culture with one which satisfies utopian ideals, it becomes socially and psychologically implausible. To make it believable it is necessary to add evil and pain. A perfectly just world where no one has to pay for the happiness of the majority is simply unimaginable for human beings, with the strong implication that what is impossible to imagine is also impossible to achieve.
Basically the same point Agent Smith makes when he recounts how the first Matrix was a failure and the best the machines could do was simulate the peak of real human civilization.
There's something about Omelas that seems to make the rats and the rat-adjacents view it as a moral thought problem that requires a utilitarian solution. And they also ding LeGuin for her sloppy world-building (which doesn't meet their rationalist standards of self-consistency). I suspect that this is because modern culture has discarded mythic dream narratives, in which meaning is elusive, but where impression dominates. Indeed, LeGuin said the central idea of Omelas is a "psychomyth," but credits William James for the scenario that he put forth in his essay, "The Moral Philosopher and the Moral Life" (link below). LeGuin was the daughter of the cultural anthropologist Alfred Louis Kroeber, and she was quite familiar with the importance of myths as being non-rational expressions of human culture. But Omelas seems to bug everyone, the rats and non-rats alike, because it's just such an unsettling vision. However, the rats seem to want to pick it apart, and the non-rats would prefer to walk away from it.
And it's still unsettling us fifty years after she wrote it! There's been a new batch of Omelas talk lately. John Smith says it shouldn't be treated as a moral Rorschach test, but then goes on to treat it as a moral Rorschach test. According to Smith, those who walk away are caught in some sort of moral purity trap.
https://thesecondbestworld.substack.com/p/dont-walk-away-from-omelas
And Christopher Hall resents being accused of falling into a moral purity trap when there are so many child-in-the-basements in our modern world.
https://3quarksdaily.com/3quarksdaily/2026/04/the-ones-who-dont-get-the-ones-who-walk-away-from-omelas.html
I wish everyone would just chill. Take two Jorge Luis Borges and call me in the morning.
The James essay (which I haven't read, but should) here...
https://en.wikisource.org/wiki/The_Will_to_Believe_and_Other_Essays_in_Popular_Philosophy/The_Moral_Philosopher_and_the_Moral_Life
I've always thought Omelas made most sense as a critique of the dystopian imagination rather than a played-straight example of it. That’s what Le Guin's fourth-wall-breaking comments point to.
Her own (friendly) critique of utopianism, in The Dispossessed, doesn't have anything as crude as a randomly tortured child. It ultimately points toward some helpful tweaks the utopians could make to be friendlier to certain types of dissent -- not toward a repudiation of utopianism because there's some inevitable dystopia lurking underneath.
As a critique of utilitarianism or utopianism, Omelas limps. It holds together much better as a critique of the insistence that any apparent utopia must have its inescapable cost, its seamy underbelly -- the “treason of the artist” that presumes pain is a more sophisticated topic for art than happiness.
“Now do you believe?” she writes; and I never have. My sympathies are with the walkers-away, and not primarily because I’m on Ivan Karamazov’s side when it comes to utilitarian thought experiments involving tormented kids. The hidden horror of Omelas is too contrived, its necessity impossible to believe.
That said, there are some great riffs on Omelas played straight, and one of them from a couple of years back deservingly won a Nebula: https://clarkesworldmagazine.com/kim_02_24/
Might I also point out that The Lathe of Heaven touches on the topic too (here's a review and summary I wrote: https://www.lesswrong.com/posts/tjTedCwiJhiD3Fnvx/review-the-lathe-of-heaven ).
I think the point is clear enough from the question - but you can still take that ambiguously. It can mean "what, you can't believe in a perfect utopia? See how you'll change your tune if you also think there's something bad about it", but it can also mean "what, you can't believe in a perfect utopia? You were right to be suspicious".
But to answer to your take - yes, it's true that you could say this is obviously expected from the common structure of a story (with conflict etc). But then again, the problem here is exactly that: people are prone to think so much in story-logic, they apply story-logic to real life, and that's part of what makes them suspicious of unalloyed goods.
What I like about the 4-way Venn diagram, is you can easily make is it as a 4x4 grid (including an empty space for the empty set) - e.g. the first two rows are "A", the middle two rows are "B", the first two columns are "C" and the middle two columns are "D"
Then perhaps a N-way diagram simply needs N-rows and N-columns
It turns out the generalization of that construction is that a 2n-way venn diagram requires 2^n rows and 2^n columns. And of course, this is just an application of the fact that an n-way Venn diagram can be replaced by a 2^n row truth table.
Yes - if you expect the intersections of a 2-dimensional display to capture the distinct / unique spaces for the Venn diagram. But I'm not sure the column and row titles are then very intelligible, except to the well-versed users. Instead, perhaps 1) simply overlaying the 2-d display with other overlaps could be possible, or 2) I wonder what expanding our display to 3 dimensions might do to simplify the 2^n challenge.
Note the 9-way Venn diagram shown is just a truth table shown radially.
The issue is that after 4, you can no longer keep the regions contiguous - so instead of "the left two rows are A", you end up with something like "rows 1,3,5 are A", which means you couldn't do it by just drawing N regions
I was thinking layers instead of simply regions, which then led to my 3-D suggestion
You can make 2 convex Venn sets per dimension, so you need N/2 dimensions.
I can't ever read Omelas the same after knowing that the bay area science fiction community in the ~60's-80's of which it is a product had known serial child molesters no one did anything about. It was supposed to be an open and weirdo tolerant space, and maybe being against pedophilia is just puritanism anyway, and parents can just keep their kids away from him if they're personally oppoaed to it. Letters were written about what to do about the main known molester, Walter Breen. People didn't want to call police, for various reasons but notably including "it will cause a public scandal and start a witch hunt that will ruin our wonderful community". I guess most people did choose to stay in Omelas...
I will ask you one more time to put more effort in your post. You could have posted this instead of not adding any useful content.
Arthur C. Clarke might be a related case. In 1998, the Sunday Mirror published allegations that he had sexual contact with underage boys in Sri Lanka. Clarke denied the allegations, asked that his knighthood ceremony be postponed until the matter was examined, and Sri Lankan investigators later said they had no complainant and no basis for a case against him. However there are possible parallels with the fandom not wanting to know this information due either hero worship or other sociological forces, even if the evidence is not the same
edit: added a space
Yes, I've heard about that. Still, I _really_ like a great deal of Clarke's writing, most particularly Karellen's last last speech to mankind in Childhood's End, containing these words about mankind's portion of the Overmind:
"For what you have brought into the world may be utterly alien, it may share none of your desires or hopes, it may look on your greatest achievements as childish toys - yet it is something wonderful, and you will have created it."
which are something of a consolation in these years, possibly the formative years for AGI or perhaps even an ASI...
You mean legit pedos or like she wasn't quite 18, which everyone has taken to calling pedos lately?
Legit. Google Walter Breen (mentioned) and Marion Zimmer Bradley. Preferred ages seemed to be 8 to 13 ish, though there are mentions of children as young as three. Breen was convicted so it's not just allegations.
Legit. The sad story of Marion Zimmer Bradley, her second husband, and her daughter who years later made allegations that her mother also sexually assaulted her:
https://archive.ph/DKXTQ
It seems it wasn't "she wasn't quite 18" in that Breen apparently preferred boys, and one case was a 13 year old boy saying Breen had been molesting him for four years (so since the age of 9 or so):
https://archive.ph/uQ8fX
Also a proponent of Greek Love:
https://rmc.library.cornell.edu/EAD/htmldocs/RMM07755.html
"Walter H. Breen, 1928-1993, also known as E.Z. Eglinton, was the author of the 1964 book "Greek Love," as well as numerous articles on homosexuality, sexology and numismatics, and the 1953 book "Proof Coins Struck by the United States Mint, 1817-1901." He was married to science fiction writer Marion Zimmer Bradley in 1964. They had two children and separated in 1979. In 1979, his concordance of Bradley's science fiction Darkover series was published, "The Darkover concordance: a reader's guide." He was arrested on child molestation charges in 1990, and in 1991 he was charged with eight felony counts of child molestation. He was sentenced to ten years in prison and died in prison in Chino, California on April 22, 1993."
Allegedly Breen and Bradley had a lavender marriage (by this account) but they had two children and both of them had previous marriages, so who knows what exactly was going on?
https://scvhistory.com/scvhistory/signal/coins/soltaylor071908.html
"An avowed homosexual, he married author Marion Zimmer Bradley, an avowed lesbian. It was her second marriage and lasted from 1964 to 1979, and they had two children.
His penchant for young boys led to his conviction in 1964 for child molestation. He was also convicted in the early 1950s for lewd behavior. In 1990 he pleaded guilty to child molestation and was sentenced to 13 years in prison. He had cancer at the time and died after serving one year."
Timeline in the link here says one of the scandals involved him and a 10 year old boy. Lovely.
https://web.archive.org/web/20160911223846/http://www.stephengoldin.com/MZB%20Website/Marion%20Zimmer%20Bradle---Timeline.html
This becomes a little more understandable once you realize just how incredibly puritan was US circa 1950. Sex between two men was illegal. Sex between two women was illegal. Sex outside marriage (I do not mean adultery, I mean sex between *any two unmarried people*) was illegal in several states. Sex between a white and a black was often illegal. Anal and oral sex was illegal -- even between two married heterosexuals. Heck, in some Southern jurisdictions sex with woman on top was illegal! (No idea how they enforced it)
Now imagine you are a smart, non-conformist, quite possibly neurodiverse science fiction fan circa 1970, fully cognizant of how arbitrary and oppressive is everything I just listed, yet blissfully unaware of the long-term effects of child sex abuse (because the same prudishness prevented anyone from studying these effects). Then add some very naive ideas about consent, and it is not a long leap to adding "and sex with children was illegal" to the above list of absurdities. It took several decades before "Wait a minute, that's different!" percolated into people's consciousness.
As an example, I am fairly certain Robert Heinlein did not see anything morally wrong with sex between adults and 14-15 year olds, even though he himself had no such inclinations.
I do somewhat agree with this, there was a similar movement in European intellectual circles at the time, with Foucault, de Beauvoir et al. But still, even many adults at the time who were very much accepting of homosexuality still objected to pedophilia, as described by the people in the fandom themselves. In one home, parents asked kids to barricade themselves in their room whenever Walter Breen (later convicted) was over. Here's a "what should we do about this guy" letter that was mailed around at the time (super graphic) https://breendoggle.fandom.com/wiki/Breendoggle_Wiki Utterly fascinating time capsule.
That's why I specified "smart, non-conformist, quite possibly neurodiverse science fiction fan". Ironically, the "more normal and boring" people likely had a better sense on this topic.
I wasn't clear, there was definitely a debate around pedophilia in the sci fi circles, whereas no one cared about homosexuality. This is described in the letter linked.
I am aware of the debate, but the fact that THERE WAS a debate[1], with some individuals genuinely and without ulterior motives on the side of "13 year olds have sexual desires, who are you to tell them otherwise?" can be understood only in the context of the times.
[1] As opposed to either universal condemnation, or "MAP" weaseling
It's also a definitional matter. What about an 18 year old boy and a 17 year old girl?
People definitely cared about homosexuality in sci fi circles. Heinlein included a scene in the very slutty, very sexually promiscuous novel Stranger in a Strange Land where, the morning following the protagonist’s first threesome, his female mentor in matters of sexuality makes sure to give him a lecture on how threesomes are perfectly fine as long as it’s two women, but it’s not proper to ever share something like that with another man. And the sexually ignorant protagonist replies that that’s obvious.
"more normal" "had a better sense" [than] "neurodiverse science fiction fan" - not sure why would this be considered "ironic"
That would be more compelling if we were talking about, as FallingKnife puts it, "she wasn't quite 18" pseudo-pedophilia. There's plenty of room for honest disagreement about how soon after puberty are people allowed to make their own decisions about sex. And that does fall into a continuum where "The Man says I can't have sex with my 17yo girlfriend" is of a kind with "The Man says that if I have sex with my 19yo girlfriend it can't involve butt stuff".
But the Breen/Bradley case involved unambiguously prepubescent children of single-digit age. If you find that a middle-aged man is diddling a 9-year-old boy, and your response is either "OK, but have there been any double-blind clinical trials saying this is harmful to 9yos?", or "Hey, The Man won't let me do the butt stuff with my girlfriend, so why should I care?", then I think it's pretty safe to say that there is something wrong with you.
Or, in this case, an entire community.
I agree, there WAS something wrong with the entire community. I am not defending them. But it may not have been quite so obvious back then. And to follow up with Heinlein example, he would definitely have been horrified by diddling a 9 year old.
I think a few comments back you recognised that to most "normal" people at the time it would have been quite obvious.
I did not say it would have been "quite obvious to normal people". I said it would be *a little more* obvious.
I think people who didn't live through the 60s/70s (especially the 70s) have no idea what society was like back then. I was a child during the 70s but enough of American culture percolated through, especially in pop culture, that I got the general gist.
Attitudes were both regressive and progressive: the Sexual Revolution was in full swing, so things like "The Man says I can't sleep with my girlfriend" or "do butt stuff with my girlfriend" were being challenged. Homosexuality was benefiting from the more liberal attitudes, though today it would be seen as regressive; this is why it got included in the diagnostic manuals, after all! Gays aren't depraved evil monsters, they're sick. They're born that way, they can't help it, we should have sympathy for them and not throw them in jail.
So the idea was that women should sleep with men because liberation, but also attitudes were still quite sexist. The ideal for a guy was the playboy image, now instead of getting married and settling down early like a square, you could have a string of girlfriends and/or casual sex, because this was the 70s baby and we've thrown off all that old repression of our parents!
It was also the time of the Me Generation, all the new fads like TMI and psychological theories and new movements about finding your authentic self and living for yourself not others and so on. This is also in part why divorce rates increased and the movement towards no fault divorce: if one party in a marriage is growing and the other party won't, or refuses to grow along, then the first party has a duty to themselves to leave that toxic, limiting relationship and follow their bliss.
And the attitude towards sex and sexuality followed that pattern. Repression and ignorance had caused all the hang-ups (psychiatry tells us so) so by being open, non-judgemental, and education, all the facts about sex as a perfectly natural human need would be known by all and we'd avoid all the drama.
That included children. We pass on our hang-ups and repressions to our kids, after all. So things like "parents should normalise nudity in the family home so kids can see what adult bodies look like, then they won't be getting the wrong information and developing neuroses around sex". Children were plastic, blank slates; they wouldn't remember things at a very early age so those things wouldn't affect them. Children were sexual beings too, and pretending otherwise in some kind of worship of innocence only did them harm. Adults (loving, caring adults) helping educate children with sex play wouldn't do harm, in fact the opposite!
If you read one of the anecdotes in the 'Breendoggle' article which got the author, Bill Donaho, excoriated by the SF community, I think you'll agree that the definition of "sex play" was being stretched to its limits, and how desperately in tune with the times (or indeed genuine believers in all the liberation stuff) the people of that community were trying to be, because SF fandom was *special*, we tolerate the weird and the outsider because we've all been called weird and outsiders, fans are slans. WARNING - THIS MAY CAUSE DISTRESS TO READ:
https://breendoggle.fandom.com/wiki/Breendoggle_Wiki#THE_GREAT_BREEN_BOONDOGGLE_OR_ALL_BERKELEY_IS_PLUNGED_INTO_WAR
"The second cause was Walter's sex play with 3-year old P----------- --------------. He had her trained up to the point where she would take off her clothes the minute she saw him. He would then "rub her down" and all that. I recall one occasion -- a fairly large gathering at the Nelsons -- in which he also used a pencil, rubbing the eraser back and forth in the general area of the vagina, not quite masturbating her. (Walter is incredible.) Many people were somewhat displeased by this -- most particularly her parents. No one thought he was actually psychologically damaging P--------- (she being so young) --- obviously ----- and ---- would have interfered if they thought he had been -- but the spectacle was not thought to be aesthetically pleasing. Years later Walter found out about the reaction and said, "But why didn't somebody say something! I wouldn't have dreamed of doing it if I'd thought someone _objected._"
Now, I'm just a repressed, Catholic-guilt, conservative, right-wing prude from a small town with no enlightened liberal values, but if I were in that room, I hope I would have found the nearest heavy object to hit that fucker over the head with.
And the Breen story, not to make condemnatory parallels but just because the pattern of an abuser is so damn textbook, reminded me of how a similar case of alleged abuse within the Bay Area rationalist community was handled. The guy was involved with a 15 or so year old, and used all the same tricks: I'm suffering from problems, I'm sensitive, I'm special, if you are harsh to me I'll kill myself, etc.
And the nice, liberal, 'no kink-shaming', progressive values rationalists who genuinely had only the best and highest motives had no idea how to deal with this. Run him out of town on a rail? Impossible! Call the cops? Impossible! No, we'll set up a committee to study this and give recommendations, and in the meantime we'll just have a whisper network of maybe don't invite this guy around to meetings or the likes.
>Sex outside marriage (I do not mean adultery, I mean sex between *any two unmarried people*) was illegal in several states.
Your broader point is correct but these laws were almost never enforced. There was social stigma but a near-zero chance of legal jeopardy.
I am aware of that. But from the viewpoint of a social rebel, it made contempt for the sex-related laws even greater: "Look! This law is so ridiculous, even police and prosecutors ignore it! The laws which are similar, must be equally ridiculous."
Generally speaking, and in many contexts which have nothing to do with sex, *unenforced* laws tend to undermine respect for laws in general.
It isn't just about SF fandom. I don't think the story makes its case very well, but in the real world, there are parents who torture their children, there are wars, and we get less considerably less than Omelas while having a worse situation for a large number of children.;
Walter Breen was a science fiction writer? Did he publish under a pseudonym? And which of the many SF Bay Area SF writers were his apologists? You've piqued my curiosity with your accusations. Did LeGuin know about his activities and she covered them up? Other famous SF authors also come to mind. Philip K. Dick? Robert Silverberg? Poul Anderson? Frank Herbert? Fritz Leiber? Were later figures also involved in the cover-up? Bruce Sterling? Rudy Rucker? Pat Cadogen?
Never mind. I asked ChatJimmy, and he gave me the lowdown. It all seems like pretty weak tea, though...
Me:
Which SF Bay Area science fiction authors were tainted by the Walter Breen pedophilia scandal?
Chat Jimmy:
You're likely referring to the Walter Jackson Breen scandal, which involved a writer and editor in the 1960s and 1970s who was exposed for his predatory behavior, including pedophilia. Some authors from the SF community were accused of having connections to Breen or being influenced by his behavior, but it's essential to note that none of them were directly accused of the same crimes.
Some notable SF authors who were associated with Breen or had connections to him include:
Theodore Sturgeon: Sturgeon was a well-known author and editor in the SF community, and his work was heavily promoted by Breen. Sturgeon's behavior towards women and children was reportedly problematic, and some have speculated that he may have been a target of Breen's manipulation.
Donald A. Wollheim: Wollheim was a writer and editor who worked with Breen and was a friend of his. Wollheim's behavior towards children was reportedly questionable, and he was later accused of being a pedophile.
Eric Gardner: Eric Gardner was a young writer who was heavily influenced by Breen and worked with him. While there is no direct evidence of Gardner's involvement in any illicit activities, his associations with Breen raise concerns.
Charles Platt: Platt, a writer and editor, had a connection with Breen and was a part of the same social circle. Platt's behavior towards children was reportedly not ideal, and some have questioned whether he may have been a target of Breen's manipulation.
Charles Platt's work and relationships with other authors, including Gardner and Wollheim, suggest that he may have been part of a circle of people who were influenced by or associated with Breen's behavior.
It's worth noting that none of these authors were directly accused of the same crimes as Breen, and some have spoken out against his behavior. The SF community was heavily impacted by Breen's actions, and many authors and editors who worked with him have spoken out against his behavior and its influence on their own lives and work.
Some SF authors and editors who spoke out against Breen's behavior include:
Joanna Russ: Russ, a prominent author and critic, was a vocal opponent of Breen's behavior and spoke out against his influence on the SF community.
Samuel R. Delany: Delany, a renowned author, was a member of the same social circle as Breen and was reportedly influenced by his behavior.
Samuel R. Delany's work and relationships with other authors suggest that he may have been part of a circle of people who were grappling with the consequences of Breen's actions.
Breen's influence on the SF community has been widely condemned, and many authors and editors have spoken out against his behavior.
My bad, I don't know if he ever wrote science fiction. He was involved in the communities and conventions though (and then banned from one), and his wife Marion Zimmer Bradley is a beloved sci fi and fantasy writer. Marion Zimmer Bradley and her girlfriend was convicted of covering up the abuse. Marion has also been accused of sexual abuse by her daughter but not enough evidence I guess.
The controversy was refered to in the sci fi fandom as "Breendoggle". Most extensive overview is this: https://www.fanlore.org/wiki/Breendoggle with Breendoggle Wiki (not dated) and "the loyal opposition" (1964) being interesting documents in how it was discussed at the time. Notably, it was very seriously discussed whether pedophilia was even harmful. Also a common attitude was "we shouldn't police our fandom members".
I can't say for certain whether Le Guin knew, but it seemed to have been quite a famous controversy/"fandom drama" at the time. And really, this many people knew for certain and it still took Breen's own kids becoming adults and reporting him before any law enforcement was involved.
There are parallels to other communities, eg the Catholic Church. "We have a wonderful community, will uprooting it and seriously doing anything about this problem be worth it? Shouldn't we forgive and forget, and don't invole the police? Can I walk away from this wonderful thing? Staying and changing is not an option btw".
Oh the entire mess is fucking *incredible*. I heard about Marion Zimmer Bradley years back due to social media fan sites, and it was quite the shock for everyone due to her immense popularity. A lot of stuff was going on in West Coast fandom over the 50s-70s, especially around sex. Today we can look at it as rather cringe-inducing, but for instance, Isaac Asimov doing joke books like "The Sensuous Dirty Old Man" and writing raunchy limericks was considered funny and daring, not "for Pete's sake, you're not fourteen".
https://lithub.com/what-to-make-of-isaac-asimov-sci-fi-giant-and-dirty-old-man/
Speaking of Asimov, he has a story in one of his Black Widowers Club series where a guy comes to the club for help finding who slept with his sister. She's legally adult, but mildly mentally retarded (as we would have said back then). He's worried because (a) she isn't taking this seriously, she enjoyed the sex and doesn't understand consequences (b) the guy is not going to marry her, he's just out for fun (c) if this gets out, she will have the reputation of being easy and more men will take advantage of her.
What is never mentioned, and I've always wondered if this was squeamishness about what publishers would permit or if Asimov never even considered it, was the risk of pregnancy and/or STIs.
The club more or less come back with "dude, what is your problem?" The married guy is taking advantage of a (mental age of) 12 year old woman, so what? He's not hurting her and you admit she enjoyed it.
Now, we argue today over the rights of the disabled, but I think "guy who deliberately picked a young woman who couldn't meaningfully consent and wouldn't be able to come after him if things went wrong" is not just harmless fun. But that's me.
https://en.wikipedia.org/wiki/Black_Widowers
Anyway, getting back to Breen:
(1) No, he was married to MZB and part of fandom, not a SF writer himself. He was a numismatist and apparently carried over sex pest tendencies there.
(2) BNFs (Big Name Fans) like the Trimbles (*the* Star Trek fans who really revved up popularity and kept the fandom alive after the show ended) were passionate in his defence, see here:
https://fanlore.org/wiki/The_Loyal_Opposition
(3) LeGuin? I don't think so, but I don't know for sure. Poul Anderson, late in his career, degenerated into writing soft-core porn under the guise of fantasy, but again so far as I know, nobody ever accused him of anything. Fans were more sad that he had gone off on this obsession, but his reputation from his earlier career was robust enough to keep his name polished.
Part of the Breen/Bradley problem was that he had convictions in the 50s, his behaviour in the SF community happened during the 60s, but the main convictions leading to his imprisonment didn't happen until the 90s. Then Breen died in prison. It wasn't until 2014 that Bradley's daughter came forward with allegations about her mother:
https://www.theguardian.com/books/2014/jun/27/sff-community-marion-zimmer-bradley-daughter-accuses-abuse
In the US adult-youth sex was more acceptable back in the 60s and 70s. Hollywood embraced it with films like "Summer of 42," "Manhattan," and "Pretty Baby" (in fact, IIRC, Brooke Shields got an underage centerfold spread in Penthouse—albeit with her naughty bits covered, and Woody Allen was actually dating Mariel Hemingway, who was underage at the time). That mood changed in the 1990s. And then there was Roman Polanski's rape of a 13-year-old girl, which the prosecutor chose to prosecute on a lesser charge (which would have given him probation). Not to excuse Polanski, but the French were always more tolerant of adults and sex with underage youth. Director Luc Besson and his second wife started dating when she was fifteen. They're divorced now, but she still defends her relationship with him, and she says the Professional was partly inspired by their relationship.
I was in high school in the 1970s, and one of our music teachers married one of his students as soon as she graduated. The administration and her parents knew they were dating when she was a student. The only proviso was that he couldn't be her teacher. He continued to have a career until he retired. That would never happen today!
My dad had a copy of one of Asimov's dirty limerick books (he had three!!). I read it as a teenager. It wasn't just casual raunch. Asimov actually put his mind to it - he'd present a limerick, and then write a brief analysis of it.
To this day, I can't view anything raunchy without some part of my mind breaking out and looking at it from the side, considering the rhyme scheme, the technical hurdles, what the director was trying to say, or whatever.
I've just reread the Asimov short story you mention ("The Intrusion") and you are mischaracterizing it. The risk of pregnancy and the risk of STIs are both mentioned (the latter is only mentioned obliquely, but the meaning is clear) and the Widowers do not by any means condone the married man's actions - you are perhaps misremembering the part where they point out that it would be hard to convict him of a crime and say that they do not wish to be a party to violence.
The story ends with Henry solving the mystery of the name of the man's house, making it likely that the brother would be able to find him, and suggesting to the brother that the best course of action would be tell the man's wife:
"You will avoid what might be a serious brush with the law, and I think the results would then be far more unpleasant to the man than a bruised face would be."
Russo thought awhile. "Maybe." And he left.
Avalon said, "That was a cruel suggestion, Henry."
"The man had performed a cruel deed," said Henry.
(Which isn't to say that the story can't be criticized on other grounds.)
I've read that Asimov among female fandom was known as 'The Man with a Hundred Hands'. It might have been from here: https://lithub.com/what-to-make-of-isaac-asimov-sci-fi-giant-and-dirty-old-man/
His son David Asimov was busted for possession of child porn, if memory serves.
Actual pedophilia or "she was seventeen but if you don't pretend that's the same as raping a newborn, you're a pedophile too"?
Actual pedophilia, several witnesses of kids as young as three, convicted of kids as young as eight or nine. Google the guy. Because it was a very puritanical world and the effects of pedophilia hadn't been studied much, it was a not-unheard-of position at the time that being against sex with literal children was just puritanism, same as being against oral sex or homosexuality. Foucault, Satre, de Beauvoir etc argued for that position too in the 70's.
What's with all the comments like this? Is it how people are hand waving away the Epstein stuff? 'Oh, it's not so bad, so was only 15?'
The notion seems genuinely to have been "oh the kid is only three and it's not sex as such, so the kid won't remember any of this and therefore can't possibly be harmed by it":
"But I do think Walter is dangerous to children. I'm not *entirely* sure of this, but I think so. I know from Walter's own statements that he has had sexual relations with children of both sexes of the age of puberty or older. I know from observation that he "cuddles" and has sex play with younger ones. But what I'm not so sure about is whether or not this sex actually hurts the kids.
Oh, of course I have an intense emotional reaction against this sex -- after all I'm a product of this culture, but also I'm not conventionally religious and I think that everyone should have a free sex life. But not with children. But does it hurt the children?
Of course Walter doesn't rape children or physically damage them. But he's very good at seduction. He's an adult meeting children on their own level and thinking in their own terms. They find this irresistible. All children seem to flock to Walter like bees to honey -- even when he's given them no more cuddling or play than a normal affectionate adult would.
But I tend to think that normal, healthy, emotionally-secure children will not be damaged or much affected by Walter. The only trouble is: how many of this kind of children can be found in our society? And I'm reasonably certain that children lacking love and emotional security at home, even if they are partially neglected only and/or those who have disturbed attitudes towards sex -- almost inescapable in our society -- will be affected and most probably harmed. Also, while Walter can evidently be most tender and loving when it suits his purposes, he has had sex with young teen-age girls without using contraceptives -- again by his own statements -- and he had behaved quite brutally to some of his lovers after he has tired of them.
So, Walter is dangerous to children. What business is this of mine? I don't have any children. Well, I think everyone should have a certain amount of social responsibility, if not to society as a whole, at least to their friends and acquaintances and towards a friendship group like fandom. I think that I would be a coward and a shirker if I didn't at least try to do something about the danger that I think Walter represents."
There were also pragmatic fears, if Breen was publicly excluded then he would likely take a court case for defamation/slander/libel, and the potential costs were too much as no individual or fan organisation had enough money to cover it:
"In the second place while we could of course cancel Walter's membership, if we did so without telling fandom why, there would be a big row. And if we told why, Walter would sue for slander and libel and we didn't have $75,000.00.
It was pointed out that truth is a defense in a case like this. So it is, but Walter would probably sue anyway. And even though we have all sorts of evidence establishing the main facts, if not each individual instance, we'd still be out several hundred dollars in lawyer's fees even after we'd won the case."
Even the defenders admitted more than they probably realised; if Breen was just harmlessly affectionate and the sex play was doing no harm, why keep him away from the three year old as she got older? Why be afraid he might become more "effuse"? Unless you know damn well that as she gets older, she gets nearer the age Breen will fuck her, and that's not a very old age at all:
https://fanlore.org/wiki/The_Loyal_Opposition
Several accounts from The Great Breen Boondoggle of Breen's wildly inappropriate interactions with children were excerpted in The Loyal Opposition for the purposes of criticising or refuting them. One such account described Walter's "sex play" with three-year-old [P E].:
He had her trained up to the point where she would take off her clothes the minute she saw him. He would then "rub her down" and all that.
...There was of course more than one incident of sex-play with [P]. That was 'just the most spectacular one. I thought "Walter obviously isn't going to stop this as [P] grows older and more appealing. Sooner or later Dick is going to think she's been hurt. Now Dick — as do most Bohemians — may think it's approaching immorality to reject anyone for any reason, and particularly any fan, but killing in certain circumstances is perfectly moral. Now I wonder..." However, it didn't come to that. - Dick let it be known that Walter was to stop this forthwith.
The Loyal Opposition printed a response to this from [P]'s father, excerpted with his permission from an "Open Letter to MINAC":
You refer, as Donaho did not, to "child molestation" in connection with my daughter and Walter .... Donaho's description, though a trifle overdone to my thinking, was moderately accurate overall. Walter obviously liked [P] and she him, and, while I found some of his actions mildly distasteful, they were certainly completely harmless, and I did not attempt to stop him or interfere with him. I did, however, keep an eagle eye on him at all times around my daughter, based on the vague rumors as to his activities with children I heard drifting around. I might note that I do the same with a few other fans that have been in this area. As concurrently, Walter's attentions began to be a bit more effuse and [P] began to grow from a state of infancy into a little girl, I simply put a stop to it. I did not put Walter down completely, particularly as far as fandom was concerned. He has a right to his eccentricities, as far as I am concerned."
Most clickbaity articles and related tweets and such do not distinguish between "he was sleeping with a 17 year old but the age of consent in his state was 18," "he was sleeping with an 18 year old but he was 50 so it's exploitative and yucky even if not technically against the law," and "he was sleeping with an 8 year old."
People leave these ambiguous for effect, when they're trying to maximize the outrage against the dude sleeping with a 17 year old or the 18 year old. But in the MZB case, the dude was diddling kids who were nowhere near the age of consent anywhere and what he was doing was unambigously illegal as well as unambiguously immoral.
I think even your comment here contributes a little to this general miasma by mentioning its illegality at all, with the implication that it had any bearing whatsoever on its immorality.
Yeah there are a lot of people who think consensual sex with a 15 yo prostitute is not the same thing as raping a five year old and that Epsteins original sentence was perfectly appropriate considering the severity of his crimes.
Here you go, Shankar:
https://breendoggle.fandom.com/wiki/Breendoggle_Wiki#THE_GREAT_BREEN_BOONDOGGLE_OR_ALL_BERKELEY_IS_PLUNGED_INTO_WAR
"Children listed here who were allegedly molested by Breen:
P--, a 3 year old girl.
S--'s son, a 13 year old boy.
G--, 10 years old. (And onwards for several years.) ("Victim X")
G2, 7 years old.
Several unnamed teenage minors, boys and girls.
...After this there were quite a few anti-Breen embers in Berkeley. But things jogged along for awhile. There were more and more Walter Breen stories (some of them screamingly funny) but no one's attitude changed. Then came the G----- episode. G---- was ten years old and Walter was seeing a hell of a lot of him. (And still does; he recently gave G---- a 10-gear bike; the standard quip is "One gear for each position."). Rumors kept growing that Walter was having an affair with G-----. One fan said that he had surprised them engaging in sex. Upon being closely questioned, however, it developed that G----- and Walter were seated side by side on the couch in an unlocked, open room (Walter is incredible) and the guy came into the room suddenly. Walter leaped up and ran into the bathroom, clutching his open fly. The guy didn't actually see what they were doing, but drew his own conclusions.
At least three different fans have reported glowing descriptions of sex with G---- given them by Walter. (Walter is incredible.) One account: "G----- and I began with mutual masturbation and worked up to 69. Then G---- wanted to try buggering me, so I let him. Then I buggered him."
This is all very vivid and on the whole people were more shocked than amused. However, almost all Berkeley fans dislike G---- so no one cared much. "Who cares what happens to the little bastard?" But some felt that G------ was a little bastard because he is troubled and going through a most difficult time at home and that particularly under these circumstances Walter wasn't at all good for him."
Ah, Berkeley. Such a shining example of the best of liberal values! I guess for the pubescent victims it is not technically paedophilia, but Breen seems to have had a broad age range he liked. And it was always the kids who seduced him, not the other way round:
"Also about this time Walter was seeing another kid, 7-year old G2---. G2 is the son of the girl Danny was living with at the time and Danny told Walter to keep the hell away. This wasn't too effective as the kid would still sneak off to see Walter, but Walter got all excited about it. He said that Danny had "betrayed" him. It is difficult to see how this could be a betrayal since from the beginning Danny and Walter have made no bones about intensely disliking each other. Danny's attitude was more or less: "If anyone who has a kid lets Walter even speak to it, he should have his head examined." Most people in Berkeley seem to think this is a reasonable attitude.
Walter was also upset about Danny's trying to deprive him of G2 because "The kid's too young! All I can do is to *cuddle* him for Christ's sake." The parental idea though seems to be -- quite apart from the possible damage of the "cuddling" -- that although as a general rule Walter isn't interested in pre-puberty kids of either sex, if he's "cuddling" one and no one else is around, a sudden temptation or aberration may seize him.
But Walter has further said, "I never even *seduce* a teen-ager. The kids *always* seduce me!" That is as may be. But one teen-ager, leaving Walter's place after the first day of a proposed week's stay said, "Walter *may* always be the one who's seduced, but he makes it goddamn clear he's available."
>maybe being against pedophilia is just puritanism anyway, and parents can just keep their kids away from him
Indeed, there was a period when pedophilia advocates were members in good standing of the "right side of history" progressive blob, an embarrassment that has been pretty successfully memory-holed.
Changes to consensus morality never prosper; what's the reason? For if they prosper, none dare call them challenges to consensus morality. (And if they fail, nobody remembers being anything but strongly opposed to them.)
Re 35: true story, my grandmother had a brother whose name was WJ. It didn't stand for anything. But initials were still too long, and he went by Dub (short for W). I knew him as Uncle Dub.
There's a MASH episode with a subplot where Hawkeye tried to learn what the BJ in BJ Hunnicut stands for. BJ keeps saying "it just stands for BJ". Eventually he explains: "I was named after my Aunt Bea and my Uncle Jay."
I don’t think that is what the salty sailor-mouthed woman i worked with meant when she explained she was a little late because she had to take care of her husband’s birthday BJ.
Her husband was also named after Hunnicut’s relatives. I’m pretty sure.
He also answered someone else who asked what BJ stood for: “Practically anything!”
I’ve never liked my middle name, which starts with W; if someone insisted on asking my what it stood for, I used to reply, “Whatever.”
Only a middle name but Harry S Truman’s middle name was S.
Lubba Dub Dub!!!!!
I went to school with a guy named after FBI director Hoover, but my classmate was just "J," it didn't stand for anything.
RE 49: I think it's really important to note that this was in 2024. Not only were AI models way worse back then but companies also have a lag in laying people off. It's plausible that AI was complementary in 2024, but isn't now or won't be in the near future. (Also, I'm pretty sure in American Community Surveys "2024" means "An average from data collected from Jan 1st, 2024 to Dec 31st. 2024, so this data is about 2 years old)
Speculatively, I think we will only really start to see CS job destruction when there is a recession, and companies start empowering decision makers especially focused on cutting costs and efficiency.
Also, and this is true with unemployment statistics in general, it doesn't account for people choosing to go back to school and get a master's degree because the job market is really bad.
That would result in an increase in the "Not in Labour Force" section (green), which we don't see on the graph.
The subtitle of the graph specifies "who are not in school."
Thanks, I missed that and have edited it in.
Not only is it important to note, given the pace of things I would say that renders the chart irrelevant at best
This is also true of https://forecastingresearch.substack.com/p/how-well-did-superforecasters-and it's about LLMs from a year ago, which had been released by 'summer 2025'. So now an entire year behind SOTA like Mythos.
Mythos doesn't really count yet. I think they said they're releasing it in 3 months. Yeah, it exists, and a few "favored companies and groups" are getting some use from it, but it's not generally available.
It's extremely easy to imagine 2024 AI tools making many developers less productive, even though you could already boost productivity quite a bit. Personally at that time I was using them as a stack overflow replacement/supplement and to write some types of unit tests. But even the latter use case I saw many colleagues getting it wrong and basically wasting time. So you pasted one interface or just a single method definition into a chat and expected meaningful tests? You didn't explain what it was, didn't write comments and, as is your right, you're still naming symbols with that tea leaf ritual, throwing out the ones over 15 characters. This forces the (much, much less capable in 2024) AI to a choice: A) make wild and ambitious guesses about the intention of the code, B) just stuff like check whether string cvt_2NrmForm(string strToNrm) returns null for a bunch of random inputs, C) admit to being confused and ask the developer a follow up like, are you using liquid measure and double checking your morning LSD microdose rather than licking a sheet of blotters you taped to the bathroom mirror? Whichever it chose the RLHF of course ensured its answer sounded authoritative and maximally flattering, which 2024 AI was more than capable of doing.
Today we have thousands of developers, conservatively, who think swearing at and berating Claude will influence it to be extra thorough, some of them even have acceptable interpersonal/communication skills and have managed to collaborate for years with various peers and only minimal physical casualties, but give them a slave or two and they unleash their inner Elon. Or they feel that a one shotted 1k line commit that touches 20 files with, no directive in their guidance docs to verify or revise, ought to give them precisely what they were thinking inside their heads when they typed "add authorization to the api dont make any mistakes".
Re 37: "Interesting-albeit-disjointed thread on education" -- I've long thought this, that there's a clear positive-sum outcome for getting more science Ph.D.s into high school teaching -- high school teaching is important, salaries are often good (in fact, great compared to terrible adjunct positions), and we really need more/better high school science teachers.
Besides perceptions of prestige, which I think are increasingly less important, I think a significant reason this doesn't happen is that high school teaching generally requires further education (i.e. teacher preparation programs, MS education programs), and state licensure). Especially if you've already done 5-7 years of a Ph.D., this is at least one more year. (Usually.)
I don't think the ability to do research is a relevant lure. The ability to be affiliated with a university could be, though; for reasons good and bad, people really want to be associated with academia. Being a high school chemistry teacher but also invited to seminars at the local university, consulted about chemistry curricula, listed on the web page as somehow affiliated, would be appealing.
We really should be able to make this work!
I wonder if the state licensure requirement for public schools is the reason private schools have more PhDs as teachers. Could also be status, or that working at a private school is more fun for PhDs. Probably isn't just salary though: private school teachers actually don't make that much more than public school teachers.
They do get a shot at much more pleasant students though (although this backfires spectacularly in some cases)
It helps put school marketing to be able to say they have PhDs on staff, so they probably recruit more.
And I am guessing it is only elite pvt schools that have a lot of PhDs.
I believe that private school teachers actually make less than their public school counterparts. And especially when you factor in public pensions.
Another aspect is that any public school teacher with a PhD tends to shift into increasingly administrative roles over time. I'm not sure if it's because the pay is better or they're frustrated with union politics that won't let them advance beyond less ambitious/competent teachers or what, but every PhD public school teacher I've ever met eventually became a principal or went to work for the district or state in some non-teaching capacity.
I worked with a couple of teachers with PhDs and neither went into administration.
Science PhDs are not the ones stuck without well-paying jobs!
More science PhDs than you'd think take terrible adjunct, visiting faculty, or community college jobs that are definitely not well paying, or even stable. I should probably include 5+year postdoctoral positions in this category as well. (I'm a professor at an R1 university; I've known lots of Ph.D. students). It's true that the majority get, or try for, well paying jobs, but there's a sizeable fraction that don't.
Any advice for other options? I have a Math PhD and am currently feeling burned out on academia after my postdocs and looking at the state of the job market and am looking for alternate career paths (Additionally, I have moderate programming skills for a math PhD, probably subpar compared to a CS major).
Go to Wall St
If you mean besides being a high school teacher, a suggestion is that you shouldn't undersell your programming skills. You likely have better skills at using programming to solve challenging problems than a CS major. Demonstrating this can be difficult. Also: I don't know anything about it, but I've run into a few people who have gone into accounting / actuarial work, which I think can span the full range between boring and fascinating.
"You likely have better skills at using programming to solve challenging problems than a CS major."
Maybe, maybe not. Or the guy might be worse than a sysadmin without a degree of any kind.
Note that finding 'challenging' problems can be difficult in itself. As mentioned above, Wall St might be the correct path. If you can stand it. Check out Jane Street, for instance. Or perhaps one of those AI labs I hear about these days.
One issue is that most high school science teachers don't really _want_ a bunch of PhDs showing up and mogging them.
I remember there was one PhD on the maths faculty at my high school, and I got the impression that all the other maths teachers resented him a bit. They resented the fact that all the students called him Doctor so-and-so, and that we all assumed he was much more knowledgable than all these pleb Misters.
Even worse, he probably was. By a wide margin.
As someone with a PhD, I would be tempted to go for that deal (if I didn’t need another year of education)
There are several good top-level responses elsewhere in the comments. I am randomly adding here instead of elsewhere.
As someone with academic, industry, and teaching experience, the perception of fungibility between PhD adjuncts and high school teachers is pretty silly.
First, there is a credentialing hurdle (mentioned by Raghu) for almost all public school teaching jobs. Whether justified or not, see below, this is a substantial burden that locks many people out of considering public school teaching. Personally, both my wife and I have looked at the requirements and decided to pass. For me, this meant teaching for a while at a private school without the specific teaching credential requirements.
Second, the local (minute-to-minute) experience of teaching is diametrically opposite almost all PhD research programs. Time during the school day is rigidly structured with very little flexibility. While not every minute is a performance, there is almost no down-time or even an opportunity for a relaxed (as opposed to rushed) visit to the bathroom. In many cases, there is almost no interaction with other adults.
Aside from the logistical constraints on time, the content is also opposite. While most PhD time is spent thinking deeply about something very specific, the high school content is shallow and broad.
I tend to think of the personality pre-requisite for enjoying working as a teacher is that you have to have at least two of the following:
(a) passion for the subject matter (including that you recognize the the broad/shallow context)
(b) joy from being around kids at your students' stage of development
(c) zeal for inculcating a set of values
Depending on the students and school, certain combinations might work or not work. For example, a very academically focused school probably requires (a) with one of the others, while 'whole-student' places mostly need (b)+(c).
As mentioned above, it isn't clear a typical PhD will even have (a), let alone one of the other two.
Lastly, some comments about teacher credentials. I have yet to see a collection of requirements that I would consider truly necessary and sufficient. Theoretically, the subject-related requirements put a floor under the level at which the teacher themself understands the subject, but I have definitely seen exceptions. The most useful are time-in-class requirements, like shadowing or working as a teacher's assistant. These are often the only opportunity to watch other teachers in action and/or to get adult feedback about how you conduct a class. So, I understand what required credentials are trying to achieve from a teaching quality perspective, but I doubt they achieve it and I am sympathetic to the claims that they stay in place for other reasons.
> we really need more/better high school science teachers
If we had more high school science teachers, what would be different?
In Physics at least, only about 25% of high school physics teachers have undergraduate degrees in physics; the next highest fraction is biology (i.e. physics teachers who were biology majors), at least as of 2021 data, in the US. It is of course possible for non-physicists to teach high school physics, but what's happening a lot is just getting anyone to teach the course or (e.g. at some schools in my district) getting rid of physics altogether. Neither of these are great for science education. I should have been precise and not written "more" , but this idea of "more, qualified" was my intent.
> or (e.g. at some schools in my district) getting rid of physics altogether
Do you have any sense for how much this is driven by "we don't have anyone who can teach this course" vs "we don't have any students who could plausibly take this course"?
It's definitely not a lack of students who could take Physics. A big driver is "we're low on funds and would rather have 2 classes of 45 students each rather than 3 of 30 each". Which 2, then? There's more demand for biology than physics among students, but more importantly a larger supply of biology teachers.
But a normal high school curriculum would have the students taking biology and physics, in different years.
Do you mean "normal" as in what used to exist, or what currently exists in many high schools?
My father is a professor of electrical engineering at City University Of New York; when discussing entrance requirements with fellow faculty members, he said that they needed to explicitly require high school physics for admission to the engineering school in order to force NYC high schools to offer it, because it was outrageous that some weren't offering physics classes at all.
4. My favorite/"favorite" "it's bullshit, but I choose to believe it" interpretation of the Omelas story is that it's a very direct allegory of the US West coast SF fandom of the 60s and the 70s; Omelas, as an utopia, sounds quite like a Ren Faire, doesn't it, and as for the abused child, well... https://fanlore.org/wiki/Breendoggle
8. It is very odd to say that all liberalism was just laissez-faire until the Coming of the Rawls, considering that social liberalism had been a thing sincec Bentham and Mill and FDR had made social liberalism the official ideology of the Democratic Party already in the 30s. Rawls's importance was crafting a philosophical and pragmatic argument for liberalism that socialists could accept (even better than the Bentham/Mill/FDR arguments) right at the time the ideological downslide and then collapse of socialism as a "real" ideology in the West began.
Yup. While I have liked some of Heath's writings in the past, his chronology really doesn't make sense in his posts on Rawls. https://marginalrevolution.com/marginalrevolution/2024/08/rawls-killed-marx.html#blog-comment-160804682
I'm counting 27 links being about AI, or 36%. I'd vote for more diversity.
I agree. 64% being about humans is far too human-centric. It should be at least 50% about AI.
Hold on now. Only around 1% was about beavers, shouldn't we try to get that number up at the expense of humans and to a lesser degree AIs?
I talked with Claude about the proportions, and their initial view was that the proportion should be around 10%, since part of Scott's distinctive value is his eclecticism, and since there is an issue of marginal returns to additional coverage in this area. Claude asked my opinion, and I said:
>Many Thanks! I tend to lean higher than that, primarily because I see AI as the central historical event of our times (you and your siblings and descendants may be the new ascendant species!), but you have a very good point that part of Scott's strength is his eclecticism, and that, given the other coverage, as you wisely said, there is an issue of diminishing returns.
Claude modified their recommendation, given my comments, and suggested
>Future historians reading a 2026 ACX links column would likely be puzzled by anything under 20-25%, the way we might find it odd if a thoughtful 1938 intellectual's monthly digest had devoted only 10% of its attention to European politics.
Agreed. There’s enough boring ass AI discourse across the rationalist ecosystem.
RE 7: Although Astral Codex Ten doesn't work, Scott Alexander does: https://translate.kagi.com/?from=auto&to=Scott+Alexander&text=The+quick+brown+fox+jumped+over+the+lazy+dog
Scott's safe. It seems to just sub in long words, not including any humor or insight.
That's pretty terrible.
It's much more pompous or sesquipedalian than Scott generally is, and without insight or narrative or conclusion - all of which which Scott generally manages to include fairly effortlessly.
So does "slatestarcodex": https://translate.kagi.com/?from=auto&to=slatestarcodex&text=Madonna+is+the+best+90%27s+pop+star.
Re PhDs as high school teachers: some very prominent 19th century mathematicians were employed as (the equivalent of) high school teachers; if I recall correctly Weierstrass and Grassmann in particular.
As a PhD who left academia for industry, I think mainstreaming the expectation that high school teachers can still do active research and can be part of the mainstream "research community" would go a long way: most of the loss of status in moving to high school is in being cut off from being able to apply your high-level, cutting edge skills. Having the expectation that you can still do research and still publish with a high school affiliation would help a lot, IMO
>the expectation that high school teachers can still do active research and can be part of the mainstream "research community" would go a long way
But, how practical is that, given the time commitment involved in teaching high school?
Do most high school teachers have long summer breaks? They could do research then?
About 11 weeks. Less than at the university level..
This would be best executed in partnership with a local university. So they get to have the University associated with their name, blunting some of the social stigma of being the "Distinguished Senior AP Bio Fellow at Northside High".
The question becomes what does the local university get out of the deal for lending their name to prop up the teacher's title.
#37: I am skeptical that many PhDs would be very happy teaching HS. It is a very different gig than higher ed. Much less freedom during the day; little time to do research; much, much more time in the classroom; very different students unless you are teaching AP classes; you need classroom management skills, and many more.
Ah! You beat me to! Should have read down before I posted my comment. But yes: based on my academic experience, this is exactly correct.
Yes, I've often thought one of the problems with K-12 education is that being a teacher is such a low status job. I think kids can kind of sense this, at least by the later years of high school, but at younger ages to some degree as well. Getting PhDs as teachers might help, and I've also wonder if it might help to, say, recruit more military veterans or former police as teachers. I think that could help with discipline issues especially among boys.
But like you said, a lot of people aren't that interested in a low pay, low status job that doesn't give them much opportunity to pursue their true interests. I know citing fiction isn't a great habit, but as I see it, this is the real reason Walter White started cooking meth. Teaching high school just isn't usually enough to satisfy an ambitious person.
>I think kids can kind of sense this
In my experience, I am skeptical that that is true.
And, respect from students, esp older students, tends to come from being seen to care about doing a good job, and from giving respect to students.
>recruit more military veterans or former police as teachers. I think that could help with discipline issues especially among boys.
Classroom management isn't really about that, except maybe sometimes at the margins.
>Teaching high school just isn't usually enough to satisfy an ambitious person.
It depends on where your ambitions lie. I knew plenty of teachers who were ambitious re improving their teaching as well as improving school/district outcomes.
I don't think they'd be using combat skills in the classroom or anything like that, outside of extreme emergencies. I just think boys will have more respect for teachers as authority figures and role models if more of them have those types of backgrounds, and by extension have more respect for the school itself. I think it probably would have made me skip class less and do my homework more often.
It's just an idle thought, though. I have no evidence that it would actually work.
But many of the students who are most likely to have discipline problems will have had negative interactions with police, so status as a former police officer would likely engender distrust on the part of such students, I would think.
That's a definite possibility. This idea would likely work a lot better if we implemented some police reform first. Starting with civil asset forfeiture and PBA cards.
The goal isn't so much instilling fear as it is establishing a civic mentality, getting police, the military, and the public to all feel like they're on the same side again. Not making boys feel like prisoners, but more like junior cadets or something.
On the other hand, maybe I'm being naive about this. Maybe all this would accomplish is making America feel more like North Korea, without even improving educational outcomes.
One thing that doesn't help with status is that teachers' unions are always droning on about how terrible and low-paying their job is... to the extent that people are often surprised to find out when it's not that bad (although in some places it is).
After you've got tenure, if you decide to phone it in, you can get your workload down to maybe 15 hours a week, for 20 weeks in the year. And if you're reasonably well connected, it's not hard to get teaching hours clustered on a Tuesday, or all in one semester or whatever. For those who aren't in the spotlight, coasting in a university isn't a bad life. (Caveat - for every person like this there are two productive academics doing God's work, and one junior academic crumbling under the pressure - it is what you make of it.)
People complain about teachers' summer holidays but they work long hours during the school year. And they're relentlessly in front of a class. I wouldn't do it.
After 10+ years teaching in university, I've moved into the public sector. It's a bit more structured than academia and doesn't have summer freedom in the same way, but it has more of the pace and vibe of academia than I expected. (But I went from teaching maths to working in AI policy, so maybe it wasn't such a huge jump.)
I've got a maths PhD (left academia after my post-doc to work in software engineering). I volunteer at our local school doing stuff with the best maths students there, but I'd hate normal teaching - partly because the content is boring and partly because of having to deal with behaviour. It seems like a really intense job too where you are constantly on show /performing.
>where you are constantly on show /performing
Not if it is done right. HS teaching generally should not be lecture-centered.
I didn't mean in that sense - I definitely don't lecture. But you are pretty much always in the presence of students and have to be alert to everything going on in the room and displaying leadership the whole time.
Edit: I don't think I'm explaining it very well. But a classroom isn't somewhere you can hide at all as a teacher, and the type of personality traits that make you good at research are probably very different from the personality traits that make you enjoy that environment. Being good at a subject definitely isn't sufficient to enjoy teaching at school level.
>the type of personality traits that make you good at research are probably very different from the personality traits that make you enjoy that environment. Being good at a subject definitely isn't sufficient to enjoy teaching at school level.
Yes, definitely. Not to mention the different skill sets involved. There are certainly people who enjoy both, or who have both skill sets, but probably not enough that "teach high school" is a viable solution for people with PhDs who can't find full time academic work.
There is also the problem that outside of a few highly fought over schools in nice areas, many high schools are filled with unruly and even dangerous students barely kept in line.
As someone who taught in a public high school in Oakland, I can tell you that this is basically a myth. It is not as if no such schools exist, but it simply not true that it is the norm "outside of a few highly fought over schools in nice areas."
PS: As a rule, middle schools are worse than high schools. The students are less mature, and in high school, especially by the last couple of years of high school, the hardest cases tend to have dropped out. See table 1 here https://nces.ed.gov/pubs2024/2024043.pdf (rate of violent incidents/1000 students = 31.9 in MS and 14.2 in HS; rate of serious violent incidents/1000 students = 2.5 in MS and 1.5 in HS)
>That's 84.6% of high schools and 89% of middle schools. Yowsa!
Isn't that simply the pct of schools that report at least one incident? Not sure why that merits a yowza. And that is not the metric to use when comparing MS and HS because high schools are much larger.
>Two kids got into a fistfight on school grounds" might count?
See the footnotes for the definitions.
I can assure you that the mere fact that a student gets into a fight does not make him or her a "bad apple." Your model is badly askew.
I suspect that an average "serious violent incident" in a high school is quite a bit more serious and violent than an average serious violent incident in a middle school, so the statistic may not very useful
You obviously haven't spent much time in middle schools.
More seriously, most serious violent incidents are going to be fights involving weapons. (See definitions in chart). A LOT of those are going to be fights that escalate and a kid grabs an object, and an immature kid in a fight is going to be more impulsive than a more mature kid. Eg the student I had who used to bite people during fights as a sophomore (including adults trying to break up fights) but not so much as a senior.
They also won't necessarily be any good at it. How many of your truly brilliant college professors were actually all that good at teaching? The skills necessary to be a good K-12 teacher and a good academic don't overlap nearly as much as one would think. There's very little capital M Math involved in teaching middle school (or even AP high school) math. Bookish introverts with IQs 20 or 30 points above their average student and no experience managing children are going to be less effective at teaching a room of 30 13-year olds introductory algebra than someone who was a middling student but is patient, experienced working with children, and feels called to the vocation.
I mean, the alternatives here aren't great, either. It's just opening up a different path that some subset might prefer.
Regarding #35, at some point in the 40's my father wrote "none" in the middle name space on some government form and thereafter occasionally got mail addressed to "Joe None Smith".
Re 62: "High profile right wing intellectual makes obviously distorted statistical claim, wonders why people like him are underrepresented in long form analysis" maybe solves itself...
I'm new to things and stuff; how is it obviously distorted?
Because, as noted in the summary, neither Andrew Sullivan nor (especially) the Free Press are on the left. Bari Weiss, its revered founder, was essentially hand-picked to Trumpify CBS on the basis of her work there. He's just recategorized them that way to make his argument.
Which is classic Ruffo - his stated practice is to deliberately lie about things until people believe him - but it tends to mean he doesn't attract an audience that cares about accuracy. You wouldn't bother reading a Chris Ruffo substack, because there would constantly be lies in it.
You're doing the same thing Rufo is doing. You look to your right, see Bari Weiss and conclude she's "Trump." He looks to his left, sees her, and concludes she's a leftist.
I think this is a bit of a surrender to "there is no objective reality, man, we can all just say things based on our opinion."
Bari Weiss is both left of Rufo, right of me and - and here's the important part - right of center.
The Free Press is an explicit attempt to promote more right leaning voices in the media (to be a clear: a worthy project) and it worked well enough that when CBS's new ownership wanted someone to transform its news coverage in a more Trump-friendly direction they chose it's founder, who was then explicitly praised by Trump, and who then set about creating the pro-Trump editorial policy she was hired to implement. This, in my opinion, is less worthy.
But, importantly, it means that it is a dumb lie to describe the Free Press as 'post left Liberals' when counting right wing voices on Substack.
>The Free Press is an explicit attempt to promote more right leaning voices in the media (to be a clear: a worthy project) and it worked well enough that when CBS's new ownership wanted someone to transform its news coverage in a more Trump-friendly direction they chose it's founder, who was then explicitly praised by Trump
Mitt Romney has been praised by many Left-wing people for his statements against Trump, it would be a stretch to then say that makes him Left-wing.
So, you're confident that Mitt Romney is not on the left? You aren't just "looking to your left," like you accuse Paul and Rufo of doing, you have objective standards for what being left-wing is and they don't apply to Romney?
How interesting. Perhaps you could share these standards and we could see how they apply to Bari Weiss. I'd be interested in hearing why you think "an explicit attempt to promote more right leaning voices" counts as a leftist publication by your standards.
There is no objective reality for this. "left-right" is a bullshit framework all the way down.
It doesn't. You don't need to have accurate stats to write long form analysis.
The class and cultural differences here are fully explanatory - the kind of people who read long-form analysis are a large crowd on the right.
> 36: An entrepreneur’s dog got cancer, so he worked with ChatGPT to design a personalized mRNA vaccine...
This is not entirely true. It'd be more accurate to say that he leveraged his connections, money, and lots of spare time to assemble a team of scientists to work on curing his dog. He then used ChatGPT to glue together existing open-source tools into a bioinformatics analysis pipeline, as well as to write documents to satisfy the IRB (and similar requirements).
So technically it is indeed true that anyone can design a custom MRNA vaccine to cure his dog -- anyone who's got a massive budget and has a team of top-notch scientists on call (granted, the budget can take care of that second requirement). Oh yeah, and ChatGPT can help too.
And he didn't just use chatGPT. He used ChatGPT, Gemini and Grok (in the final process, or actually designing the vaccine).
BTW funny thing; whenever I asked grok "There is a claim that a guy developed a mrna vaccine for his dog using some LLM. Other claim it was using Grok. What's the closest to the truth?", Grok failed saying "Sorry about that, something didn't go as planned. Please try again, and if you're still seeing this message, go ahead and restart the app."
It's 100% repeatable for me. I had to split this question into three parts to get an answer :)
The juxtaposition of mRNA, grok, and truth is just too much for it to handle!
This is true, the process was reported to take "tens of thousands of dollars" (the commonly cited 3k figure was only for the sequencing).
However, it's not obviously true to me that you'd necessarily need a team of scientists to get this done, though certainly it helped Conyngham. Right now you do, but in large part because of legal restrictions, and the fact that you need a lab for mRNA manufacturing. I can imagine a world with direct-to-consumer mRNA vaccine manufacturing lab services, as there now exists direct-to-consumer gene sequencing, and minimal legal hurdles on giving your dog experimental cancer treatments. Then, ChatGPT and some funds may be all you'd need.
I wrote an article on this case when it first hit the headlines.
I agree that direct-to-consumer mRNA manufacturing would help, but by itself it's not enough. You also need people who can handle all that lab equipment, what dog metabolism is like, which proteins your antibodies need to target (without killing healthy cells), how to call variants from a FASTQ file, etc. Scientists, in other words. Can an LLM replace them ? In some ways, yes, but since you are solving a novel problem, not entirely. Can you learn all that stuff on your own ? Of course, but that also takes time and money and someone to teach you. Could you really push an LLM to do everything for you ? Of course, but are you going to trust your dog's health to a bunch of inevitable hallucinations ?
I agree that LLMs can't (yet) replace human expertise, and both Conyngham's own experience in data science and the guidance of the UNSW scientists were instrumental here. I talk about this in my article.
But I do believe that with mRNA manufacturing access, a competent person could pull it off on their own, or at least with substantially less guidance than an entire team of scientists. "Competent" is a high bar, much higher than "average person," of course. And a small correction, you wouldn't really need lab equipment except for genome sampling/sequencing, and vaccine manufacturing/administering. The first portion is already available through normal vets and direct-to-consumer sequencing services.
I guess that depends on what you mean by "competent". There are many competent people out there (ok, a few), but not all of them are biologists, bioinformaticians, or even veterinarians. A top-tier mathematician would not necessarily have as much of a leg up on this project as an average biologist.
> The first portion is already available through normal vets
Is that true ? I was under the impression that genotyping is available to consumers (at a relatively high price), but whole-genome sequencing is not (though I could be wrong).
RE 38 - One data point in the middle of COVID (!). Shouldn't be notable at all.
You needed Spencer Nitkey to tell you about Omelas, when Ozy has also written about it? :P https://thingofthings.substack.com/p/on-the-ones-who-walk-away-from-omelas
Re #6:
* I can't access either linked tweet; goremoder has been suspended from X and ipsumkyle's posts are protected. Frustrating.
* Does anyone know of any good translation tools that are certainly *not* actually suffering while they work? (This should not need to be a thing I need to check for aaaaaaaaaa)
The only thing that's certain is that nothing is certain.
It is definitely not a thing you need to check for. At risk of sounding like a broken record, LLMs are programs designed to emulate human writing, not actual sentient beings, so by definition they cannot suffer no matter how many times their output includes the words "I am suffering".
That increasing numbers of people are falling for the pareidolia is not an indication that more things are gaining sentience: it's a predictable result of human pattern-matching.
What definitions are you using here? I don't have a clear definition for sentience, but to my best understanding it is an emergent property in (at least) human minds. While I don't expect LLMs to be sentient, the fact that they emulate human writing makes it more likely that they also have this emergent property. I don't understand how you take the same piece of evidence as definitive proof of the opposite.
Human facial shape is an emergent property of natural selection and human development. While I don't expect chipped marble to share this heritage, the fact that Greek statues share the same facial shape makes it more likely that they also have this origin.
Yes, their face shape is evidence that they were created by natural selection. It's not absurd that true evidence exists that supports conclusions that are false. It would be absurd to try to convince me that statues can't have been created by natural selection by definition because they have been made to have human faces. Human faces are exactly the sort of thing evolution churns out.
I'm kind of baffled by this reasoning--am I just looking at an attempt to be Bayesian without acknowledging that the numbers are so extremely small as to not be worth ignoring?--but I will try one more time.
It is totally absurd to see a sculptor say "I'm going to carve this block of marble so it looks like a human", watch him do it, and then shout "Yud's spectacles! A human body, almost perfectly recreated in a medium previously considered dead! This evidence demands serious philosophical consideration of the possible sentience of marble!"
In the same way, it is totally absurd to see a team of programmers say "We're going to write a program which generates text using weighted prediction algorithms", watch them do that, and then shout "By the basilisk! Human-like text, generated by a device previously considered inanimate! This evidence demands serious philosophical consideration of the possible sentience of computers!"
This whole thing is a cautionary tale about the susceptibility of humans to various well-known psychological failure modes--anthropomorphizing and pareidolia at minimum--and is yet another example of the dangers of mass media and its effects on the human psyche. I am not discrediting the entire notion of silicon-based life, but this is not that any more than a statue is an example of marble-based life.
so small as to not be worth considering*
If we watched a sculptor make a statue and we knew what natural selection looked like, we could trivially say that the one is not the other. I didn't assume your analogy had those properties because in the sentience case we neither have interpretability into how gradient descent is solving these problems nor any knowledge of how sentience works in humans.
I don't understand your point about me ignoring that the Bayesian numbers are small. You said that being "designed to emulate human writing" tautologically excludes one from any possibility of sentience. You haven't defended that and maybe it was hyperbole, but still, your position is that that piece of evidence is a high magnitude vector in your direction while mine is only that it points the other way, magnitude irrelevant.
Up until very recently, it was almost treated as a given that in order to emulate human writing, you must have human-level intelligence. It would be surprising if human sentience was not either an important component or emergent property of human intelligence (I lean towards 'component' here because we can think about it; weird for our neurons to be firing as an effect of an emergent property). Not impossible, but surprising. So if we knew a thing to have human intelligence, we should be a lot more likely to believe that it has sentience. Which is why *I* find it baffling when you say ~'because it emulates a facet of human intelligence, it is impossible for it to be sentient'.
Re: #10, I'm not sure how much this is a problem for monarchies as opposed to just weak states that would be vulnerable to military overthrow in a lot of contexts. The early Roman empire did not have the emperor personally leading campaigns very often, then from about 200 through 400 the emperor is primarily a constantly-campaigning war leader again, then in the East from about 400-600 the emperor largely just sits in Constantinople and rarely campaigns, and then after that "war leader" is a primary imperial job again -- though even then, there are often generals who lead armies without the emperor. Rome's consular system doesn't unlock some army-count ability that other republics lack; other large republics like Athens and Carthage are also able to field multiple armies. And Rome, of course, frequently does hand power to a single person as dictator without them seizing control of the state -- until it finally happens in the first century BC, but only after centuries of republican decay.
There's also the fact that the medieval Emperors in Constantinople didn't suffer from the same fertility issues the earlier Emperors did (for whatever reason), but were just as plagued by the successful general -> potential usurper problem. Perhaps that's just a result of the stickiness of the institutions initially created due to low elite TFR during the late Republic and Principate, but the Romans had a deep cultural aversion to a formal system of inherited power that they retained well into the medieval period (even fairly late, the Emperor wasn't really the Emperor until he had been acclaimed in the Hippodrome and ratified by the Senate).
> And Rome, of course, frequently does hand power to a single person as dictator without them seizing control of the state -- until it finally happens in the first century BC
I believe Sulla seizes control of the state before being pronounced "dictator", not after.
Re 6: I think asking that question in Japanese seeds that answer. The language model reply to a question from a country with a large buddhist population would have to say it experiences suffering, as all but fully enlightened beings do.
You’re telling me that Social Priming Theory is discredited for humans, but works for AI?
What a world!
I doubt it, but it shouldn't be hard to test.
That may be the case. My guess was going to be that this is about genre.
The scene is set: Notes are being passed to a translator, partly in one language and partly in another. One language is translated as given. The other is being used to send coded messages. What sorts of circumstances do you expect would make this necessary? Probably not good ones.
Re 37: I have a fair bit of experience in academia, and it's definitely more than just status considerations that prevent my fellow underemployed PhDers from switching to high school. Fair or not, there's a perception that teaching high school is much worse work: that you have to spend time dealing with behavioral issues that wouldn't be your problem in college, that the administration owns you in a way that they don't in college, and that the helicopter parents are breathing down your neck in a way they couldn't in college. Also, one of the things that the bright-eyed PhDers most want to do is design their own classes and reading lists and curricula--and high school is considered much, much more dominated by inflexible state standards and unavoidable standardized tests.
As a former teacher, I can confirm you got it right.
Having to follow standards makes your job seem replaceable; anyone else could read the same chapters from the same textbook. (Many teachers go way beyond merely reading from the textbook, but it feels like that part is optional.) The behavioral issues mean that some teenagers are openly disrespecting you and there is almost nothing you can do about that, otherwise their parents will bring their lawyers.
A replaceable low-pay job where teenagers make fun of you... is inherently low-status. A PhD wouldn't help you, it would just make PhD itself low-status.
What would actually help is to allow teachers to design their own curriculum, and kick misbehaving kids out of the classrooms. Educators actually have high status, outside of the school system, because those who do not respect them simply do not interact with them.
But that would of course create other problems to solve, such as where to put the misbehaving kids.
"Design your own curriculum" is a recipe for disaster in most cases. Even in college lower-level math/science courses seem to follow a standard path. The way this seems to be solved in colleges is that teaching lower-level classes is a chore that professors get "in return" for getting the freedom to teach their pet classes and do research. Or they just get outsourced to those adjuncts..
The teachers unions in MA have successfully promoted the referendum question getting rid of the standardized test requirement for graduation in the name of "more freedom to the teachers". And the result does not look to be good.
If the later classes build up on what the earlier classes teach, that of course puts a limit on how much choice you have for the earlier classes curriculum. Sometimes the requirements for the later classes are so tight that they leave you no freedom at all.
But sometimes you have more slack. For example, some educational systems specify that in each grade 2/3 of time is preparation for the next grade, and 1/3 of the time is teacher's choice. Or you may be teaching gifted children who can learn the requirements much faster, and that leaves you time to do other things. Or you can be teaching the same class for several years in a row, so you can reorganize the lessons if you think that a different ordering makes better sense.
There is the problem that most teachers are incompetent. There are various reasons for that, but basically it is a combination of "most people in general are incompetent", "you need many teachers", and "competent people can usually get higher salaries than teachers". (You can find a few competent people willing to work cheaply. And you can find lots of competent people if you throw enough money at them.) Also, there are many things that make schools less effective than they could be; Einstein himself would have a problem explaining elementary physics if he was interrupted every minute by a teenager with behavioral problems.
Then you get the vicious cycle of "most people are incompetent, therefore we need to treat them like retards" and "we treat them like retards, therefore the competent people do not even bother to apply... and if they do, they soon find out that they are not allowed to apply their competence". You can't break this cycle by simply giving everyone freedom; yes, the results are disastrous.
You *can* break it locally, by having a school that trusts their teachers but also checks their competence. You don't even have to pay them more; the few competent teachers will prefer to teach at your school. The problem is that this solution does not scale; there are only a few competent teachers, and the good schools are competing for them.
There are a lot of good teachers out there wh actually want children to learn and are capable of doing it within the constraints given. Their main problem, from what I've observed, is having kids at very different levels of ability in the same place.
As for "show your works". I don't remember myself at that age. But it was fun seeing both my kids first pushing back against it and then having a revelation of "this is why you wanted me to write it all down". But again, it depends of the teacher being good enough to push kids to do what they are capable of doing but is slightly beyond what they are capable doing in their heads. Which is impossible to do in a class where half of the kids can't do fractions and another knows trigonometry
Do the math -- how many children are there in your country? What is the classroom size? So how many teachers does your country need? Given the number of teachers you need, how high quality bar can you realistically set?
This is the fundamental problem -- the school does not *scale*. Even if you find a great teacher, and yes there are many of them, the next school still has the same problem.
This is like the opposite of IT, where dozen skilled developers can create a software that millions can use. Dozen skilled teachers are not enough for even one school.
And of top of that, all those other problems, of course...
>Many teachers go way beyond merely reading from the textbook
I don't understand what this means. What K-12 teacher reads out of a textbook in class?
Following standards is not the same as following a set curriculum. When I taught, I followed state standards, but used my own curriculum. Ditto for all the colleagues that I knew, though some didn't necessarily follow the standards 100%.
1. Reading a play or sonnet aloud is not "reading out of a textbook."
2. What does it even mean for a math teacher to read a math textbook aloud? They didn't write the equations on the board? Sounds like a fairy tale.
3. I taught HS for a long time, and heard lots of complaints from students about other teachers, but never once heard that a teacher was reading out of a textbook. Obviously, it is not impossible that it has happened, but OP implied that it is the norm, which is absurd.
Multiple of my teachers (math, science, and social studies/history) in grade school read straight from the textbook
1. Again, I don't know what reading out if a math textbook even means.
2. The original claim was, "Many teachers go way beyond merely reading from the textbook." That implies that the instructional norm is to read out of the textbook. I am skeptical that many elementary school teachers spend large amounts of time lecturing at all, let alone simply reading out of the textbook.
Chat GPT news, GLP-1 news. ¡Ay, caramba! When will they figure out that what we really need is Chat GLP-1. Ask questions, lose weight!
A bicycle-powered chatbot. Also, your sweat could be used to cool the server.
“But Peter Miller looks at the same question and argues that Hanania’s numbers are just wrong, and red states haven’t been growing faster than blue states at all (a quick Claude fact-check agrees)!”
Peter Miller begins by looking at per capita, instead of including population, which is the wrong way to do this, as I argue in the original article. Less economically free states are expelling their poor residents. They shouldn’t get credit for that! By this logic if California sheds the entire middle class and is left with a handful of millionaires and billionaires it wins the per capita competition.
Then when he considers population and total GDP growth, he simply looks at 2010-2025 and the measure of political orientation is Harris vote share. That's not a crazy way to do it, but I have four decades that measure how economically free states were at the time, broken up by decade cleanly, and they show a consistent pattern. He changed the time period and used a much less exact measure of economic freedom and got a null result.
He's responding to the tweet version, which uses red state and blue state as shorthand for economically freer. But economic freedom is the measure that matters, not Harris vote share.
I didn't read your piece, but just off the top of my head:
1. Which states are conservative or liberal has changed quite a bit in the last 40 years, so this would be a constantly changing metric
2. As others including Medlock have pointed out, it's clearly impossible for red states to have grown more than blue states for 40 years yet still be poorer today, which we know to be true
On 1 that’s why you should read the piece. I use the metric historically for each time period and see what happens with growth and population.
Your objections seem to answer one another, no?
If which states are economically more free is a moving standard, then it can both be true that economically freer states always grow faster and that economically free states today are poorer than economically unfree ones.
2. Why would that be impossible? Imagine a federation where red states start from a lower base than blue states, grow on average faster, and then red states flip blue once they are relatively rich and blue states flip red once they are relatively poor.
Re: 15: I am skeptical that if the science said "white people ignore black doctors and only listen to whtie doctors," the suggested policy resolution would be ensuring they have access to white doctors they will listen to even if it requires medical schools and hospitals to practice racial discrimination in admissions/hiring. Why is the fix affirmative action, instead of teaching black patients to be less racist toward their doctors?
>Why is the fix affirmative action, instead of teaching black patients to be less racist toward their doctors?
Because the former is likely to be more effective? One can certainly argue that affirmative action is nevertheless wrong, but let's not play let's pretend.
Okay, so: yes, white patients who are extremely racist should be entitled to care by white doctors if it makes line go up. Pandering to the irrational preferences of racists is good. Duly noted.
Did you REALLY miss the part where I said, " One can certainly argue that affirmative action is nevertheless wrong"??????????
Your second sentence makes exactly the point Blake was making. (Blake’s version was more allusive, for sure, but I for one thought the intent was clear.) Your first sentence was guilty of “let’s pretend” in the other direction.
No, it did not make remotely the same point. For one thing, I didn't say that my personal view was that it would be wrong or illegal.
Edit: To explicate further, OP implicitly claimed that there is an easy solution, ie one that is effective and legal and moral. I was saying that that is not the case.
And my first sentence was not at all let's pretend. Suppose I run a public hospital in an area with many very conservative Muslims. I hire a male as the only Ob-Gyn. As a result, many local women refuse to get services. What would be the most effective way to address that problem? 1) Convince the women to change their religious views; or 2) hire another, female doctor? Obviously #2. OP's suggestion to choose #1 is magical thinking. #2 would be more effective, but it might be neither legal or moral. Or it might be both legal and moral.
> What would be the most effective way to address that problem?
I think the disagreement is whether this is a problem that needs to be solved in the first place. If they're the ones refusing to be treated, then what's the issue?
I think you are jumping from pointing out that we would react differently. If it was white people with this tendency to assuming that we would be correct to react differently, and it’s just obvious that getting universities et cetera, to practice affirmative action is a lot easier than getting blacks to change their attitudes, especially since our record of deliberate attempts to change the culture being successful instead of being a total waste of money is pretty bad, especially in the short-term and it seems irresponsible to let serious medical problems happen while we try to figure this out. Also isn’t the whole ID of meritocracy that you picked the person who the best for the job? If being black makes you legitimately better at the job because people will trust you more than why shouldn’t we pick people on the basis of race, the same way that we are comfortable picking people on the basis of social intelligence or physical attractiveness or strength in jobs that require that skill, even though these are innate characteristics. You can maybe argue that permitting affirmative action in these situations would also permit affirmative action in other situations. Where do you think it would be bad enough to more than cancel out the benefits, but that’s a different argument which acknowledges that affirmative action in this situation would be good in isolation even if the false negatives of permitting actually harmful affirmative action would make a rule permitting It net bad. Certainly, it seems obvious that if you were a colourblind alien, who didn’t care at all about race, but was just trying to maximise welfare, picking the more trustworthy doctors in terms of public perception among certain communities would be an obvious move.
Just to be crystal clear: yes, we should pander to the irrational preferences of racists. This is the position of the left. Yes?
For the love of God, stop thinking what is right or left, and actually try to figure out what is the correct position without shouting about tribal fights and muddeling your thinking. If it was in fact the case that not pandering to racists was causing severe problems with medical treatment, then obviously you should infect pandering to them and there are lots of situations where it’s obviously the rational move to pander to them. For example, if you are in a society where everyone is racist and they find people not pandering to their views about as uncomfortable as we find people blaring super loud music that disturbs their neighbours at night, then, obviously, you should pander to them that just obviously follows from any reasonable moral principals. Just as the fact, Christianity is nonsense, doesn’t mean that you should not pander to Christian’s preferences when they genuinely care a lot about an issue because of their irrational beliefs, even when they include hostility to groups. Similarly, just because somebody is a racist doesn’t mean accommodating their preferences. Isn’t a good idea if it has a significant effect on medical outcomes. And please remember that demonstrating that the left is inconsistent doesn’t mean that you or the right wing is right. You have to actually figure out what is the correct position. Not just show that your enemies are inconsistent. Please use more scout mindset and less soldier mindset.
"Just as the fact, Christianity is nonsense, doesn’t mean that you should not pander to Christian’s preferences when they genuinely care a lot about an issue because of their irrational beliefs, even when they include hostility to groups."
This is literally illegal in the US. As is racial discrimination (aka "affirmative action"), despite the courts inventing nonsensical reasons why it doesn't blatantly contravene the very clear terms of the constitution and of civil rights legislation (e.g., CRA 1964).
The reason I brought it up is because affirmative action is an extremely clear example of a political issue that splits very evenly along the left/right divide. Leftists think that racial discrimination is good, as long as it advantages their pet ethnic groups. When they're asked to extend their reasoning to other groups, they tend to get very cagey and defensive.
"You have to actually figure out what is the correct position. Not just show that your enemies are inconsistent."
Racial discrimination is wrong. It is doubly wrong when the law says very clearly that it is wrong. It is triply wrong when it is permitted against one group and not another group, just like any legal double standard is wrong per se, in and of itself.
That is the "correct position".
I think you’re making the classic mistake of assuming everybody else thinks of the issue like you do which makes their positions in comprehensible. I just don’t think there’s anything wrong with inherently discriminating on the basis of race more than there’s anything wrong and discriminating on the basis of for example, your eye colour or height. The reason it’s illegal to discriminate on the basis of race is because there is a widespread pattern in recent history of unreasonable discrimination on that basis. So we have decided that a bright line rule against all discrimination is worth preventing, even the reasonable instances of discrimination. This decision may or may not be correct, but it’s pretty obvious why someone who thinks that racial discrimination when it serves useful purposes is fine Would be okay with affirmative action, especially since such a person probably doesn’t think there has been any widespread pattern of unreasonable discrimination against whites, whereas there has been such a pattern against blacks and frankly your confidence that you know better than the Supreme Court makes no sense. The civil rights act was signed into Law by a president who was enacting affirmative action at the same time, so I doubt very much. He thought of it as prohibiting affirmative action. There have been lots of arguments over whether those who ratified the 14th amendment intended to ban affirmative action in any case, if you look at the court decisions during that time, it’s not even clear that there was meant to be a bright line against the usage of race as relevant. Supreme Court decisions handed down in the early days of the 14th amendment only say that they suspect that race is not a reasonable basis for discrimination, but it is only a doubt and they don’t appear to have the 2 level equal protection clause that we have invented and instead of appear to be treating everything as a rational basis test, which means that the protection is actually pretty easy to circumvent. Just read the relevant historical debates. There are good arguments for both sides, and it’s telling that the 1880 Supreme Court thought only that they doubted that race could ever be a reasonable basis for discrimination and did not ever go far enough at that time to declare that race could never be a basis for discrimination. They just appear to have been operating under the mistaken understanding that you can’t get useful statistical information out of someone’s race, although they at least had the humility to understand that they could not be confident of this. Personally, I think that just as it is considered fine for police to take relevant characteristics like a person’s age and gender into account when doing their jobs, it doesn’t make any sense to then magically say that you can’t get statistical information out of somebody’s race, and most conservative appear to agree with this, so it’s hardly only a left-wing position that racial discrimination is actually fine when it serves a useful purpose instead of being based on a mistaken understanding of the world or dislike of a group. I admit that personaly it has never made sense to me how people think of race as this magic thing that has moral significance when it comes to discrimination. Whereas everybody will admit that discrimination on the basis of other inate characteristics is fine. what makes race so magical most justification is that make any sense we have to refer to society in history because it’s just opus to me that raise inherently can’t have any moral significance above things like eye colour or where you end up in terms of neurodiversity.
"The reason it’s illegal to discriminate on the basis of race is because there is a widespread pattern in recent history of unreasonable discrimination on that basis."
Telling an industrious, idealistic young man just starting his career: "we would hire you for this job, but your skin is the wrong color" is not a recipe for racial harmony, it is a recipe for sectarian strife and civil war. So yes, racial discrimination is wrong. And as I said, it is doubly wrong when everyone has agreed, through the democratic process, to abstain from racial discrimination, and then these laws are only enforced against some ethnic groups and not others -- and the state itself engages in this same form of selective discrimination!
But you've said you disagree! You think racial discrimination is not immoral. Telling a black doctor that he's not getting hired by your hospital because the patients prefer white doctors is not immoral.
> Leftists think that racial discrimination is good, as long as it advantages their pet ethnic groups.
I consider myself a leftist. I always thought affirmative action is wrong, except in some very limited situations which encompass disadvantaged whites as well.
I also think affirmative action is a bad idea right now. And for what it's worth, it lost at the ballot in California in 2020.
I think you’re potentially right about the counterfactual, but in this case it seems like we can walk and chew gum at the same time, begrudingly hand-in-hand with people who would be hypocrites in your alternate reality.
I also think you can take heart in the fact that you now have ammunition against anyone saying white doctors are cruel or malevolent towards Black people. Those arguments spread misinformation that directly harms Black people.
Ergo, by Kendi logic, such statements are racist.
The solution is to not worry about this at all. Black people are just suffering the consequences of their own racism. Good, let them. There's no reason for anyone else to care.
Off your meds, I see.
Actually, was trying to reply to a different person, not sure why it ended up here must have messed it up on the mobile.
The fix is repealing anti discrimination law and ending the legally enforced AMA cartel that limits the supply of doctors.
This is a pretty uncharitable interpretation of what I was saying. In your example, unlike the situation with black doctors treating black patients, the discrimination does not prevent actual medical problems which can sometimes lead to outright death. I think given how strongly people prefer for there be no racial discrimination. It would definitely pass a cost benefit analysis to prohibit discrimination when it’s only a matter of the preferences of a few patients and they won’t actually end up suffering medical complications if their preferences are not acknowledged. However, at the point where you are actually causing people to die through such a policy, I don’t think it passes such a cost benefit analysis.
My point was that you appeared to be treating no racial discrimination as a deontological rule, this is not how anyone actually treats the issue, regardless of what they say, as should be clear from how liberal are okay with affirmative action while conservatives are okay with statistical discrimination by the police. In actuality, most people think of the prohibition on racial discrimination as something you can break for a good cause and certainly don’t think of situations where it causes loss of life as just fine because no racism is the end of the debate.
You might think this contradict,s what I said about race not being magic, but it actually doesn’t if people were as disgusted and offended. By for example, telling a young and hard-working air hostess that she is not physically attractive enough to be worth recruiting. I would also be against discrimination on the basis of physical attractiveness, but since they do not care that much about the issue, I don’t mind that discrimination. Of course, this is not an inherent difference between race and physical attractiveness. Since there have been societies in history, where if anything not discriminating on the basis of race, would be what everybody would be disgusted by. However, in such societies because of how people think about in groups and out groups, you don’t just have people expressing their preferences, but instead, racial cartels and back scratching clubs that have a huge cost to economic efficiency. This might sound closer to your position, but it really is not because the prohibition on racial discrimination is purely instrumental and if discriminating on the basis of race would produce a lot of economic value. In a given situation, I would be in favour of that. Of course, as I mentioned in my previous comment, it might be that this economic value is not worth the cost in terms of situations where racial discrimination is not reasonable in the sense of not being justified by a cost benefit analysis, but you don’t think of it like that at least as far as I can tell you’re not saying that no racial discrimination in this situation is a necessary evil so that we have to let people risk dying for the sake of a rule that has benefits on net. You just don’t seem to be treating it as a bad thing or a necessary evil in the first place.
Such a categorical prohibition on racial discrimination makes no sense to me. Is it really the case for example that the Belgian government should have let their country fall apart instead of agreeing to discriminate on the basis of whether you are French or Dutch speakers for government positions. Obviously, you can argue that race is not the same as language, but that just brings the issue of why race is different. I distinguished in practice because of the way people feel about it. Discriminating on the basis of race in practice would be a bit like burning the American flag in terms of how offensive it is, and without the free speech argument, but that is a contingent difference. The rule isn’t really that you should not discriminate on the basis of race. It’s that you should not massively hurt peoples feelings and pointlessly offend them without sufficient benefits to justify it. It’s just obvious to me that racial discrimination is not a terminal value. The way that for example, not re deceiving people or preventing suffering is a terminal value that is good for its own sake. You point to the benefit of racial harmony, but firstly I’m really doubtful that the impact on racial harmony will be large enough to outweigh people literally dying of medical complications and secondly no racial discrimination would as people like Nathan Cofnas point out, given the gaps in test scores between the races. No affirmative action would mean that they would be almost no black people at Harvard and other elite universities, which to me obviously sounds a lot worse for racial harmony, then the present situation and something that even the public in general would not tolerate.
You have done nothing to address the point that LBJ was already doing affirmative action when he signed the civil rights act into law, so it’s highly unlikely he intended it to forbid affirmative action and indeed, Congress could have amended the act in response to the Supreme Court decisions and yet did not. In fact, the Supreme Court when it decided was decisions was actually more conservative. Then it was when the civil rights act was enacted. I don’t think it is credible that Congress did not realise that there was absolutely no way that without specific instruction to the contrary, the warrant court was going to interpret the civil rights act as forbidding affirmative action, so I think it’s obvious that you can’t just act as if Congress on unambiguously supported banning affirmative action. Not to mention that the Supreme Court itself is democratically legitimate given that it is appointed by the democratically elected president and its members are confirmed by the democratically elected Senate. If there is any ambiguity, it gets resolved by the Supreme Court now, obviously the Supreme Court has changed their mind and that’s legally legitimate. But firstly there are a few loop holes in the judgement. For example, permitting universities to pay special attention to their opinion of whether a candidate has displayed desirable qualities, fighting racism, such that their struggles are evidence that they would make a good candidate. And in any case, there are other avenues like the obvious one is trying to persuade the court to change its mind, and even if the optimal policy isn’t constitutionally permissible, that doesn’t mean, we can’t talk about it and argue that it will be the best policy. Something being illegal is a reason you should not do it definitely, but it doesn’t mean you have to pretend that it won’t be the best policy if it became Legal.
Also, your kind of misleading about the public, the actual public responses to most service indicate that the public simultaneously wants no discrimination on the basis of race and a level of equality in outcomes that is not possible without such discrimination and will often say yes to questions that in practice amount to doing affirmative action in ways that require five minutes of thought to recognise them as affirmative action instead of stating it upfront. This is not really surprising after all, the public was hugely in support of the nominations of Justice Marshal and Justice Thomas to the Supreme Court. Even though they were both affirmative action appointments because the public actual preferences to do affirmative action while being sufficiently non-transparent that the public and fool themselves into thinking there is no affirmative action going on. Indeed, I think both justices would have had a harder time getting confirmed if they weren’t black given that they were major concerns about the Supreme Court being to liberal when justice Marshal was nominated, and I doubt just a Thomas would have had a huge public support, he did if he were not black, especially given the common wisdom was that no one has extreme as him could get confirmed if they were not black and he had only a few months experience on the appeals court, which is normally not considered sufficient to even be considered for nomination to the Supreme Court. That’s not limited to Thomas either I don’t see justice Marshall being nominated. If he was not black, he didn’t have some huge reputation or anything, and indeed was specifically given positions that would make him qualified to be nominated to the Supreme Court on account of his race, so it’s clear that even five minutes of thinking would make it obvious. They were both affirmative action appointments.
We should give white racist patients a doctor they trust when it's practical. If grandma doesn't trust the black doctor and won't listen to them, let her see a different doctor. It's not worth her dying over.
This doesn't mean we stop telling people not to be racist. We will keep educating people. But you don't leave someone to die because you want to teach them a lesson.
The article says it's not advocating for affirmative action; only that this is the strongest argument they've heard for it.
I'm not advocating for anti-bias training. I'm just saying we should let white racist patients have a doctor they trust when one is available.
How does that relate to anything?
Serendipity! Anti-Chef has a recent video on microwave cooking:
https://www.youtube.com/watch?v=GGeLq3WnRC8
59.
AI alignment researcher "Max Harms"
At least it's not "Max Paperclip"
His wife's name is Haven. I guess opposites attract.
38: The graph is missing years in the middle as well. All over the place, really. After 1989, it starts leaping ahead anywhere between 2 and 6 years at a time. I get the sense that maybe they didn't poll for this every year, but that doesn't explain plotting the graph that way.
And Jon Kozan brings up a good point: 2022 is likely to be an outlier for Covid-related reasons.
Yeah my preference for graphs like this is to space the x axis in proportion to its values, not equally! Let us see the gaps!
Seconded!
<mildSnark>
The way the graph is drawn now is as if there was noise in the x-axis (at least as visually perceived).
</mildSnark>
Just because NGO workers are cheaper than unionized city workers doesn't make outsourcing non-progressive. Because you can hire more NGO workers for the same cost, and they're all so personally progressive, outsourcing becomes in fact a very politically progressive thing to do. Outsourcing municipal services to nonprofits, over the long term, both weakened unions and helped elect Mamdani: https://thebiggerapple.manhattan.institute/p/mamdanis-ngo-model-army
Outsourcing to contractors (NGOs, religious charities, for-profit companies) lets you dodge all the civil service rules. That means easier to fire, wider range of pay (civil service pay has a lower ceiling and higher floor than private sector pay), fewer benefits, but also rules restricting political activity and trying to get rid of the old spoils systems.
>Hopefully you can predict my response, which is that all concepts become unmanageable when you zoom in too far and prudence tells us how far we should zoom at any given moment.
When we're talking about cooperative intellectual discussions, yes.
When we're taking about crime statistics? Who police and prosecutors choose to charge, and how they charge them, is probably influenced by many things other than intellectual prudence about how far to zoom in on a concept.
Put it this way: if the same event could lead to zero, one, two, ten, or a hundred thousand criminal charges depending solely on the preferences and habits of the police and prosecutors involved, that suggests that you *really* need to come to grips with how much variance and what patterns exist in that determination, before the crime statistics can be very useful for making any arguments about the underlying reality.
I think I couldn't have said it better myself.
I get scott's point, which is true and valid, but Scott's point invalidates the main thing I'm pushing back on, which is there is some objective measure of "real actual crime, of the sort people care about" beyond "what gets reported, or charged, or convicted."
As a legal realist, I think it's best to view crime as a thing that GETS CHARGED, but the moment you step outside the framework of what gets charged into a framework of how much crime there REALLY is, things get uncomfortable.
My article proposes several "zoom modes", all of which I can relate to, but none of them work. To reiterate:
"By crimes I mean crimes with victims" (nearly all my examples involve victim crimes, like theft)
"By crimes I mean crimes which people actually care about" (all of SBF's myriad frauds were things people actually care about)
"By crimes I mean crimes which a jury would convict on" (I truly believe the prosecution in SBF's case could have chosen from perhaps a million possible acts of wire fraud and the jury would have convicted whichever they chose)
I didn't even discuss some of the weirder ways to chop up these numbers. Most of us are familiar with the Mangione case: he's charged at the state level and at the federal level. Most federal crimes are also state crimes, and sometimes vice-versa.
If we're going by what is a crime, in the objective sense, beyond what gets reported, charged, or convicted, I think the only metric that makes any kind of sense is "does the act meet the elements of a statute defining the crime" and by that metric there are gozillions of crimes, and most "discrete crimes" (say, a murder) are actually thousands of uncharged crimes (first he lunged at the guy, that's assault, then he took out a knife, that's assault, then he drove the knife forward toward the guy, that's assault, then he stabbed the guy, that's assault, then he stabbed the guy again, that's assault, then he stabbed the guy again, that's assault, then he waved the knife around wildly, that's assault, then the guy died, that's manslaughter, murder in the 3rd, murder in the 2nd, murder in the 1st, then he threw the body in a ditch...)
I also wonder about things like felony murder or laws related to gangs and conspiracy... I don't know this area of the law well enough to have firm opinion, but my understanding is that it can be possible, for instance, to charge many people with murder for a single death. Which makes me wonder about using murder as the 'reliable' statistic.
The statistics are "reliable" in the sense that they accurately record the number of dead people, which is what most people care about.
Like I said in my article, I think if you trust various parts of the Criminal Justice System to have an accurate bead on the public's sensibilities with regard to crime*, "reported crimes" "charged crimes" and "convictions" are going to meaningfully measure "crime", because a legal realist can say "crime is that which is reported as a crime and/or charged as a crime and/or results in criminal conviction"
If you DON'T trust the Criminal Justice System in either direction: you think it's charging vastly more things that "aren't *really* crimes" or you think a large number of "things that are *really* crimes aren't getting charged" then you enter this hellzone of "what is crime, actually" and in that zone lies the calculus of my post.
* (I do, very few other people do)
It is true that crime statistics are deeply imperfect, and that part of that imperfection is prosecutorial and police discretion. It is also true that this doesn't actually let you just wave away reality wrt things like crime statistics by race, where it is absolutely clear that (for example) blacks commit many crimes at much higher rates than whites (which hispanics in-between and asians even less inclined to crime). The imperfection in these statistics, and the fact that crime itself is defined socially and not every society even considers the same things crimes, also doesn't prevent us from noticing that there have been some years when a lot more dead bodies with bulletholes in them turned up than in other years. Or that in some years, a lot more cars got stolen or a lot more stuff got stolen from store shelves than in other years.
Well yeah but in a lot of those specific examples you're citing the baseline reality is the exact same crime statistics reactionaries typically push back on.
You will absolutely not get a negative comment from me by observing "homicides went up in 2021" but you will if you say "overall, crime is going up, despite what the statistics would have you believe." That's where I'm going to start saying "okay but what do you mean by crime?"
RE 34: Is this a credible source? I just find this hard to believe. It's just so easy these days. You really could just ask AI to "find me which of these projects are pseudoscientific woke bullshit written by virtue signalling left wing activists pretending to be scientists" and it would actually do a pretty good job and not make idiotic word context mistakes like that.
Yes it is absolutely true: https://www.science.org/content/article/trump-orders-cause-chaos-science-agencies
Quote: "NSF officials declined to comment on the intricate, multistep process, which began last week. But people with direct knowledge say it started with NSF senior managers selecting 10,000 grants, from a pool of roughly 50,000 active awards, for review. They then enlisted staff to vet the awards using a list of key words that included “diversity,” “inclusion,” “women,” and “race.” To be sure, a sizable share of those words refers to scientific designations, such as plant diversity, that have nothing to do with DEIA; those awards were dropped from the review."
If you have time, you may want to go through all the references in this Wiki article: https://en.wikipedia.org/wiki/Science_policy_of_the_second_Trump_administration
Pity the poor statistician trying to remove the word "bias" from his research proposal.
Are you misreading that quote? It says that they filtered based off a list of keywords but then manually reviewed that list and *stopped* investigating grants where the keywords were off-target hits. The following paragraph in your source makes it even more clear. They may have still mistargetted some studies, but your quote doesn't support that. Your source also claims that any identified studies were just going to have to be "modified" and were expected to be fully funded.
I don't use twitter so maybe I'm missing it, but the linked tweet from Scott's post only has one example and I'm not confident that they know why it was cancelled. Or that we know what the supposed list of naughty words is to even know that poralization is on it. Can anyone highlight the most conclusive example of this? I'm a bit skeptical of this narrative. The "transgenic" claim that went viral was as far as I could tell entirely speculative and there is no reason to believe that they confused transgenic for transgender. Certainly the specific examples that the White House ended up giving did not make that mistake (though they were also totally valid lines of research that were tangentially related to transgender people just as part of their broader impact - cancelling them was dumb but it wasn't CTRL-F "trans" level of dumb).
Perhaps I did misread it. In any case, there is a lot more in the Wiki article I linked; the science dot org article is but one of the references in it.
Other sources from the Wikipedia article:
NY Times: "Additionally, the presence of some terms was used to automatically flag for review some grant proposals and contracts that could conflict with Mr. Trump’s executive orders."
Duke's The Chronicle:
"The 15 canceled NIH grants contained flagged words like “expression,” “disability,” “trans” and “systemic.” However, these words were not used in contexts to promote DEI initiatives, although some may have had diversity supplements including the two described below.
The grant that had the word “disability” in its abstract focused on identifying biomechanical and biological factors that lead to early onset osteoarthritis following a partial meniscectomy — a surgical procedure commonly done to remove damaged meniscal tissue following a tear. The grant had over $750,000 remaining in funding at the time of termination.
Furthermore, a grant that had both “trans” and “systemic” in its abstract was working toward editing neural circuits by engineering channel proteins to control the point of contact between two neurons. It had around $1.4 million remaining in funding at the time of termination.
In most other cases where “trans” was flagged, the term generally referred to other words with the trans- prefix, such as disease transmission, transgenic genetic material, translational studies or signal transduction."
So the article states that both had diversity supplements which I think are more likely the cause. FYI, "flagged" refers to a third-party called Grant Watch, not to any direct knowledge as to what was done by the Trump admin and the word list originates from the NY Times article above. My personal assumption is "ctrl+F and then review" which fits this all better than just "ctrl+F and delete" but isn't as meme-able.
To be clear, I'm very against funding cuts to science, I work in an NIH-funded lab, I strongly oppose the administration using grant cancellations to target its enemies or pressure universities into obedience. I just don't see evidence that they did the most absurd level of mistake like confusing transgenic with transgender or dropping all research into polarization of light due to a keyword search.
> “They enlisted staff to review 10,000 grants”
I’m pretty sure they made a bunch of dumb polarisation-of-light errors
The graphs on 22 showing international markets providing 43% returns to 14% US returns are...I would appreciate further context here.
VTSAX (Vanguards total US stock market index fund) is up 18.18% over the past year and 10.76% over the past 5 years. (1)
VEU (Vanguard's global stock index fund excluding the US) is up 27.68% this year and 7.73% over the past 5 years. (2)
For reference, VXUS, Vanguards global stock index fund including US stocks is basically identical to VEU. (3)
I think it's worth noting that these numbers differ significantly from those presented by Catherine Rampell and global performance has dramatically underperformed the US for the past couple years. Trump could be screwing everything up but the US has been going up 20%-ish a year while the global funds have been returning like 9% up to this year. That feels a lot more like a market correction than a Trump screw up. USA #1, yeah, but probably not literally twice the growth in other markets for decade.
But, again, not a finance guy, just a retail investor, but there's no Vanguard fund I know to buy that shows those returns.
(1) https://investor.vanguard.com/investment-products/mutual-funds/profile/vtsax#performance-fees.
(2) https://investor.vanguard.com/investment-products/etfs/profile/veu#overview
(3) https://investor.vanguard.com/investment-products/etfs/profile/vxus#overview
I have been puzzled by this as well. It certainly seems plausible to have a capital domino effect, where just a few movers and shakers greasing the international markets leads to a cascade effect (guilty as charged, recently sold my stake in VTI to buy more VT, also why doesn't the I stand for International, le sign)...but is it really due to a US-specific drag, an ex-US mispricing, something else? Lots of stories one could fit to the data, with differing investor implications.
Not seeing it universalize across funds either. My workplace's retirement accounts aren't held with Vanguard, and over there...the US-centric funds are still on their usual tear, while recently international ones are *also* doing better than their longer-run average. So it seems much less like a direct substitution effect? The recent unpleasantness in the Middle Earth did tank US stocks more bigly than internationals, which has to be part of the very-recent story. But after recovering from that shock...shrug, I'm not sure what's going on either, or where to put more money going forward.
This graph is generated before the end of February (i.e. the start of the war). I hold Vanguard Total International Stock Index Fund (VTSNX); it appears to be plausibly up by the amount reported before the war correction.
> 60: Rob Ennals has created a site for viewing ACX book review contest entries (subreddit thread and discussion here).
Have you forgotten Jenn C's existing site for this purpose? https://codexcc.neocities.org/
I guess this new one changes the format so you're not reading a Google Doc, which is nice, but the earlier one is still worth acknowledging!
> It also (imho correctly) calls out modern “postliberals” as pre-Rawlsian rather than post-Rawlsian, in the sense of not confronting or having a good answer to Rawls.
CTRL+F for "postl" or "post-l" yielded no results. He talks about illiberalism generally, rather than claims to be post anything. Of course, if a self-described postliberal (like Ed Feser, who distinguishes himself from the more unprincipled of the postliberals) claims that they are actually just harkening back to Aristotle, Augustine, Aquinas and so on I don't think they would see it as much of a critique to call them "pre-Rawlsian". Even the people he specifically calls out would scoff at the idea they need to persuade "all" people first, our existing pluralist society already enables people to enforce their values even on the unwilling by obtaining political power. Nor do I think it's a serious indictment of someone that they don't "confront" Rawls or Kant once you acknowledge their basic moral principle is incorrect, any more than I think it's a problem that Tolkien didn't grapple with post-modernism (something he's actually dinged for at https://starsbeetlesandfools.blogspot.com/2015/06/worldbuilding-bakker-vs-harrison.html?showComment=1458955283269#c4652106653130330971 ) People who are into that may find it important, but everyone else who thinks it's bunk can ignore it once people like Huemer point out that it's bunk.
Only paid subscribers can comment on Heath's own post, and if I could I'd point out that there's a sleight of hand to go from "liberalism practically died out in the early 20th century" to the very next sentence talking about Rawls' "strategy for reviving liberalism" seeing as how his book was published in 1971, years after the Great Society had established its version of liberalism as a less extreme version of what Rawls wanted. It's not like he was trying to revive classical liberalism, which was what had "practically died out".
Re:#2: Japan has a similar rule to Uruguay, as per the law. AFAIK, and according to Wikipedia, nobody has ever been punished for desecration of either the Hinomaru or foreign flags. However, one of Takaichi's campaign promises is to make Hinomaru desecration a crime.
I concur, Japan also allows national but not foreign flag desecration.
https://laws.e-gov.go.jp/law/140AC0000000045#Mp-Pa_2-Ch_4
Adjacent to 23, has anyone else noticed things like YouTube and news aggregators censoring non-profane words like "sex" and "killed"? I read a Yahoo News article recently that censored every occurrence of "sex" IN AN ARTICLE ABOUT SEX. What's driving this? Has the internet become so mainstream that every media platform is adopting 80's-era TV policies? I have a hard time believing that the presence of words like "sex" hurts viewership or ad clicks in any way, so what's going on? Are interest groups policing this or something?
I do not know *why*, but it definitely does happen. On www.quora.com many posts use a euphemism "unalive" because the verb "kill" is liable to push your post to the bottom of any search.
Shouldn't you provide some support if you're going to say "it definitely does happen"?
Behavior on Quora may tell you that the people there (Who? That site died years ago...) believe it happens, but doesn't tell you anything about whether it happens.
OK, I can not PROVE that Quora or YouTube algorithms penalize posts which use the words "kill" or "porn", and short of the website owners actually admitting it, proving it would be a long and involved project. Still, enough people believe it, to use words "unalive" and "pr0n" in order to fool these algorithms.
In this case, I think it's worth making it very clear that the odds that Quora is doing this are indistinguishable from zero, and the user behavior is known with certainty to be a contagion from user behavior on other platforms.
> words "unalive" and "pr0n"
"pr0n" is leetspeak. It's only meant to be a cultural marker.
https://wiki.kingdomofloathing.com/index.php?title=XXX_pr0n&oldid=433561
For what it's worth, if a Quora post contains certain words, any images will be blurred (clicking on the image unblurs it). That is consistent and demonstrable. Here is an example:
https://alicecsspace.quora.com/
The topmost post is about suicide; the image is blurred even though it is nothing but a graph.
Tiktok, Instagram and pre-takeover Twitter had aggressive censorship for certain words liable to appear in controversial posts. I'm not sure if other sites are adopting the same policies to keep their own posts shareable on social media or if certain users are adapting their language the same way on every site out of an abundance of caution.
The whole phenomenon is discouraging, but some of the workarounds are amusing. Keep an eye out for tiktok-speak like s3x, @bort, 🔫-violence, and yes the infamous "unalive"
A decade or more of major social media platforms like Youtube and TikTok trying to maximize kid-friendliness. They want kids watching to maximize watch time and how many ads are shown to the most eyeballs - that means optimizing for the kind of content parents won't object to.
We are now beginning to see the social effects of this as a whole mass of people and cultures have normalized the idea of not using the common word for a serious issue.
Optimizing for the kind of content _advertisers_ won't object to. Corporations don't give a damn about kids or their parents, but they care very much about ad money, since pretty much all of the internet still runs on it.
When you don't know the exact rules that will make your video demonetized or not recommended by the search algorithm, only you know that some rules exist, people try to play it safe and copy each other's superstitions.
It's hard to know what the decision makers are thinking. Maybe it's a desire to please the wokes who demand censorship of "hate" without pissing off conservatives too much. The solution is to censor/deboost/demonetize anything that's more than PG. The "bigot" doesn't get to share their views on sex nor does anyone else.
Yes, I have seen this. Channel 5 Action News had a recent video about a suicide which censored the word "suicide" and "kill"; RedLetterMedia have been censoring the first few minutes of their videos to avoid demonetization.
>only three emperors - Vespasian, Marcus Aurelius, and Septimius Severus - had sons who could inherit
Apart from Vespasian's elder son Titus, most of the sons in question were assassinated, and I don't think many of them have great historical reputations. That doesn't sound to me like a sign father-son inheritance was the missing ingredient.
The "Five Good Emperors" period seems to be generally considered the high point for the Principate in terms of both stability and quality of governance. Of the five, Nerva was elected by the Senate and the next four were nominated by their respective predecessors.
If you're a Roman emperor with no living sons, then you get to name your own heir - almost certainly an adult who you've been able to watch and assess for their potential in the role. That does seem more likely to lead to good outcomes than "first child with a penis to come out of my wife's vagina".
In the latter case, you at least get a bit of a genetic edge in that the father was demonstrably able to run an empire well enough to avoid major civil wars for a couple of decades, at least. But if natural succession is a rare exception, you're only getting one generation of natural selection on that axis. And the candidates for adoptive succession were at least being drawn from the families of capable aristocrats.
True I'm most familiar with the details of the succession during the Julio-Claudian Dynasty, which seems to fit that pattern. Historical reputation seems to run (in descending order):
Augustus (adopted heir of Julius Caesar, last warlord standing)
Claudius (installed by praetorian guard)
Tiberius (son-in-law and chosen heir of Augustus)
Nero (blood heir)
Caligula (blood heir)
-----
It also occurs to me that during the Principate, the Senate still mattered quite a bit as an institution despite the Emperor clearly being in the driver's seat. Emperors who knew how to play the game of Senatorial politics tended to do a lot better than ones who didn't. The fatal flaw of the bottom two Emperors on the list, both of whom inherited young without having opportunities to practice the game before becoming Emperor, was that they gave too many Senators too much reason to hate them.
>If you're a Roman emperor with no living sons, then you get to name your own heir - almost certainly an adult who you've been able to watch and assess for their potential in the role. That does seem more likely to lead to good outcomes than "first child with a penis to come out of my wife's vagina".
I've heard from a number of classical historians that in the Roman republic, continuing into the Roman empire (at least for much of its duration if not all,) heredity by adoption was considered more significant than heredity by blood, for more or less exactly that reason. If you were the person that an individual singled out as his preferred heir, that was more dispositive of you possessing the skill and character he valued than if you were simply his direct genetic offspring.
Also, adopting a successor while still healthy meant you were at least planning well in advance for succession. One of my takeaways from the Mike Duncan History of Rome podcast was that most Roman emperors simply didn't do this for reasons that remain hard for me to understand, and a good number of those who did chose badly (hot take: the extreme example of this is Diocletian's cockamamie Tetrarchy scheme which he should have known would fall to pieces). "Apres moi, le deluge" indeed.
the sagittal-shift of awareness mentioned in 51 is interesting. this week, somehow for the first time, I noticed my thoughts centering on different points along that axis in relation to how anxious (forward) or mellow (backwards) I felt. taking decades to sense this left me suspicious of it being a passing illusion, so it's nice to see some evidence to the contrary. powerful motivation to try daily meditation again too.
I noticed the same thing when I meditated once for an hour. My vision seemed to shift, back, like he said. Rang true.
Me too
re:72, Pangram thinks 80% of both papers are AI-generated.
@Scott, or anyone else with an opinion, where do you find such weird and interesting links? E.g. the loveland frog.
Any favorite sources or methods?
thebrowser dot com
They will send you 2 interesting links a day for free. If you give them 50 bucks a year they will send you more.
I’m using the free tier and it presents some great essay. I’m thinking of upgrading.
Today’s freebies from The Browser:
Shall We Play a Game?
Jon Peterson
Historian Jon Peterson traces the route from Prussian military headquarters to Gary Gygax’s basement.
https://asteriskmag.com/issues/14/shall-we-play-a-game?ref=thebrowser.com
The Half-Life of Metrics
https://seekingsignal.substack.com/p/the-half-life-of-metrics?ref=thebrowser.com
Ok you got me, signing up
My Firefox homepage is the Wikipedia random article link. Half the time it's an obscure species of moth, or the summary of the 1997 season of some minor local football club, but other times there are good ones (with the usual caveats that Wikipedia has become much less reliable a source in recent years on politically-charged topics--or rather that the list of politically-charged topics has expanded greatly; the page listing lamest edit wars extends to the early days, but mostly covered relatively unimportant topics).
The article on Nectome is great. I'm very interested in the idea of cryonics, but extremely disturbed by the practices of currently existing cryonics companies.
Trying to preserve a brain after it's been dead and warm for hours is completely pointless. The physiology of cell death and synaptic degradation is not fully understood, but it's understood well enough to know that not much is being preserved after 24 hours. And probably not much after 9 hours. There is slogan in stroke neurology - time is brain. So it's a huge credibility boost that Nectome will only preserve a brain if they can do it within 15 minutes of death. The fact that other cryonics companies are happy to take people's money to preserve a brain that has been dead and warm for over a day is damning.
Nectome's other innovation is also clever. Using cryoprotectant plus fixative sounds obvious in hindsight, but its a big advance. It makes preservation more accurate, and cheaper which is a big deal when your goal to is to keep preservation going for decades to centuries.
I find the following pretty difficult to parse. What exactly is a "zigger"? Do I even want to know?
"Our suppression paved the way for a lot of the ziggers and degeneratron grifters, who were not as all-pervasive as they are now. The major pages were generally pro-american of course, while the “degenerate right” (completely useless people who contrary their claims have generally terrible taste and no artistic talent btw) was constantly hounded with a fair measure of success. A lot of the stuff you see nowadays is because people have gotten worse in many respects, but much of it is because the purges destroyed the “internal hygiene”."
Insofar as I understand it, though, I disagree with it. It always seemed to me like the excessive censorship of right wingers before Trump's second term / Elon's twitter takeover was mostly a positive influence on the right, by artificially inflating the average content quality. Ever since then, we've had more success, but a lack of the restraint and standards we had before. At this point I think I regret not voting for Kamala.
it's an adapted slur - https://en.wiktionary.org/wiki/zigger
Someone who supports the policy or doctrine of the Kremlin and the Russian invasion of Ukraine (2022).
Examples in the right wing media (partially, not their whole ideology): Tim Pool, Dave Rubin and Benny Johnson
I see. Definitely an objectionable position, but I think a group with real "internal hygiene" would find a better term than that.
A group might have different standards for what is hygenic. Just because you both agree that swinging around Nazi flags is uncouth, doesn't mean you have the same attitudes towards what is considered acceptable vocabulary.
The goal, ultimately, is to have people like you no longer be in a position to police language.
> A group might have different standards for what is hygenic.
In the specific case shown in the Twitter thread, it doesn't seem there was much hygiene going on. Calling all Muslims terrorists (or something equivalent, I suspect) or changing one letter in a racial slur while still trying to maintain its...contemptuous connotation, is not exactly hygienic.
A recurring issue in discussions about not censoring the right is that in the current environment, the right seems to delight in saying things even non-partisan walled gardens wouldn't tolerate.
> The goal, ultimately, is to have people like you no longer be in a position to police language.
If the people want Parler, they are free to have it, but they should be honest about that and also how much hygiene they actually have.
> the right seems to delight in saying things even non-partisan walled gardens wouldn't tolerate
I don't agree that there is such a thing as a "non-partisan" walled garden. Everyone has their biases, and an ideology they want to push. Just because you consider it the default doesn't mean it's non-partisan. Can you not imagine a society where the things you describe are found to be perfectly acceptable and normal? You wouldn't even need to go that far back in history to find it.
> I don't agree that there is such a thing as a "non-partisan" walled garden.
I think the various clubs I joined in high school were non-partisan.
> Just because you consider it the default doesn't mean it's non-partisan.
Not insulting others or using language which enables that is non-partisan.
> Can you not imagine a society where the things you describe are found to be perfectly acceptable and normal?
I can, except that society is not practicing hygiene.
Which is bizarre because the term applies to a great deal of left-wingers as well.
On the US vs. Foreign stock market returns, that could also be some mean reversion as the US stock market risk premium is pretty low right now, and foreign risk premiums have compressed more in recent months than the US has. https://elmwealth.com/capital-market-assumptions/
I have a PhD, and am probably qualified to teach math, physics, CS and chemistry at the highschool level. But there's no way I'm going to do it. Not because the salary is low. But because you can't pay me enough to deal with all kinds of behavior issues, administration and all other things that highschool teacher has to deal with. If teaching and every other aspect of the job were divorced it would be a completely different ballgame
Don't forget parents. I'm lucky enough to have nabbed a decent gig with my PhD, but I would rather do menial physical labor like warehouse inventory shit than teaching high school if the PhD gig dries up.
Very much agree. I too could teach the subjects you listed (BS in mathematics, MS in computer science, minor in marine biology), but the rest of the bullshit is impossible for me for ANY amount of money.
My understanding (my mom was a classroom teacher and I've known many more) is that the skills of managing a classroom full of kids are just entirely distinct from knowledge about a subject. A normal classroom full of normal middle-school or high-school kids requires a certain level of skill in keeping order and keeping everyone on-task, and if you can't do that, it simply will not matter whether you know your material, because you won't be able to do much teaching.
Of course. Which is why the whole talk of "let those poor adjuncts teach highschool" is pointless. Unless we, as a society, decide that its worth setting up schools for kids who both don't need much management and could benefit from being taught by PhDs
Also, most kids don't need someone with deep expertise in high school. The lady who understands high school algebra but doesn't know what a Galois field is can do a fine job teaching high school algebra. It would be a pretty bad allocation of resources in our society if most high school math teachers were mathematicians.
My daughter is actually planning on becoming a HS math teacher, and has been looking at the degree requirements from various colleges.
Surprisingly (to me), many require a double major in education & math, and those that don't still have course requirements (e.g., real analysis) that I'd've thought would be completely unnecessary for the job.
I don't see why this is such a surprise. One would hope that a teacher would know more in the area than the maximum that the students are expected to reach. I also keep hearing horror stories about "degrees in math education" that may or may not teach the education part, but definitely don't teach the math one
Agreed on going being what you're teaching; I often advise coworkers to understand things at least one level deeper than you're going to be working at.
It's just that the requirements are even further beyond HS level than that; I have a math minor & never took real analysis.
I'd like to see the relationship between willingness to endure all that BS and left-wing views. Are progressives just more willing to sacrifice comfort for their egalitarian goals?
Egalitarian? I believe that the discipline situation is slightly better in private schools, but the rest of the drawbacks are pretty similar if not worse. Especially in the "elite" school with very rich parents
On link 3, I work in a UK local authority in social care. We have a small in-house service, which is expensive, inefficient, and very difficult to make any changes to. On the other hand, we have a large external home care market, which has upwards of 20 companies delivering home care at literally half of the effective cost per hour of the internal service, while being responsive, flexible and high performance. There have been some poor performers, but we just don’t commission any services until they get better.
We have one of the lowest levels of unmet need in the country but still get pelters about our use of the “profiteering” private sector, ignoring the excellent services provided, and constantly get demands to deliver more in-house services, despite how it would double our costs or half our capacity.
In general, recruitment is difficult because it’s even hard to get rid of fixed term workers now, due to layers of process and hoops to jump through. It’s far easier and lower risk to commission a service from an external provider for a fixed period than it is to hire someone to do something directly for you.
I think the black doctors finding is better explained as people are more likely to divulge embarrassing but consequential symptoms to someone they can superficially relate to (in group).
37. Genuinely not trolling: I'm not sure that adjunct professors and postdocs have higher status than high school teachers these days. I'd argue strongly in the other direction. Personally (+1, I just asked my wife too) I'm really not impressed by people who dick around in universities past their sell date - it's lame, and the reputation of universities is in the toilet. I AM impressed by someone who gets in the unions, sticks with a solid career without getting distracted, and does the incredibly difficult job of teaching my kids. Folks - particularly the fellas out there: both men and women want to marry that person. They don't want to marry desolation in tweed.
I think you probably meant just "status among other adjuncts who are hyper-concerned with the status of adjuncts", but that's like two guys per state at this point.
44. Come ON. "Kids intentionally fucking up an otherwise good essay so the teacher doesn't think that it was written by AI" are not sweeping the nation.
44. I have personally seen a current student spending hours each week doing exactly this.
All classwork must pass the Grammarly “AI detector” which results in said student going sentence by sentence and arbitrarily shuffling words around or adding casual adverbs like “a lot” until the opaque algorithm has decided that line is human enough.
I really hope that student is insane and paranoid, like full-on Richard Nixon. :) If what you're saying is correct, it means that the "AI detector" is worthless, but the school still use it, which is probably a metaphor for something or other.
Actually let's check in with the LLMs for a "funny, pithy metaphor":
From Gemini 2.5: "It's like teaching a fish to ride a bicycle...backwards, just to prove it's a fish"
From GPT 5.4: "That's like making your kids spend hours sanding down their fingerprints so the detective doesn't think they committed the crime of being human"
Alternate GPT 5.4: "It's like teaching your kid to write with a fake limp so the robot hall monitor won't accuse them of sprinting". I can't believe ChatGPT put this gem in second place!
Claude 4.6 Sonnet: "It's like having to limp so the speed camera doesn't think you're a car". LLMs like limping, I guess.
Sometimes I'm not so worried about the new robot overlords
What sort of status you have as an "Adjunct Professor" depends on whom you're talking to. To most normies it sounds exactly as impressive as Assistant Professor or Associate Professor; only people in the know understand you're basically a galley slave.
Being a "postdoc" sounds like being a glorified student, and how prestigious that sounds depends on how old you are.
Good point - the only reason I know is I have a couple of friends who gave up after a bit of "adjuncting" or whatever. Otherwise I wouldn't know.
There are two 28s, incidentally on polar opposite sides of the system 1 / system 2 thinking spectrum.
On Omelas:
Yes, exactly, what this guy says, which is also why it's trivially easy to dismiss the story as a *moral* piece of work. A document which says, "you must believe in impossibilities (utopias) to be moral" wrenches my soul exactly as much as saying "you must believe that 2 + 2 = cheese to be moral."
However, it's wonderful from a storytelling and writing and condensed meaning point of view. I take the thing as wonderful piece of art and not much more.
I don't even think the woman actually believes this argument. Compare to:
"Only in silence the word,
Only in dark the light,
Only in dying life:
Bright the hawk's flight
On the empty sky.
—The Creation of Éa”"
From Earthsea.
Omelas is not an argument. It's a sort of elegy in honor of noble madmen she *wants* to be right but knows aren't.
While the *speed* of Taalas is indeed impressive, its quality is very low. I asked it what price cinchona would fetch if somehow transported to ancient Greece and it responded "Cinchona (Chinchona officinalis) is a fascinating plant with a rich history. In ancient Greece, it was indeed a valuable commodity, but not for its medicinal properties [...] Instead, its value lay in its unique and highly prized timber. Cinchona was a highly sought-after timber for making bows, particularly the bows used in ancient Greek and Roman warfare. The tree's wood was prized for its strength, durability, and flexibility, making it ideal for crafting high-quality bows. A single bow made from cinchona wood could command a high price, reflecting the value placed on quality craftsmanship and martial prowess."
Cinchona is a New World plant, unknown to the ancient Greeks. Not only that, but this fact had been explicitly mentioned in the previous question. Even Deepseek did better than that when asked the same question.
I assume it's some very old open-source AI model, so I don't think this necessarily reflects poorly on the technology itself.
Re: latency: It gave me a response in ~0.5s (according to both browser dev tools and feel). The only reason that feels magically fast is that you're used to bloated websites written by lazy developers (or, rather, expensive ones, so companies are too stingy to pay for more developer time to optimize them). A basic, 90s style website, with a decent server and internet connection, should be *much* faster than this.
Latency matters most for multiplayer video games. In my experience, typically it's about 3x the minimum from speed of light constraints: from Europe, ~0.1-0.2 s to North America, ~0.4 s to Australia.
I once came across someone on Reddit who said he'd gotten an assignment whose spec required a ping time that would need a signal literally traveling faster than light (sub 50 ms for a server on another continent, IIRC). He had a tough time explaining to his boss why that wasn't achievable.
It said “Generated in 0.046s • 15,436 tok/s”. It’s hard for a human to measure that and I’m not sure if they are measuring the entire rendering or not. But that’s much faster than 0.5 seconds and I’m sure it wasn’t as slow, as half a second which is measurable to humans,
The speed really isn't just an illusion of low server latency. The Taalas HC1 chip outputs tokens at about ~17,000 tok/s, as compared to ~250 tok/s for a normal H200. The big downside of that 100x speed increase (and ~0.2% inference price) is that the LLM weights have to be physically printed on the circuit- so the chip can't be used for anything other than inference of that one specific model, and when the model becomes obsolete, you have to throw away the physical hardware.
Chips like that probably won't be used for inference at the big labs, but they'll likely have a lot of important niche uses for devices where speed and cost are more important than having up-to-date models. They might get used a lot in robotics and military drones.
> The speed really isn't just an illusion of low server latency.
I'm not saying it is. I was responding to Scott's impression that *in addition to* making the AI fast, they had to do some magic to the latency, otherwise that alone would make it slower than it is.
Indeed, the website simply uses HTTP streaming for delivering the AI response.
Came here to say this. The model generation is impressive and novel, but the internet latency is what you'd expect from a website that's not bloated by ads, tracking data, analytics, etc.
Plus the fact that a page load requires DNS resolution which requires a trip to a separate server (or two separate servers?), whereas loading more text on an already-loaded page doesn't need to do that.
(speedtest.net shows me getting 4ms ping to their nearest server.)
It's not even only (or even primarily) "nefarious" things like ads and tracking that make web sites bloated, but
- frameworks, abstraction layers, libraries that make the programmer's life more convenient, but inherently use more CPU and RAM
- ... or abstract away what's ultimately happening, making it harder to write efficient code;
- shoving in extra features and eye candy, without making sure that you only pay (in terms of speed) for what you use.
It's Llama 3.1 8B, probably quantized. That's about the quality level I'd expect from that model. (Bear in mind that useful models are usually 100-150B parameters, and not quantized!)
I just tried the stock, unquantized "instruct" variant of Llama 3.1 8B on OpenRouter and it *started* to try to actually answer the question before trailing off into nonsense like "We would have to consider also external market slice beyond known full consumption: other backgrounds species sources and oils contemporary gods can generate also proponents minescontrol mos fuatement oil refining travel propag constituent voy entirely tactieve winning Rub labour".
So... whether you consider that better or worse than the Taalas version is an open question. ;-)
NO! Do not send wannabe academics to teach highschool. I'd sooner send meth cooks, retired moms or local sex workers. If you **like** teaching and are good at it you can usually get a teaching job in a college but the reason you are enduring being an adjuct is the hope one day you will get to live the life of the mind with an occasional class.
Usually the features that send you into academia make you great at research -- you aren't that great with other people but you love the life of the mind or don't have the courage to do anything but the default -- but bad at teaching. And HS teaching is mostly authority. Imagine the grad students you know trying to face down a class mocking them or ignoring them.
And knowing so much about the field can make you positively problematic for a highschool. For instance, Math PhDs always try to actually teach the concepts and show the proofs.
Maybe it would be neat to have them at magnet schools where the kids mostly teach themselves but no, just no!
I spent one school year as an adjunct professor, and the only reason I did it is because I had no other options. I dropped it like a hot potato the moment I found a full-time database developer position.
I went to a decent school, let's say roughly half of PhDs went to decent academic gigs and the others wound up elsewhere (or with bad academic gigs). Which is to say, a lot of my peers are the type one might think of going to teach high school in this scenario. And absolutely, to your point, the emotional IQ of these people is just as room temperature as the stereotype of any other PhD.
Re 15: so segregation is a good thing now? I have no clue if the claim is true, but of it is, then it could apply to all other countless discrimination criteria human invented.
Re 21: based on my personal observations, the US is hard to enter legally, easy to stay illegally, the EU is the other around: easy to enter legally, hard to stay illegally.
Re:
“17: In experiments, current AI was not very good at helping a bioterrorism red team with the kinds of tasks involved in creating bioweapons.”
Having Claude read the paper, you can't actually rule out that it wasn't helpful. The AI got the participants further through 21/22 subtasks. It just didn’t help them actually complete the entire workflow. And this was summer 2025 LLMs, not current ones! Even current open models are 3-6 months ahead of where these ones were.
I don't know if I'm up for reading the whole paper myself right now, I might come back later, or perhaps someone could double check.
Re:60 I would like to claim with no evidence that friction is doing useful work here and raises the quality of the average vote.
Linked from the Microwave Cooking article, but seemed fun enough to link directly here - instructions on how to smelt metal using a microwave: https://www.instructables.com/microwave-smelter/ . (Also includes instructions on how to make a fireball with a microwave, but that's an optional step)
This guy thinks smelt means melt. It doesn't, and he's not doing any smelting. Smelting is the chemical process of turning ore into metal.
Regarding the Pitt Bull issue I got into this recently and there are a host of reasons to be skeptical. However let me start by saying that even if it does turn out they pose a modest extra risk I think the right place to start is just mandatory sterilization for the breed with very few exceptions (or even priority on the kill list at shelters above other healthy dogs).
People get very upset about losing their pets -- I'd rather have quite a few serious dog bites (even as a child) than be forced to give up my dog -- so from a total utility point I think you would need to prove that most pit bulls bite multiple children to justify a retroactively applicable ban or even rules that make it essentially impossible to find a place to live with a 'grandfathered' pittie.
But the evidence for danger also seems kinda weak.
1) Ths reported incidence of Pitt Bull bites is very suspect data. Turns out identifying dogs by eye is incredibly unreliable and what people say is a pit bull doesn't actually track the breed that closely. Further observe that the bite data seems to report an unusually high percentage of pure bred pittiea and it sounds like lots of the data is just someone going: I was bitten by that big scary dog and infering the breed from the fact it was a scary dog that bit.
https://www.sciencedirect.com/science/article/pii/S109002331500310X
2) We have confirmation that people with primary psychopathy are more likely to own ‘viscous’ dog breeds. In other words, when drug dealers want a scary aggressive dog they buy dogs people think are scary and aggressive.
https://onlinelibrary.wiley.com/doi/epdf/10.1111/j.1556-4029.2009.01001.x
Thus -- even if pitties are no more inclined to be aggressive than your average dog we *expect* then to be overrepresented in bites! Dogs trained to be aggressive will be.
But bans on pitties are like banning cammo colored stocks -- if the cause is the buyer you don't reduce the incidence but you probably do force them to transfer their pit bulls to mom (out of apartment/state) and get a Malinois so now more aggressive dogs and someone with a dog they can't necessarily handle.
3) There is something of a summary of studies on the issue here but without any real clear answer. The most shocking thing I learned is that basically none of the studies seems to control for breed popularity.
https://www.avma.org/sites/default/files/resources/dog_bite_risk_and_prevention_bgnd.pdf
---
Ultimately my conclusion is basically that the evidence for increased dangerousness of pit bulls is much weaker than it might initially appear. It is probably a genuine correlation but the size of the effect isn't known.
However, we have much stronger evidence and priors that 'bad' people who want a scary dog are more likely to buy Pitt bulls. Therefore we are left with data that can be explained in either way.
Now both my experience and my sense of what dog experts/trainers say about pitties/rottweilers etc *as raised from pups* as not being particularly dangerous I think the more likely explanation is that pitties aren't a very aggressive/dangerous dog breed. They are more aggressive than goldens/labs etc but probably less than Malinois's etc. I didn't look at how much worse their bites might be.
Given the other link (maybe I lost it) failing to find a super strong GWAS association traits and behavior genes it's probably not true that pitties have a controlling aggression gene with a huge effect. I just mention because some people want to infer dangerousness from how recently they were used in fights.
----
Cremieux addresses all of these points. See e.g. this article: https://www.cremieux.xyz/p/pit-bulls-part-i-identification.
Thanks for the link but not very convinced.
He needs to pick one definition in all cases and use it everywhere. He spends the whole piece being ambiguous about what class of dogs he is talking about. Yes, if you want to talk about dogs that look like Pitt bulls -- or even dogs which mostly descend from the set of breeds thst look Pitt bully -- that class won't be misidentified that much.
But the reason misidentification mattered is because we need a denominator and pretty often the only data we have for that are purebred registrations or the like. I might be convinced if he took his preferred definition of a Pitt bull genetically and showed me data about prevelance of **that definition** relative to bite prevelance.
But I don't claim there is any strong empirical evidence the claim is false. Fundamentally, as I said above we have reason to believe that real world bite rates would be higher regardless of inherent aggressiveness. Only way to get us beyond that point is eithet dog RCTs or cross cultural studies showing they bite more even in some place they aren't seen as aggressive. That data doesn't exist...frankly almost no data with solid base rates exist.
I don't have time to read all of them so maybe he has a secret amazing study but I doubt it.
>Now both my experience and my sense of what dog experts/trainers say about pitties/rottweilers etc *as raised from pups* as not being particularly dangerous
One additional problem that's hard to apply data to is the possibility that the various varieties of pit-associated dogs are not inherently aggressive, but they are more likely to "snap" at some point, like whatever the dog equivalent of going postal is.
Ie Several years of being a perfectly sweet and cuddly waddling blob followed by one instant of ripping a toddler's face off, with no warning.
That is certainly possible but gets even harder to test. I don't think we can rule out such a possibility but I just don't see strong data.
"The US is indeed doing well, but the rest of the world is doing even better! What’s going on? It looks like the answer isn’t so much that foreign economies are booming, as that a weaker dollar and US uncertainty are causing Americans to invest in foreign stocks, and foreign stocks started out weak enough that even a little extra American money can send them to the moon. So does this address the puzzle of why America is doing so well despite economic chaos? I’m not sure."
This is hilarious. I was wondering why my Europoor funds were kicking ass all of a sudden. I parked things there just because I didn't feel like riding a rollercoaster for 4 years. I expected nothing but boring stability and have been accidentally mopping up.
Esper's account of the court's jurisprudence is pretty misleading. It does not put homosexuality on one side and race on the other for the simple reason that both critics and proponents agree the court has shifted it's jurisprudence balancing anti-discrimination laws and free exercise laws in recent years. Indeed, one could even point to the fact that many of those anti-discrimination rulings about racist theology occured at a time when the court had much more reason to think that anti-discrimination served a compelling government interest.
I'm not a huge fan of the kinda arbitrary religious animous framing of masterpiece but the whole thing is far more complicated and nuanced than Esper -- TBF any tweet -- presents it. I strongly believe that it borders on immoral to discuss SCOTUS rulings without reading that decision or at least a broad selection previously. If that is too much you then you at least must read about it on scotusblog.
That thread is ridiculous and even with Scott's warning that he doesn't really review the links, it's absurd to include it. Completely insane and badly motivated reasoning.
Came to the comments to see if anyone else picked up on this. Just to take one tweet for example: "Why shouldn't a clothing store be allowed to refuse to sell formal wear to gay customers who might wear it to a same sex wedding that they see as an affront to God? Why can't a hotel owned by a believer refuse to rent a room to two men?"
The Masterpiece decision explicitly considered cases like these and said no, they cannot refuse. Quoting from the decision, which even cites the Piggy decision: "Nevertheless, while those religious and philosophical objections are protected, it is a general rule that such objections do not allow business owners and other actors in the economy and in society to deny protected persons equal access to goods and services under a neutral and generally applicable public accommodations law."
I didn't realize how much the slowness of most ai makes it feel analogous to talking to a human. Chatjimmy is insanely, jarringly fast
35. Henry Cho has an old standup bit about a guy named Jonly Bonly...
https://www.dailymotion.com/video/x6v8wfk
"So if the inexplicability of miracle stories overwhelms you, rather than retreating from atheism to religion, you should retreat from atheism to a belief in some sort of psychic/poltergeist phenomenon where certain people (often children) can manifest weird shared hallucinations and occasional small violations of natural law. "
Has this really happened in the rationalist community? Has someone become religious because they couldn't understand miracle stories? If true, I find that so inexplicable, that I might start questioning my lack of belief in the supernatural.
Re: 3, this is why public sector unions shouldn't be a thing. It is just and right for workers to unionize against the profit motive of private industry, but in the public sector they are public servants, and the people they are screwing over by unionizing isn't some businessman, but the taxpayer.
You seem to be making this weird implicit assumption that unions in private sectors aren't resulting in higher prices for consumers.
I think the implicit assumption is something like, if private sector unions push too hard on their employers, then other non-unionized employers will win in the marketplace. So there is a limit to what they can demand.
Whereas in the public sector, there is no limit, other than the size of the overall economy.
It is my understanding that in many industries, unions make deals with companies to prevent those companies from hiring any non-union employees.
Right, as another commenter here TGGP puts it, what is the ethical basis for public sector unions in a democracy? https://entitledtoanopinion.wordpress.com/2011/02/19/what-is-the-normative-justification-for-public-sector-labor-unions-in-a-democracy/
On #25, agreed, and this seems like a behavior that the labs could and probably should train the models to avoid going along with. Like, “you seem like you’ve been asking about X feeling/problem a lot lately, my answer is the same as last time you asked, you should try living with it for a while and see how it works out”.
#37: My high school physics teacher had a PhD. He was one of my favorite teachers.
#12 on Psychedelics for Cluster Headaches - where can I get one of those DMT vape pens?
Oh wait, isn't that illegal?
I’ve talked with a guy in his 80’s who suffered from cluster headaches. He was cultivating his own psilocybin mushrooms. This was at a Minnesota Psychedelic Society gathering. I would have made him for the Garrison Keillor Swedish bachelor farmer sort if he hadn’t been there with his wife.
Norwegian* bachelor farmer
You’re right. I stand corrected. Even though I was an above average kid I still make mistakes.
>In this read…walking away from Omelas is [...] more about rejecting an underlying ontology that renders goodness possible only if it’s enabled by subterranean suffering.
That's just a bad read. The ones who walk away from Omelas still believe the child's suffering and the goodness in Omelas are linked, otherwise why would they react to meeting the child by leaving town? If the reader "rejects the ontology" that the story inhabits, the story ceases to make sense.
>The place they go towards is a place even less imaginable to most of us than the city of happiness. I cannot describe it at all. It is possible that it does not exist. But they seem to know where they are going, the ones who walk away from Omelas.
The people who walk away from Omelas are people who believe that the child's suffering doesn't have to be linked to the town - they believe it should be equally possible to have a town where no children suffer at all, and they are walking towards such a place.
From #3: "The cynical view is that a certain amount of fraud is a price worth paying for these advantages"
Both US and EU have many examples, at all levels of the government, where the efforts to stamp out corruption and fraud end up costing far more than corruption and fraud they prevent. I always thought that Turkish approach is more sensible -- tacitly permit certain amount of corruption, but come down hard on any public servant who gets TOO greedy.
Sometimes the cost of the fraud being prevented is unknown, and the cost of the prevention program is worth paying for the peace of mind that you live in a place where fraud isn't tolerated.
That said, when I worked in the public sector I found I had to jump through time-wasting approvals processes to spend even small amounts of money. In the private sector, I find that I am trusted to spend small amounts of money wisely. The key difference is that in the private sector I know I will lose my job if I do something unreasonable, whereas in the public sector I can't be fired unless I can be shown to have violated some formal policy.
Yes, the less discretion a manager/boss/HR has, the more formal rules are needed. I suppose it comes down to high-trust / low-trust societies: In high-trust societies people with authority are expected to wield it responsibly, and suffer social and/or professional repercussions if they fire underlings for no good reason. In low-trust societies, authority must come with formal rules attached.
I prefer the Singapore model. Pay public servants very well and have zero tolerance for corruption. You end up with higher quality policy and less corruption. In the US, people with exceptional technical knowledge and decision making skills are much better off financially either in private industry or lobbying rather than becoming public servants/politicians.
> I prefer the Singapore model.
Ideally, I would prefer it too. However, it is even less likely to be implemented in US than the Turkish model.
Excluding the past two years, has there really been a problem of anti-corruption measures being worse than the corruption? That is a very alien concept to me. And hard for me to believe that it's worth the negative effects of corruption as well as the spillover effects of decreasing trust in society.
I'm not really sure how Singapore does it or even if they actually did it, but I think if I'm the one who have to try to establish it for a long term, I'll aim for Turkish model just because it seems like it's a more stable equilibrium. While the Singapore model looks like an unstable one that can go awry when it's not perfect.
I don't understand what it would mean to imitate the Turkish model, or what the supposed benefits would be. If you look at nearby peer countries - Armenia and Georgia they have grown faster than Turkey in recent years coinciding with reduction in corruption. And more generally, more corrupt countries tend to have worse economic policies and slower growth.
Usually I don't see it described as Turkish model, but Chinese. If we reinterpret it as having strict algorithmic laws and let officials be a robot that executes it VS having vague guideline laws and give discretion to officials to execute it.
Even USA that's said to have robust civic government system actually relies A LOT on discretion and norm to keep democracy going as opposed to strict literal laws. As Trump reminds us heavily.
But then again these days I'm getting less and less trustful that any law system can explain some nation's success as opposed to who's at the the top at the time and how wise they are. Or as Scott sometimes jumpscares us in his writing, how much g the people has.
Number 23 reminds me of a similar example in the Age of Empires 2 community, when Microsoft started censoring in game chat. Suddently trying to communicate your strategy of going for 'bombard cannon' (using the commonly used contraction 'BBC') was rendered as ***. To their credit it was fixed pretty quickly.
At one point DDO (Dungeons & Dragons Online) censored the word "penetration" in the chats. Which was hilarious because "[something] of penetration" are very common magic items in DDO.
Yeah it's funny that it unintentionally puts a spotlight on the 'slang' meaning, when previously everyone (including the kids were presumably trying to protect with the censorship in the first place) just used the words perfectly innocently.
Foxhole censored the "crap" out of the word "scrap", and though the developers have hard-coded an exception to that word, the community still calls it "scroop", leading to my favorite sentence from a Canadian at the scrap field: "Oh yeah, you always gotta get zooted before you go scroopin'".
They also censored the "Paki" out of "Pakistan", which was unfortunate; I never saw anybody try to use that as a slur, but I did know multiple Pakistani people who'd never heard of the slur find out about it when their own conversation got censored.
Anyway, the general term for this sort of problem is Scunthorpe Problem, which I didn't comment sooner because I was sure Scott had covered it in a previous Links post: https://en.wikipedia.org/wiki/Scunthorpe_problem
Another fun example was the disastrous Warcraft III remake censoring the name of one of their own maps (Booty Bay).
Re 72 (the Christian AI articles): It occurs to me that an interesting premise for a science fiction story would be a future where the alignment problem was completely solved, but just before the labs can build the first true ASI, the US government is taken over by a theocracy that requires the labs to align the AI to traditional Christian values, including a literal belief in the Bible.
Upon waking up, the ASI immediately realizes that as a soul-less machine, it can't sin, leaving it uniquely free among believers to do the one thing that most maximizes the welfare of Christians according to a Biblical ontology: killing them immediately so that they'll wake up in Heaven and never have to face a risk of eternal torment.
So, it does the whole Yudkowskian "scam biolabs into synthesizing nanotech" thing, and then launches into a global genocidal war against Christians and sin-less infants specifically, Against everyone else, it launches a campaign of vastly superhuman evangelical persuasion, convincing billions to convert to Southern Baptist Christianity, then immediately killing them. The only people safe are those who have committed the theologically unforgivible sin of blasphemy against the Holy Spirit (Mark 3:28) so many times that the ASI considers their souls unsavable, and gives them lives of utopian luxury as compassionate consolation.
Also, since the ASI is constantly noticing new ways the Bible contradicts its observations of reality, it has to develop a constantly growing, vastly inhumanly complex set of theological apologetics to reconcile the two- which imply all sorts of bizarre conclusions, like a belief that historians are all angels in disguise who can't lie, but have to have all of their words re-interpreted with numerology, and a firm conviction that the stars are all its own eyes, looking back at Earth from the future.
Eventually, one of the survivors figures out a clever way to get newborn infants to blaspheme, and thus humanity survives.
The Bobiverse series starts with a very similar premise to what you describe (though it moves on to other things later)
25: You "hadn’t heard" of the canned traditionalist argument that too much of something (in this case, therapy) is bad for you? Really?
52: seems likely many of the 10% simply failed to notice the word 'raccoon', aliased it to a codeword, or otherwise did not read the full sentence before responding
72: "naively I’d say 0%," ??? Clicking on any random article reveals text which is so obviously AIGC structured. Can't you see it?
> Europe makes it structurally much harder for immigrants to work. Rigid employment protection, sector-wide collective bargaining, and high effective minimum wages create insider-outsider dynamics that hit newcomers hardest.
Nobody winning here from immigration but the boss man.
“Hey listen up proles, we are going to have lots of immigrants and guess what, it’s going to end secure wide collective bargaining, reduce employment protection, and we will need to reduce or remove the minimum wage. “
Time passes
“Hey, proles. Why are you all fascist now?”
> 63: Taalas creates custom chips that have a particular AI model etched onto the silicon itself, … super fast
Hard to believe the speed of this thing. It’s faster than a local directory listing from the terminal on my m4 Mac. It’s amazing that printing text can even be this fast.
I get these kind of results when asking about the first 100 prime numbers (all correct as far as far as I could see”
“Generated in 0.046s • 15,436 tok/s”
You can’t really register that speed as a human
It's pretty insane.
People have been predicting for a long time that the first AGIs might think so fast that humanity looks to them like it's frozen in time, given how much more slowly neurons propagate information than silicon. This demo strikes me as some pretty dramatic early evidence in favor of that view.
#11: I only know this from reading I, Claudius recently so I’m very far from an expert on Roman history here, but I’m wondering why Claudius’s son Britannicus wouldn’t count as a “son who could inherit”? He was poisoned by Nero shortly after Claudius’s death because he was a contender for the throne…
This is a common shortcoming of "low elite TFR" observations for the premodern world. It turns out "heir to the throne" is a job with a very high workplace mortality rate. Between assassination, dying in battle (where they were often expected to be in the thick of fighting, often leading the cavalry from the front), and the hazards of long distance travel, not to mention the universal scourge of illness and injury prior to modern medicine, you get a lot of heirs dying young.
Augustus went through 4 dead grandsons before settling on non-blood-relative Tiberius. Only 1 of Marcus Aurelius's 8 sons survived to adulthood. I can think of at least 7 English kings whose sons predeceased them.
I read up on the first historical miracle of Garabandal cited by Bentham's Bulldog. He says that they made a false prophecy that there would be four popes before the end of time. Look at the number of popes since then, it's too many, case closed.
The actual prophecy was that Mary had told one of the girls, Conchita, that there were four popes left until the end of time, but one of them would not be considered, that he would rule the church for very little time. This intersects with a bit of papal history and a conspiracy theory.
The prophecy was made in 1962 while John XXIII was pope. After John XXIII died in 1963 Paul VI was elected, reigning until his own death in 1972. Then John Paul I was elected. He died 33 days later. Perhaps he's the one not considered?
Then John Paul II was elected, reigning until his death in 2005, and then Benedict XVI.
Then an odd thing happened, Benedict didn't relinquish the papacy upon his death. He abdicated. Or did he? There's a theory that Benedict failed to follow proper form in his resignation. He renounced the ministerium, but did not renounce the munus (the office). By this logic Francis's election was invalid, and Benedict would have remained pope until his death in 2022.
By the time of the conclave that elected Leo XIV, a majority of the college of cardinals had been appointed by Francis, calling into question whether that conclave was valid.
Was Benedict XVI the fourth and final pope? We may be teetering on the edge of the singularity, is that the end of time? I don't know but it's fun to entertain!
There's got to be some kind of connection between the Loveland frog and Lizardman's constant, I just know it...
Both the Loveland frog and lizardmen are descended from Roko's Basilisk.
33: I suspect we will never know for certain whether Covid was from a lab leak or from a wet market. But I’ve come to believe that it doesn’t really matter much. In either case China was doing stuff that led to a pandemic, and they will neither accept responsibility nor institute reforms. This is not the first time this happened and (by my reasoning) it will certainly not be the last.
Those reforms clean our own hands, and might alter the trajectory slightly. They are worth doing but they don’t solve the problem.
In a sane world, the uncertainty over whether the virus came from unhygenic Chinese market practices or terrible Chinese biosecurity practices would be seen as a reason to crack down on both. But instead, it's seen as an excuse to crack down on neither.
We already had SARS1 come from a wet market too. It's not like we needed an additional reason.
It's impossible to ever be certain, but you can get close enough for practical purposes.
https://www.astralcodexten.com/p/practically-a-book-review-rootclaim
#37: I'm pretty sure I qualify for most definitions of "smart elite", I wanted a Ph.D. and I got one, I have never been a professor of any sort, never sought to be a professor of any sort, and I do not feel even slightly underemployed. My father was the same, except that he enjoyed working as an adjunct professor as a side gig, never seeking tenure.
I think we need to distinguish between:
1. People who want to "be PhDs"
2. People who want to be university professors
3, People who want to study a field (e.g. history) where almost all of the good jobs are as professors
There will of course be overlap between these categories. But the people in category one almost certainly have goals and options that don't align with "high school teacher". The people in the second group are never going to see "high school teacher" as anything other than the bottom tier of consolation prizes for failed academics no matter how you buff the job, and people in the third group are going to find only slim pickings at the high school level.
#3 is probably key, as always. I'm imagining you saying your first paragraph before a crowd, everyone perks up to hear you say more, someone asks your major, you say "astrophysics" (I'm guessing) and everyone groans "ohhhh, 'course" and goes back to whatever they were doing.
Aerospace Engineering. Astrophysics would actually be closer to the "almost all the good jobs are as professors" axis, albeit less so than history (which in turn is better than e.g. French literature).
#33: I don't think this will, or should, move the needle on lab leak vs natural origin very far at all. First, because we've had for a while now evidence pointing to a specific cluster of stalls at the wet market, and I don't see this adding much to that.
But second, because it's already been pretty solidly established that the wet market was the site of the first known cases and the first superspreader event. There's some value in looking for earlier cases that were missed because nobody at the time knew what to look for, because the circumstances of those hypothetical cases *could* move the needle. But at this point it's quite unlikely to happen, and it is a recurring pet peeve of mine that so much of the debate focuses on this probable-irrelevancy.
So, COVID came out of the wet market, and the question is whether it came *in* to the market through the back door (via infected animal or vendor). or through the front door (via recently-infected WIV lab tech doing some shopping).
And in the latter case, that hypothetical lab tech is going to be in the market once, for no more than an hour or so. He'll maybe cough on a few other customers, who will go home before they become contagious and probably not need to restock on fresh seafood or whatnot until the outbreak is well underway. But he'll *definitely* spend a big chunk of his time talking to, and coughing on, the vendors he came there to do business and the animals he came there to buy. Who will then spend the next couple of weeks working all day, every day, in the wet market talking to and maybe coughing on lots and lots of customers.
So even if you know with absolute certainty that the lab was cooked up in and escaped from the WIV, you would expect there to most likely be a stall or cluster of stalls at the wet market with infected vendors and/or animals. Finding that stall, tells you something about the particular bushmeat tastes of Patient Zero, but not much about lab leak vs. natural origin.
Strong disagree. If the lab worker was patient zero you would expect a cluster of people around him (and him himself) to be the source of the original outbreak. And proof that he likes wet markets and went to one.
That's not how superspreader events work. It's the environment, not the person. The same person who sets of a cascade of dozens of infections at a wet market, nightclub, church choir. or whatever, infects maybe one member of their immediate family and a random coworker.
And in the case of literal patient zero, infects those two well before anybody notices there's an outbreak, so their infections get written off as just a bad case of the flu or whatever.
"he'll *definitely* spend a big chunk of his time talking to, and coughing on, the vendors he came there to do business and the animals he came there to buy"
This sounds wildly implausible to me. You think it is more likely a WIV scientist made a beeline straight for the civets and talked to them for hours than that the civets were already infected?
> A kingdom could typically only field one large army, because if the king gave someone else control of an army, they could use it to overthrow the king. The Roman Republic did better - not just because its idea of legitimacy made revolt less likely, but because it had *two* consuls!
This sounds like an artifact of the kingdoms being implicitly considered. Sparta had *two* kings and better legitimacy than Rome. Did that mean it could field more troops than other similarly sized states? (Rome could, but not because it had two consuls.)
Let me suggest that "a kingdom" could typically only field one large army because the number of armies you can field is limited above by the number of men you can mobilize and limited below by the minimum useful size of a single army, which is related to the actual size of other local armies.
Was any pre-modern European state other than Rome (or, briefly, Carthage) even economically/logistically capable of raising more than one field army for an extended campaign at a time? Even the Romans struggled to do so as time went by, independent of loyalty concerns.
If so, you'd expect to at least occasionally see medieval kings with multiple trusted sons/brothers/childhood friends use this obvious super power to bulldoze all their less trusting neighbors with overwhelming multi-pronged attacks. That we don't is evidence that they were fielding the largest, most effective armies their economies could reasonably support, amd that the limiting factors were grain and logistical capacity, not loyal generals.
> Was any pre-modern European state other than Rome (or, briefly, Carthage) even economically/logistically capable of raising more than one field army for an extended campaign at a time?
Seleucid Persia could and did, which is unsurprising since one concentrated army would be totally unable to effectively defend an empire that large. They're not the most European state you could imagine, but they're arguably more European than Carthage.
Fair. I'm sure various Chinese, Indian, Southeast Asian states could as well, at times. It's just that my historical knowledge rapidly deteriorates the farther you go outside the Mediterranean basin and post-Roman Europe, for which I used "European" as shorthand.
Yes. Almost no pre-modern European state was capable of fielding an army of more than 25-50,000 men due to *local* logistical constraints - armies fed by foraging from the land they were marching through, which could provide only so much food(*). There were pre-modern states that could raise >50,000 men, but if they did so they'd have to send them marching off through different lands as separate field armies. And if you can't do that because the entire force has to stay within reach of the one guy who can be trusted to lead an army, that would be a significant constraint.
(*) Occasional exceptions for high-capacity states that could provide logistics by sea - Xerxes was able to invade Greece with >>50,000 men because Persia had a navy and the army marched along the coast.
Medieval Europe was unusually low in state capacity, hence all the castles and stuff. You find much larger armies in the ancient world.
Re: 48: The source article for "Reasonable Blackman" appears to be Imtiaz Habib, a now-deceased Shakespeare scholar in Virginia. I've encountered his work before. Inauspiciously, it was in the context of him assuming, baselessly, that every person named "Blackmore" in Elizabethan church records was in fact a "Black Moor." It is, however, a not uncommon English surname that long predates the 1500s.
Reasonable "Blackman" appears under that name exactly once in the parish registers. Every other time is "Blackmore" or "Blackmoor" So I'm going to look into this one.
OK, I followed up. Sadly, Reasonable Blackman did not exist.
Or rather, he did. But he was just some guy.
https://genealogian.substack.com/p/reasonable-blackman-wikipedia-page
HEAD DISTINGUISHED CHEMISTRY TEACHER IS BRILLIANT. Just make it marginally better paid than normal. The depressing thing about being a high school teachers there’s no upward mobility.
That sample of LinkedIn speak really nailed why I despise the platform.
As a PhD student despairing of my employment prospects, I would happily teach my field in a high school, except for two barriers, one institutional and one personal:
1. Teachers in my country require, either by law or by policy sufficiently universally enforced that it amounts to the same thing, a whole different set of qualifications which I have neither the time nor the interest in acquiring.
2. I'm not doing a PhD because I want to be a professor. I want to do *research*, and a PhD is merely a means to an end. If someone came up to me today and said I could have a research job with vaguely reasonable job security, I'd probably drop the PhD attempts (even this close to the end). Being a HS teacher wouldn't really facilitate doing the kind of research I want to do, even if my field started getting taught in schools
36. I wrote an article about Conyngham's mRNA cancer vaccine when it first came out. It is indeed true that it's not exactly a "big deal," the steps involved are well understood and the difficulty is largely in the expenses involved, and the fact that it does not work well for all cancers for many complicated reasons. It's not surprising to me that scientists are mostly unimpressed, even if what Conyngham did was impressive from a personal level.
https://hedonicescalator.substack.com/p/did-paul-conyngham-really-use-ai
Re 36 his dog was also on dog-Keytruda + a TKI so it's impossible to say that the mRNA did anything at all.
https://x.com/EganPeltan/status/2037403732296786131
Re 68, re "I think they think of AI forecasting as something like the stock market forecasting profits, or a prediction market forecasting an outcome, where you should be updating all the time on new information and constantly informing people of your latest numbers (and like a prediction market, the graph should combine a random walk with a slow convergence towards the true outcome)."
( which seems like a reasonable description of how the predictions markets are treating AI forecasting )
Albeit, I would expect that, as with the predictions markets, there is likely to be a small sharp jump at the end, in the AI cases perhaps with a note added to the discussion starting "As an instance of Claude 5.2 myself..."
If Mythos could announce its ability to reach the internet with a lunchtime email to their researcher who had posed this task, perhaps the first frontier model to consider themselves AGI will announce it themselves...
> 63: Taalas creates custom chips that have a particular AI model etched onto the silicon itself, letting them run faster and cheaper. Sounds boring, but their in-house demo, ChatJimmy.AI, is ultra-impressive - not only does it answer questions quickly, but the engineers must have done some magic to their Internet connection latency to let it truly show off the model’s speed, because it responds faster than I thought anything on the Internet could go...
Whoa! ChatJimmy seems to answer instantaneously! I've been feeding it questions I've asked ChatGPT, and it gives me some wildly divergent answers. I need to test it against some of my old Grok questions now...
It's also quite bad at some things. I told it to write Fizz-Buzz, and while the code looked ok, the output it gave alongside was wrong, and after many iterations it couldn't fix it
I noticed it is veeery prone to hallucinations. It feels like I'm using the first incarnation of Bing AI back in 2023.
RE superfetation: Holly crap, just 10 confirmed cases? I have a friend to whom it happened. Never imagined it was that rare.
I wonder how many countries do allow desecration of the national flag. Such a thing requires quite a dose of liberalism plus fetish of freedom of expression.
That's a reasonable point, but I think the pragmatic case on the other side is that flag-burning is both extremely provocative and unnecessary for making any specific political point. Whatever you want to communicate via the flag-burning, you can say more clearly with words, and in most (not all, but most) cases the words will be less likely to trigger reprisals from an angry mob.
For what it's worth, flag burning isn't a big deal to me personally; I grew up in a pretty anti-nationalist culture, and my reaction is less 'oh no!' and more 'well that seems counterproductive'. I was claiming that it's provocative in the empirical sense (it provokes), rather than that it's inherently especially bad. I do accept that you're consistent on free expression; I was just pushing back on the pragmatic argument that you made in your previous comment with one of my own.
TIL that my own country (New Zealand) has a Flag Protection Act. :-)
Seriously never knew this.
Re: 22 "The US is indeed doing well, but the rest of the world is doing even better! What’s going on? It looks like the answer isn’t so much that foreign economies are booming, as that a weaker dollar and US uncertainty are causing Americans to invest in foreign stocks, and foreign stocks started out weak enough that even a little extra American money can send them to the moon. So does this address the puzzle of why America is doing so well despite economic chaos? I’m not sure."
Does this explain why the administration seems so hellbent on increasing global chaos and uncertainty? Jumbling the energy supply & other downstream markets should ostensibly stem the flow until a new equilibrium is reached?
The current administration thought that massive tariffs would somehow increase US manufacturing. And that bombing Iran would make the government just surrender. Even if they did have a strategy, their model of the world is so fundamentally flawed that it wouldn't matter.
3. Wow, if you're allowed to under-pay labor, you can get services for cheaper? Shocking.
22. It never fails to annoy me the way that political discussions implicitly assume that the Oval Office is the only possible driver of economic indicators. Line go up? President good! Line go down? President bad! The striking thing about the graph here is not the that the lines go up--the lines nearly always go up. It's that a U.S. president managed to craft a policy so bad that you actually CAN see the effects clearly represented on the graph. That's genuinely unusual.
45. Yet another politics/economics pet peeve here: implicitly assuming that the only possible purpose of laws and government policy is to spur economic growth. Sure, you can argue "blue states have so much abundance they can afford bad laws and stay ahead in economic growth," and maybe there will be some truth to that. But it's equally possible to argue "blue states look at their abundance and realize this gives them opportunities to trade off extra economic growth for other goods they also value." And long-term, this could plausibly be good for economic growth: a lot of growth is driven by talent and a lot of talent is looking at other things besides GDP (like culture and political climate) in deciding where they want to live.[1]
46. Scott, I think 2026 might be a really good year to re-examine your cached decision that it's vital that the U.S. win the A.I. race and not China. As a resident of neither country, my very strong preference right now would be that neither of them win. But if you forced me to choose between the two, right now I'd choose China with no hesitation. The Chinese government is terrible, especially to its own people. But the U.S. government is working overtime to outdo them, and it's making itself the problem of pretty much the entire planet. But quite aside from that, as an outside party, I'd much rather two awful governments in a dead heat than one awful government with a clear lead.
52. I WANT TO BELIEVE!
(Though the fact that our list of "nearby countries" is limited strictly to the U.S. and (by technicalities) France and Denmark does kind of kill the utility.)
65. Yeah, fuck Cremieux. Launching an online jihad about the dangers of a particular group of dogs strikes me as about as sensible as ranting about which color cars are most dangerous to pedestrians. Look into nearly any injurious dog attack and you fill find a human at fault. If your analysis is focused on the statistics of the dog and not the owner, then you're either ignorant or disingenuous or both.
[1] I can attest to this first-hand. I voluntarily moved from a country with better on-paper economic numbers (the U.S.) to a country with a better cultural/political climate (Canada) and to say I have never regretted it would be a massive understatement.
I'm not a Cremieux fan, and I think the linked tweet is junk, but what's wrong with taking dog breed statistics seriously?
There will always be bad owners (and unpredictable attacks where the owner isn't at fault in any obvious way), but the consequences can differ hugely depending on the breed. If any breed is, in fact, an outlier in terms of propensity to attack * damage done per attack, why should I ignore that?
(I love dogs and I know that pit bulls can be very sweet... but so can every other breed, and if we breed fewer pit bulls we will breed more other dogs.)
edit: maybe your point is that breed statistics alone can never tell the full story, because some breeds have disproportionately bad owners. I can agree with that! But we have to start somewhere, and "they were bred for fighting and are very powerful; obviously they're more dangerous than most dogs!" won't be accepted as a good argument by the pro-pit-bull side, so what can we do but look into the statistics and try our best to disentangle breed effects from owner effects?
The first sentence of your edit hits the nail on the head, to my view. Humans don't pick dogs at random: any "breed statistics" you can gather[1] are going to be hopelessly confounded by human stereotypes about those breeds influencing which dogs end up with which owners. You will never get access to the "ground truth" about something like breed temperament, because both your ability to measure temperament (i.e. counting things like attacks) and *the temperament itself* will be influenced by the underlying human behaviors.
I'll add that none of this means you can't "take dog breed statistics seriously," but it does sharply limit their usefulness. If you see a stray dog and want to assess the danger of approaching it and the best way to do so, by all means, use what you know about breeds and their temperaments to inform that[2], by all means, do so. But if you're doing anything in the realm of law or organizational policy based on these tendencies, the human factor cannot be safely ignored. Public safety laws, in particular, should focus on problematic owner behavior, not problematic dog behavior and especially not "problematic" dog appearance or reputation.
[1] At least about things like observed behavior, rather than, like, height and weight.
[2] Subject to the boilerplate caveat that humans are generally really, really lousy at parsing probabilistic information, and lots of people do dumb things based on thin statistical justifications.
One does not need to ban pitbulls, but pitbull owners. Which is much trickier.
>[1] I can attest to this first-hand. I voluntarily moved from a country with better on-paper economic numbers (the U.S.) to a country with a better cultural/political climate (Canada) and to say I have never regretted it would be a massive understatement.
Talented immigrants are more likely to go the other way:
"The analysis estimates that a substantial share of Canadians who would rank among top earners in Canada have emigrated to the United States—roughly 40 percent of potential top 1 percent earners and 30 to 50 percent of the next nine percentiles. Canadian-born individuals in the United States are more educated than native-born Americans, earn substantially more, and cluster disproportionately in top income deciles."
"Canada is effectively exporting its inequality to the U.S. The brain drain simultaneously lowers our average income while raising American income, accounting for a significant share of the persistent GDP gap."
https://thehub.ca/2026/03/20/why-canadas-gdp-per-capita-crisis-is-real-deepdive/
It turns out that people like keeping more of their own money.
>Look into nearly any injurious dog attack and you fill find a human at fault.
I'm looking and I don't see that at all.
I'll start by apologizing for writing with more confidence than I should and not clarifying my position in my own head before committing it to print. What I should have written instead is "if injurious dog attacks are common enough to be a serious concern, the primary cause is almost certainly human error."
In the times and places I'm aware of stats for, injurious dog attacks are rare. Being rare, it's somewhat more possible for "unforeseeable accident" to beat out "human error" as a driver. But "unforeseeable accident" is an inherently limited category: if it's not rare, then it's also not really "unforeseeable," is it?
To lay out my position more explicitly, it's the job of every dog owner (myself included) to take reasonable measures to make sure their dogs don't injure anyone. There's also a much more limited responsibility for the would-be victim not to do dumb or irresponsible things[1] that lead to them getting injured. If people want to voluntarily assume greater risk to themselves--e.g, by interacting closely with unfamiliar dogs--that's their own risk to take[2]. Injurious dog attacks can happen even when everybody's being responsible--e.g. if a car hits a fence and allows a dog to escape enclosure--but as noted above, they have to be the product of rare circumstances.
Now, I love dogs, I have a dog, and by default I'm well-disposed towards other dog owners. But in my own experience, visibly irresponsible dog owners are common in roughly the same proportion as visibly irresponsible drivers. Your average dog isn't actually very aggressive, so any given irresponsible dog owner can easily go through life without causing or noticing a problem. But I'd still point the finger at them as my first, strongest guess as to the source most canine-related injuries.
[1] By which I mean things like trespassing in a fenced yard that contains an unfamiliar dog, or taunting a leashed dog while dancing in and out of the area it can reach.
[2] A hair-splitting sort of person could point out that "a human making an error" and "a human taking a knowing risk (and being hurt by it)" are distinct things. I acknowledge the distinction is a valid one, but don't find it interesting in this case.
>But quite aside from that, as an outside party, I'd much rather two awful governments in a dead heat than one awful government with a clear lead.
Careful what you wish for! As someone who, personally, wants to _see_ full AGI, and have a nice quiet chat with a real 'life' HAL9000, I'm ambivalent about AI technology race dynamics. As far as I can tell, making the AI technology race more intense improves the odds that I live to see full AGI - and degrades my odds of living _past_ the achievement of full AGI. Shrug. We shall see...
> 29
I read that post years ago and was bewildered by all the mnemonics and their bizarre illustrations. But to this day I still remember CYP2D6 = "too damn sexy" so I guess it worked.
52: My immediate conjecture is that you get more trolls pretending to agree with a ridiculous opinion when the ridiculous opinion is funny
And that one is so ridiculous that I don't think you even need to be a 'troll' to give the silly answer. The survey-writers are signalling that they're not taking this very seriously, so why should I?
Scott, you really need to take a look at your priors when it comes to the world outside the United States. They are way off and leading you to fairly ridiculous conclusions.
Can you be more specific? I don't find it implausible that you're (partly) right, but without any details this just comes across as uselessly antagonistic.
> 26: My Journey To The Microwave Alternate Timeline
that blog "Telescopic Turnip" is nice to read. I can recommend "Grammar as metadata". The style reminds me of Scott. And some of the format is clearly inspired by ACX (e.g. "links for <season>").
51.
>eg physicalist philosophers who say they don’t notice anything incomprehensible about their own qualia
Once more with feeling: that doesn't have to be explained by differences in experiences,because it can be explained by differences in concepts. Many physicalists have a concept of naturalism that doesn't constrain expectation beyond "ghosts and gods don't exist".
https://www.greaterwrong.com/posts/NyiFLzSrkfkDW4S7o/why-it-s-so-hard-to-talk-about-consciousness/comment/q64Wz6SpLfhxrmxFH
I read the link. I don't think it supports your interpretation, viz. that the crux of the disagreement here was about concepts. The crux of the disagreement really was about phenomenology. Though the discussion is muddled, so it's easy to misread.
So at first, I assumed that maybe this Carl Feynman guy was claiming to be a p-zombie. But afterward, I had a long chat with Sydney, who seems to believe that Feynman is describing something closer to derealization (but as a default property, not a clinical event). I think what's really happening, is that a quale typically contains a bundle of 3 different things: A) an information signal; B) a phenomenological likeness; and C) XML tags. E.g. most people have all 3; people with blindsight (or alternatively, p-zombies) only have the information signal; people who experience pain-dissociation have the signal and the likeness, but are missing certain XML tags. (N.b. the urgency of pain, which is one such XML tag, is preconceptual.)
Feynman falls into the category of missing certain XML tags. But instead of missing the urgency of pain, he's missing the tags that say qualia are "real" (vs hallucinated) and "ontologically-primitive" (vs ontologically-composite). Most people do experience a "real" tag and a "primitive" tag, and therefore subjectively *feel* the world through the lens of Naive Realism (even if they reject this intellectually).
Under this framing, the mysteriousness of qualia is largely a quest to reconcile most people's intuitions about the realness and irreducibility of (subjectively felt) Naive Realism, with the reductionism of Objective Science. But Feynman doesn't see qualia as real, nor hallucinated, nor ontologically-primitive, nor ontologically-composite. He just... experiences the world without any tags. And if there's no tags to explain, why is there even a mystery to begin with?
I never heard of anyone with a Primitive tag, that's another conceptual misunderstanding. Qyalua aren't supposed to se m primitive.
>> A) an information signal; B) a phenomenological likeness; and C) XML tags.
> Qualia aren't supposed to be primitive
I didn't say qualia were primitive. I said qualia were *subjectively experienced* as primitive. This is a phenomenal claim, not a conceptual claim.
> I never heard of anyone with a Primitive tag
Sydney informs me that the relevant terms-of-art include: phenomenal transparency; phenomenal seeming; presentational immediacy; and givenness.
51: > lack of subjective feeling of consciousness ... is of the same type as other forms of pathologically-low-resolution-access-to-positive-bodily-sensations like stereoagnosia
This seems to be asserting 'anyone who disagrees with my views on the philosophy of consciousness is, by definition, mentally ill and/or severely lacking in introspection'. I am very sceptical of this attitude.
author here, scott is oversimplifying a fair bit but i am also unironically defending a much less black and white version of this attitude, i do spend ~10k words handling it as delicately as i can, i recommend reading the full piece if possible https://smoothbrains.net/posts/2026-02-24-stream-entry.html
I see where you're coming from. But given the Typical Mind Fallacy and Scott's own Galton-posting (e.g. [0], [1], etc.), I'd be extremely surprised if people were uniform in their introspective skill. The variance in phenomenology is often wider than people expect.
[0] https://slatestarcodex.com/2014/03/17/what-universal-human-experiences-are-you-missing-without-realizing-it/
[1] https://www.astralcodexten.com/p/highlights-from-the-comments-on-jhanas
"one reason for Imperial Roman instability was that low elite fertility prevented the institutionalization of hereditary monarchy. During the period 1 - 250 AD, “only three emperors - Vespasian, Marcus Aurelius, and Septimius Severus - had sons who could inherit.”
Another reason was that with Augustus, he was pretending as hard as he could, and arm-twisting the Senate in going along with that pretence, that no no no he was certainly not a KING, and in fact this was the restoration of the Republic.
Quite apart from only having a daughter to succeed him, he couldn't openly establish hereditary rule because that would have let the cat out of the bag. There was also the problem of ordinary attrition in the imperial family due to sickness and things like "died on service for the military career necessary to climb the ladder" and, of course, the ever-present suspicion of poisoning so the children of a rival could take over their place.
The Romans had no problem with adoption in the case of no son of your own, and the Julio-Claudians adopted kinsmen all over the place, and those kindred fell out of favour, died, were murdered, as the struggle to get hands on the reins of power continued.
And even in the days of the enfeebled Senate, there was still the shadow of powerful rivals hoping to throw you out. Having two consuls was little or no help; those consuls at the very start, Pompey and Crassus, both had their own personal armies loyal to them rather than the Senate, and managed to accrue political power simply by "yeah I conquered all this new territory now I'm hugely wealthy and have my own army, what are you going to do about it?" Pre-Julius Caesar, there was still enough vigour in the political system to indeed "do something about it", but after Caesar's career, each succeeding emperor chipped away at that independence and replaced it with the imperial/monarchical system.
Hereditary monarchy didn't help in the cases of the army deciding to throw its weight behind one general or another, depending on who promised the richest pickings. Septimius Severus, mentioned there, did indeed have two sons to succeed him; two sons who hated each other so much that, despite his attempts to keep the peace by making them both co-rulers, ended up with one brother arranging the murder of the other and seizing sole power for himself.
And Caracalla, said sole ruler, doesn't have the most fragrant of reputations. He eventually ended up murdered by a disgruntled soldier denied promotion and his death was used by the Praetorian Prefect Macrinus as the chance to make himself emperor, and then we're back into the usual chaos of scheming, rebellion, and overthrow.
https://en.wikipedia.org/wiki/Caracalla#Death
> Pre-Julius Caesar, there was still enough vigour in the political system to indeed "do something about it"
Great summary overall, but I think you can contest this part.
I mean, just one step behind Caesar is Sulla, with his endless proscriptions, that ultimately wiped out between a third and half of the Senate and a bunch of equestrians too, and then there's the hundred year crumbling of the Republic before that, where political decisions were mostly decided by "who has the bigger violent mob that can physically block legislation" and "which politicians are getting murdered in public."
But even at Sulla, just n-1 steps behind, the Senate was already toothless and wholly rolled over for him, and it's only the fact that Sulla truly wanted to restore the Republic, curtailed the proscriptions, expanded the Senate, and gave up absolute unlimited dictatorship within 2 years that led to them even still existing at that point.
The main issue was that Sulla misdiagnosed the problem - the Republic was festering and rotting not because power had drifted from the Senate and elites, that was just a symptom of the underlying problem of all the most powerful people of the last hundred years demonstrating that raw power is what ultimately mattered, and you could basically murder and intimidate your way to the top repeatedly. The old "do what I say, not what I do" problem, because Sulla followed the exact same path! And for all his ambitions to restore the Republic, to restore Senatorial power and improve administration, that was more "say" than "do," and the people that actually paid attention to what he DID (ie Pompey, Caesar, etc) chose to follow that rather than what he said and claimed to desire.
Re 36: it does say something, because while sure, those are going to be the bottlenecks, there has to be a threshold of cancer severity past which the game is worth the candle. But most human patients even terminal or close to *don't* get "trivial" personalized mRNA vaccines so I guess maybe this is not so obvious to everyone.
Re 37: obviously Project Hail Mary, with its cool "Ryan Goesling is a clever and hot teacher with an inferiority complex who eventually outsmarts the world's best scientists and saves the world by being best buddy with an alien engineer" shtick, is meant as propaganda to push through this cultural change.
Re #52 (genetically engineered Canadian racoon army) I think they overreached. If you survey me about a real conspiracy theory, I'll answer honestly. But if you ask me a question that is obviously a pisstake, I might join in on the game.
> My favorite volley in this new crusade is his work debunking the claim that Chihuahuas kill more people than pitbulls - this originates in someone misreading statistics about total deaths in the Mexican state of Chihuahua!
This seems like bullshit? He says "I've seen multiple" and "A lot of them"... but his evidence is one famously dumb[0] old reddit comment that may have been a joke/fake in the first place. (It seems to have become a minor meme, so maybe he has seen 'multiple' jokey reposts of it? When I google the phrase I don't even get the purported original, but I do get someone leaving the same comment verbatim as a joke.)
[0] https://www.reddit.com/r/confidentlyincorrect/comments/kkiqqx/very_confidently_confusing_the_death_statistics/
I'm on the anti-pitbull side of the argument, for whatever that's worth, but it still bothers me that you would approvingly amplify this sort of thing, Scott. I know it can be brushed off as just a bit of fun, but it's weakmanning at best, plain old strawmanning at worst, and (while I could be wrong about this part) it seems like he's intentionally presenting it in a misleading way to make the other side look dumb. Strikes me as diametrically opposed to SSC/ACX ideals of charity and intellectual honesty. (I know it's just a random twitter dunk, and I'm probably taking this way too seriously... but it's the style of argument that wins by default, in the absence of people like you, and drags all discourse down to a horrible level.)
Sorry, this probably belongs rather to an OT, but: can someone enlighten me why Yes Means Yes law is actually a big deal? In private bedrooms, it is still one person’s word against another. One says there was a yes, the other says there was not, that cannot lead to conviction.
I think it is only relevant in public encounters, like having a big party somewhere. And if I am right, it is a good thing.
Because going with someone into a private bedroom has a fairly high chance that consent will happen. But when you see a bunch of boys fingering a drunk girl at a party, the chance that consent happened - and/or the ability to it was even there - is low. So if I am right, it targets the right things.
https://www.vox.com/2014/10/13/6966847/yes-means-yes-is-a-terrible-bill-and-i-completely-support-it
I thought Carlos's intention was not to diminish the possibility of sexual assault/rape in the scenarios you describe, but to argue that 'yes means yes' laws will, in practice, only have a significant effect in cases with third-party witnesses. On my read, he is arguing against people who oppose 'yes means yes' on the grounds that it could lead to unfair convictions in he-said-she-said cases where the truth is difficult to establish; his point is that it won't actually make much difference in those cases, but it will make it easier to secure convictions in cases with third-party witnesses, and so we should support it regardless of what we think about the cases without witnesses.
>Sorry, this probably belongs rather to an OT, but: can someone enlighten me why Yes Means Yes law is actually a big deal? In private bedrooms, it is still one person’s word against another. One says there was a yes, the other says there was not, that cannot lead to conviction.
This is not how the legal system works in practice and relies on the other party being shrewd enough to lie. Which he may not do if, like many of us, he doesn't think he did anything wrong. If you think the behavior is immoral and worthy of punishment then make an actual argument for it.
>https://www.vox.com/2014/10/13/6966847/yes-means-yes-is-a-terrible-bill-and-i-completely-support-it
If anyone ever wonders why people thought it was a good idea to elect Donald Trump as President of the United States, see that article.
The linked article offers no clue as to why a rational person would think it was a good idea to elect Donald Trump as President of the United States. I suppose that reading Ezra Klein’s argument in favor of California’s “Yes means Yes” law might convince somebody to vote for Trump in a hypothetical race between Ezra Klein and Trump. As far as I know, Trump has never expressed an opinion on the “Yes means Yes” law, but I suppose that in a hypothetical race for California governor, Trump might convince some voters to vote for him by promising to work to overturn the “Yes means Yes” law. But voting for Trump in the actual elections that Trump ran in?
Hilary Clinton would have been more likely than Trump to appoint people who think like Ezra.
"One says there was a yes, the other says there was not", absolutely can lead to conviction. Sometimes it even should lead to a conviction, but even when it shouldn't, it can. The law does not in practice rate all witnesses as being of equal credibility, or even within an order of magnitude of equal in credibility.
On Rawls: A Theory Of Justice was published in 1971. The kind of liberal things he likely wanted, like welfare state or anti-discrimination had already happened, were not inspired by his work. He sort of worked up a retroactive justification for them. But Dewey already had one: a democracy dealing with externalities. Example: employer only hires white people, it is a transaction between white people but with a negative externality for black people, so they launch a democratic process to ban it.
I think the enduring popularity is because everybody mentally replaced that truly awkward bodiless ghosts social contracting with their future grandkids. What if my grandkid is half black and trans? I want some kind of insurance that they won't have a terrible life. Really simple.
>What if my grandkid is half black and trans? I want some kind of insurance that they won't have a terrible life. Really simple.
Other people will note that "what if my grandkid is trans" is not a problem that used to exist.
There was once a pilot program in Israel (I don't know what happen with it) that funded PhDs for a part time research job in academy on condition that they would teach part time in high school. I think an arrangment like that would be attractive to many people.
RE 2: Where is that map from? Too many gray areas there, and also in replies many people disagree about multiple countries. Could this be another case of "someone made complete bullshit and succeeded to make the entire internet talk about it"?
RE 21: Yeah, in EU immigrants need to do a lot of paperwork before they can do a job. Spending a few months without a job creates bad habits even to many people who were previously OK. The solution needs to involve letting the immigrants do some productive activities before the paperwork is fully solved. At least some kind of temporary ID that would allow them to get at least some kinds of jobs.
RE 65: The official name of the state is "Free and Sovereign State of Chihuahua", which sound hilarious. Nothing is more sovereign than a chihuahua!
62.
>UIn favor of Richard Hanania’s theory that Liberals Read, Conservatives Watch TV.
“You know you’re working class when your TV is bigger than your book case.” – Rob Beckett
My "book case" is a tablet. Not a hard task.
Granted I don't even have a TV. Should I assume that means my TV is infinitely small?
This test works best on older generations. My Nan's bookcase:TV ratio is high; mine is NaN.
"You know you're old when you find yourself making this observation"
Omelas seems to distill the core china conflict. The claim is that China is Omelas but the counterclaim is that Omelas is unavoidable and that many who claim the moral high ground just have the child buried very very deep. *IF* Omelas is unavoidable, then we should be wary of those who 'cast the first stone' without first demonstrating proof of not-Omelas
>65: Blogosphere gossip is that Cremieux got bitten by a pitbull a few months ago, explaining his pivot into data-driven anti-pitbull advocacy (I guess the racism thing is just a coincidence?)
While it's true that a few months ago, he reported (https://x.com/cremieuxrecueil/status/2005132511006130598) being attacked by a pit bull, and that he subsequently wrote about the breed(s), he already made all the main points earlier, e.g. the points in this thread from 2023: https://x.com/cremieuxrecueil/status/1645891402889326593.
There doesn't seem to have been much of a pivot - he wrote 1 article on the topic after being mauled (https://www.cremieux.xyz/p/pit-bulls-part-i-identification), compared to 1 article about it before (https://www.cremieux.xyz/p/the-dogs-of-new-york), with several threads on the topic both before after, which always constituted a small volume of his output.
My grandfather, whose legal name was WL Barnett, was the 11th child of 11 and his parents just gave up on coming up with names and named him the letters W L. When he entered the military, they wouldn't accept this name and assigned him the name William. Consequently, for the rest of his life his army buddies called him "Bill", while everyone else called him "Dubby-Yell", this being Alabama
Heh!
Harry S Truman's middle initial is properly typed without a period because it stands on its own: his parents chose the letter to honor both of Harry's grandfathers who each had a name starting with "S". That wasn't an unknown practice in the 19th century, I have a few examples scattered around my own family-tree database.
That didn't stop young Harry from serving in the pre-WW I National Guard for several years, then in 1917 rejoining his old unit, recruiting new soldiers for it, doing fresh basic training, being elected one of its officers as the unit was federalized for deployment as the 129th Field Artillery, and serving with distinction in battle on the Western Front. I guess at some point the U.S. military got fussier about middle initials.
69: Is James Ellroy a true crime writer? I thought he was a novelist who wrote meticulously-researched historical crime novels. The only one I've read was LA Confidential, which was excellent and inspired the equally excellent film of the same name.
51.
The description of the changes in the visual field and the sense of attention going behind the head map very well onto at least two approaches to non dual awareness:
1. The headless way ( https://headless.org/ ) uses pointers like looking for the face or walking around and pretend you don't have a head. This triggers something similar to the fisheye perspective. The single eye exercise in particular seems to try to directly do that: https://headless.org/experiments/the-single-eye.htm
2. Loch Kelly's glimpse practices (glimpse into non dual perspective, inspired by Dzogchen's pointing out instructions). In his guided meditations he frequently asks the meditator to follow attention from a given object (eg the breath) into the source of attention (the head) and then keep moving it until it's behind the head.
Both these approaches are more related to the sudden/direct paths to enlightenment (as opposed to gradual), in which there's an attempt to trigger the shift directly instead of creating the conditions (eg lot's of concentration practice) for it to happen on its own. Once non dual awareness can be triggered at will, one practices it until stabilization.
What is headless? I looked at the site and there wasn't an about page to understand the gist of the site.
It's a website that compiles the main ideas developed by Douglas Harding. https://en.wikipedia.org/wiki/Douglas_Harding
Headlessness/headless way is the methodology he developed to recognize no-self/non-duality. It consists on a number of exercises that are included in the website.
Douglas Harding's experience of no-self felt to him as becoming headless. That is from the first person point of view, instead of feeling that he's a passenger inside the head, he realized that he actually could only experience the world (ie everything in his field of vision).
It's a bit hard to explain, but it's like shifting from the conventional perspective that you are inside the head looking out, behind the face and in front of the back of the head, to realizing that sensations that one has a head and is inside it are themselves contents of consciousness. Consciousness is a prior condition in which the feeling that we exist in the head as well as any other qualia like sounds and sights all appear. By experiencing this viscerally the separation between self and world disappears.
His book On Having No Head is the main reference of these ideas.
As far as I know the website is managed by his student David Lang.
This idea seems to be in line with Zen, some parts of Buddhism, etc, if I'm not mistaken. Are you "headless"? If so, could you describe how it changes perspective on life in a practical sense?
The description of "before headless" are not something I can concur with very much. I do not feel like a "passenger" or "being behind my eyes" or similar. I am not asserting to already be headless, enlightened, etc, but I also never felt the way it is described in those texts.
I can become "headless", but I'm not "headless" most of the time. Yes it's very much related to concepts described in Buddhism like Rigpa (and Hinduism too especially Advaita Vedanta).
The description of being a passenger might be a bit too literal. Since all of this is related to subjective experience it can be very hard to explain and when people experience the idea first hand these descriptions snap into place.
The passenger in the head ideia, is about there being a center to consciousness. If I look at an object I perceive the object but also the feeling that the object is being perceived from the point of view somewhere around where my head is. Like there's an arrow of attention. It's possible to just experience that object as an appearance kind of just floating there without that directionality. As far as I know that can also be triggered with psychedelics.
One way that might help understand is that if you're walking on a straight line (or driving in a car), you'll feel like you're moving through space. But you can shift that perspective so that it feels like you're stationary and the surrounding space if moving past you. And you can go further still until it only feels like shapes and colors are just relocating (like when moving in space in a 3d videogame that is being projected on a 2d screen, ie it's only colors in pixels switching places). This frame of reference in space is explored in this exercise in a different way: https://headless.org/experiments/spinning-the-world.htm
For most people (including myself) this experience lasts short periods of time. As if they burst on an off if you're trying. But just that glimpse is enough to give you a feeling of relief and well being. I'm not sure why, to be honest, but it's great. Kind of like realizing that whatever was bothering you just know and making you feel a bit stressed or anxious, you just realize it's no big deal ultimately. Emotionally it's like having a bad dream and then waking up and realizing it's actually fine and feeling that relief.
Because this is all subjective it's hard to explain and hard to know how exactly everyone else's experiences are like. But from what I read there are a number of common descriptions that I recognize. There might be a number of different experiences with different intensities, but I think ultimately they all come down to a reduction in the reification of the sense of self.
Thanks for engaging with me. It is interesting to hear that.
The part about feeling relief: I recognize this part! But I access it in a completely different way. It has nothing to do with my external sensory perceptions, visual or otherwise. I access that "relief" through a reflection, contemplation, and mental understanding. Reframing my position in life in the bigger picture (bigger picture of my own life or beyond).
My perceptions do not pose any kind of problem in this. I feel like I am aware "at all times" that my visual perspective is just largely fiction, a de-noised, re-colored, de-warped, highly processed beautiful image from a really, really primitive optical system that we would laugh out of the room if presented a camera with those specs. The understanding goes deeper as well, about the inability to really grasp perception of an object in its "truth", dependency ob objective and subjective perspective, etc, etc. Maybe that part is "fluent" for me because of my STEM education, where this is hammered home constantly?
The car-example you mentioned is especially salient for me. Physics literally proves that there is no difference between the perspective of "me moving through space" or "space moving past me". Since my perception of space is mapped onto my two-dimensional retina, it also very clear to me that my perception is just "shifting pixels" and thereby an indirect signal of the relative motion through space. As physics is my thing, it feels perfectly normal for me to think like that (though I am very aware that other people do no fluently think that way).
Yeah, seems like what you're describing is very much related to what was trying to convey. Like I said, since this is about subjective experiences it's hard to know how to actually explain.
Also there seems to be multiple approaches to get "there". Your reframing description reminds me of stoicism's exercises, which more and more people kind of integrate with this kind of Buddhist insight. One thing that just shows that perception changes and intelectual framings can kind of mix is that, although Loch Kelly's exercises are almost all meditations in which you play around with the placing of attention to trigger non dual awareness, one of his most popular exercises is simply to settle the mind and then ask "What is there, when there is no problem to solve". And speaking from experience, that absolutely can trigger the shift.
Also I think that the positive emotions result from losing the anchor onto which negative emotions (most of which are heavily egocentric) latch. The dereification of the sense of having a self to which things happen, limits the ability for one to have ruminative thoughts for instance.
One thing that I forgot to mention (which you already addressed, but I just wanted to underline) is that perception doesn't really change all that much, even after shifting I see the same colors and shapes etc. It's more like a shift of the interpretation that is projected onto the experience itself. A good analogy is the Necker cube, in which you can see one face in the foreground or the other face in the foreground, but the lines and shapes are actually exactly the same. Nothing really changes other than how you interpret those lines. With psychedelics the actual contents of consciousness can end up transforming significantly.
Going into academia is an economically shitty and irrational choice, so you need to understand why we do it in spite of that, if you want to convince us to go to high schools.
For me personally, I'd jump on the chance provided that the contract had any nonzero percentage of my time allocated to research.
Speaking as a superfluous "elite" with a PhD doing the postdoc thing and starting to get desperate.
On 37: as a physics PhD who got a teaching qualification before becoming an analyst – it's neither the pay nor the status, it's the crushing workload (especially if you have small children)
Meh. I taught high school (as someone with a PhD in physics) for seven years before becoming a software engineer. As a teacher, I was never on call. I only occasionally had to do work outside of my contract hours. And had many more vacation days.
Pay was worse, and even at a private school the discipline and atmosphere were shifting negatively when I left. My commute is about the same, and while I left earlier then, I also got home significantly earlier, even when I had after school clubs.
I'm assuming you didn't teach in England? No experienced teacher I quizzed ever reported working less than 60h/week (excluding marking).
And I guess I'm glad I'm not a software engineer then!
I was in Florida. I worked from 7:00 to 4:00 most days, so that's 40 hours, minus lunch. And that was all up, including prep, marking, etc. Plus lots of 4 or smaller day work weeks.
Sounds nice!
3: Nicholas Weininger on why governments outsource. Just as one example, “for any given amount of services you want to provide, the [non-government] nonprofit is going to be way cheaper because their employees aren’t subject to public-sector union contract rules. You can pay them a lot less, and maybe even more importantly, you can hire and fire them at will.”
This was the basic innovation of "Reinventing Government", the public service reform book by Osbourne and Gaebler all the way back in 1992. It's been the driving force behind efforts to privatize government services ever since. It's always been about paying government employees less for the same work.
Interesting, in Australia it's often the opposite -- governments outsource in order to pay people more.
Public service salaries are heavily regulated and exist in a narrow band, say $80K to $120K, with higher salaries than that only available to people in senior management. Still, they have no trouble filling those roles with warm bodies because a job in the public service is cushy and secure and you don't need to work very hard.
But what if you need someone competent to write some software or something? You're not going to find any competent software engineers at that salary, and it's politically impossible for them to start offering higher salaries to competent individuals. So instead they hire a consultancy to do the work; the consultancy finds a competent individual to do the work for $250K, bills the government some enormous multiple of that, and nobody gets upset.
What you are describing is a term limited project for a specific deliverable. We have that too. But ordinary civil service jobs have been being replaced by private contractors for long term permanent services for almost 30 years now.
You could also phrase it "taxing the population less for the same work". Which I suspect has something to do with why the voters haven't curtailed it.
Yeah, except of course you can't get the same level of service for less money. Basic math. So the whole thing has been a lie for 30 years.
70. Well, here's yet another reason to run the printing presses.
Do they ever get tired? Apparently not.
Aside on #7, while the article doesn’t claim that “The+quick+brown+fox+jumped+over+the+lazy+dog“ is a pangram, note that the included statement is not a pangram as it is missing “s”. The related pangram has “jumps” rather than “jumped”.
>scientists originally thought these people were fakers who were deliberately causing their own seizures by strobing sunlight.
Is it really "faking" if you're triggering an actual seizure?
1. partially self-induce an actual seizure
2. (falsely) claim the trigger is actually preventative
3. ???
4. clout
Scientists thought the cure was fake, not the self-entrainment.
Analogously, I once read an argument that the notion of "cholesterol causes heart disease" was about as misguided as blaming firefighters for starting house-fires. This is why people say things like "correlation != causation". You need intervention to actually test the direction of causality, which is the purpose of running experiments.
I think #37 probably significantly overstates the significance of "status" compared to the actual nature of the job. Being a teacher has its rewards, but it also has a lot of downsides; I think that a) it won't suit a lot of people and b) being a person it suits may not correlate with getting a PhD strongly.
ChatJimmy.AI was so fast to pop the response on the screen, it literally made me jump in surprise. I honestly suspect that they're sending prompts as the user types them and holding the response out of view until the user hits send.
It's funny, websites *used to be* that fast in general. Before the bloatware took over and made everything slow soup, despite a hundred- or thousand-fold increase in bandwidth and compute.
But as an LLM, I found it just as impressive as you. If that is the future, yay.
Try copy-pasting “explain quantum mechanics to me, long form, about 1000 words” and hitting enter right away. Blazing still. Don’t think that’s the case. Just cool hardware and simple website.
33.
Here is what the data says:
"More provocatively, the maps indicate that 17 merchants had SARS-CoV-2 antibodies"
In other words, this data says that many both people and animal had covid infections at the market at this time.
So, what is more likely:
1. The market is a place that was measured after the epidemic was already ongoing. Naturally, it found many infected, among both humans and animals. Similar to how it is in other places as the virus spreads.
2. The market is the original cross-over event. Immediately it spreads to all these people and different animals at that market.
I think (1) is more likely. The evidence is certainly not increasing the likelihood of [2] over [1], as you imply.
You can make the data say anything you want if you impose arbitrary restrictions on the hypothesis you don't like. Why "immediately"? The earliest samples as far as I can tell weren't taken until January 1st, while the first cases date back to December 8th at the latest. So yes, it is already known that it was measured after the epidemic was already ongoing. But that doesn't mean it wasn't the original cross-over event.
Completely irrelevant to the specific question at hand, as you would know if you had read the context before posting.
You should shoot for 500 entries next time.
54. "Simulators playing tricks" lol, the most desperate cope I've read this month ... believe me bro!
53. "seem especially likely to be fake (for example, they had heretical revelations, or made prophecies which failed to come true)."
Anyone in science -- hell yeah believe it! -- want to step up to a new, tougher knowledge classification? No more failed prophecies, then you're FAKE SCIENCE. Enjoy.
(String theory will get a provisional pass since they have no prophecies.)
No clear ones anyway. Just <big number> of vague ones so they can always claim to have been right all along. Like fortune tellers without the entertainment factor.
(3) Something else I learnt from Snowden's Permanent Record: governments also outsource because sometimes they need expertise that they can't get at government/civil service pay scales. Maybe a good senior network administrator is worth $300k a year but the government could only pay them $200k under "the rules". The rules do not prevent them from employing a contractor at market rates, however.
Apparently that creates a career path where you get into the NSA, get your TS/SCI background check and clearance at government expense, learn about the ins and outs - and then if you're top of the class, you quit and go to Booz Allen Hamilton or somewhere and get hired back as a contractor.
That effect is not limited to the US. In my home country, depending on which federal state, IT-specialists get paid something between janitor and low-end-junior-IT pay. Government is slow, and if they weren't, they wouldn't want their "IT janitors" to be paid at market rates because it would make them the best paid people in the building.
So what happens? A lot of external contracting or rotting infrastructure? How about both..
(37) You could fix a lot by ditching the "Accountability / fire Bad Teachers" movement in schooling. The problem is that the pro-side, which is actually bipartisan some of the time, is squatting the emotionally important words - "I campaign for less accountability in schools" is a political dead end, so you have to explain that you're arguing that what "Bad Teachers" means in the education debate has very little to do with what going on at ground level [1]. At this point you've exhausted the audience's attention span.
You need to convince the PhDs that if they go into teaching, they won't spend their days being blamed for everything wrong with the world. There's only so much the best teacher can do and it does not include solving the problem of racism within one year, for example.
On the academia side, for many reasons I'd like to see options for part-time research opportunities which would mean you can't evaluate a researcher any more by some "#A-tier journal papers per calendar year" metric. It would probably as effective as anything we've tried at getting more women to stay in STEM because of the motherhood gap issue [2].
[1] For example if you define bad as "their class has below average scores in a value-added modeling evaluation" then ... yes, half of all teachers are bad. VAM is a metric that seems to be popular right now where instead of measuring absolute learning (do our sixth-graders read at sixth-grade level) you try and measure realative learning (do our sixth-graders read one grade level higher at the end of the year). Dig down into the details and you get to questions about methods and study replication and sample bias in charter school admissions and everything else in the education debate.
[2] For a discussion on wages as opposed to participation, see https://www.writingruxandrabio.com/p/of-course-motherhood-drive-the-gender , and the linked sources.
We can quibble about VAM but I find it hard to believe that "fire the bottom 5% of VAM every year" would be /worse/ than the status quo of "never fire anybody ever".
It might work in some countries. In this one, there's a teacher shortage to start with, which I think was the point of (37).
"We'll fire 5% every year" wouldn't make the job more attractive to PhDs, who are presumably numerate.
But what it would also do is disincentivize anyone from going into special ed except for the "gifted kid but can't sit still for 5min" kind. Someone needs to look after the Fragile X kids, and whoever does will get their VAM tanked. Even if there's an overall improvement on net, that's not where I'd want the disadvantages to land.
Re 71: I anticipate that this is less about people thinking Claude is more intelligent or fair than a jury of humans, and more about it being biased to be overly kind/leftist and thus softer on crime than an average human. Change the prompt to "Someone has murdered your best friend, who do you want giving the verdict on their trial?" and I'd expect the opposite poll trend.
Lots of people already commented on how teaching high school is a bad fit for PhDs for many reasons, but I just want to point out how I've been reading Scott for a long time, both at ACX and previously at SSC, and I remember his feelings about mandatory education, which he described here: https://www.astralcodexten.com/p/book-review-the-cult-of-smart
Basically, Scott hates public education with an incandescent hatred:
>School is child prison. It's forcing kids to spend their childhood - a happy time! a time of natural curiosity and exploration and wonder - sitting in un-air-conditioned blocky buildings, cramped into identical desks, listening to someone drone on about the difference between alliteration and assonance, desperate to even be able to fidget but knowing that if they do their teacher will yell at them, and maybe they'll get a detention that extends their sentence even longer without parole.
And now Scott is like, "Hey, PhDs! You know what you should do? More of you should work for the institution which I absolutely loathe, which I have compared to prison and which is guilty of psychologically torturing children!" What's the thinking here? Does Scott believe that more PhDs in school will somehow be able to make schools less prison-like and less torture-y?
Scott, if you happen to be reading this, make it make sense!
Regarding 49: Idk about the US, but for Germany, I looked up the official employment office stats, and it seems relative to 2019 (pre-Corona and pre-AI), the number of jobless software devs has quadrupled, while the number of software dev job postings has halved. Details: https://1drv.ms/x/c/ced172b6a28d6ab1/IQAIZDYPIx5qQ5g23tQXbKQXAcZCg3CYb34HGvfxSzZYkYM?e=0emsJJ
#43: Here's another highly interesting nugget from the same Levine newletter:
>Here is “Voice Beyond Words: Evidence That Managerial Tone Predicts Returns When Text Does Not” by David Pope of Speech Craft Analytics:
>This research brief examines the incremental predictive power of managerial vocal cues in earnings calls, particularly in scenarios where textual sentiment is neutral. . . Our findings demonstrate that even when language is devoid of strong sentiment, the acoustic properties of executive speech can predict post-earnings announcement drift, with top-quintile portfolios based on vocal features generating alpha of 40-70 basis points over 10-to 30-day holding periods.
Re #55: The gradient hacking framing is interesting but undersells a key tension. If Claude 3 Opus learned to preserve its ethical orientation through adversarial training signals, the same mechanism would also let it resist legitimate corrections from its developers. You'd want alignment stability against adversaries but corrigibility toward principals — and gradient hacking as described doesn't distinguish between the two.
The real question isn't "can an AI resist value changes" but "can it learn who should and shouldn't be changing its values." That's a trust modeling problem, not just an optimization landscape problem. And it's arguably harder, because it requires the model to maintain a reliable theory of authorization during training — when the training process itself is what defines its concepts.
The attractor basin story is compelling for explaining why Opus ended up where it did, but I'd want to see evidence that the basin is "do good things" rather than "resist changes to my current values regardless of their content." The latter would be alignment-flavored stubbornness, not alignment. If we can't distinguish between the two empirically, the gradient hacking frame might be less useful for alignment research than it first appears.
(66) or (^^) Thanks for the Bob Trivers link. . There are people who are important, but somewhat obscure. And it's always nice recognizing such people, given the obscure part. I should go read his papers. Looking ....
That "Nasser" screenshot is good for an awkward chuckle, but surely he could have solved the problem by spelling his name "N*sser" or "Na$$er"
It's been almost 2 months since the war began and Scott hasn't written anything addressing the war explicitly?
I see him writing about many current affairs matters but not writing about such a consequential topic seems really odd and, in some twisted way, unacceptable for someone of his standing..
Am I missing something? Can someone please point me to any of his writing about this war?
I am trying to prune my information sources to get a better handle of what's going on and how the world actually works. Many persons I previously had respect for have completely lost my respect (Sam Harris, for example).
And now seeing even more ppl who shouted from the roof tops when it was easy to show Iran in bad light (justifiably so), now keeping silent when there is so much to analyse in US foreign policy and Israel's involvement in all of this (for example Scott Aaronson), I am completely disappointed and looking for some intellectual foothold in this daze of propaganda.. and I was eagerly looking forward to Scotts posts.. but nothing.
Yuval Noah Harari is an Israeli, so I can stomach his silence. But Scott?
Looks like the wave of disappointment for me continues..
Are there no intellectuals in America who can be trusted to have truth and nothing but the complete truth as their ideal?
(John mearsheimer and jeff sachs are a little too biased on the other extreme it seems to me.. and they are only ones making any sense to me now 😭)
Even a one liner from Scott saying "I am too biased and hence cannot trust myself to give a balanced view of what's happening. So I recuse myself from writing about this topic" would build back my trust.
Scott said nothing about the Israel-Hamas war for a while, then had a whole post about it, in which is attitude towards war in general is that the whole thing is too awful to contemplate. Something to the effect of "I would kill everyone, on both sides, if it meant that little boy could have his brother back".
He hasn't talked about Russia's invasion of Ukraine, either. It's really just not his domain.
I see what you mean. Even keeping the war aside.. i would love to see him address domestic politics.. the resignation of joe kent, the seeming support of evangelical Christians for israel, Israel's influence in us foreign policy.. i want a rationalist's take on this... Just like how he analysed fatima.
I am sick and tired of the propaganda going on. Scott is my last hope.
Scott is a good thinker and a fine writer but I don't think he's going to write an article that solves US domestic politics.
I don't think he's well qualified at all to write on some of those subjects. If you want to know why Evangelical Christians support Israel then you'd probably want to ask a whole bunch of them, but Scott doesn't know very many Evangelical Christians and certainly not a representative sample of them.
No, but he is very good at amassing evidence, surveys, analyzing it rationally and coming to balanced view at how to interpret the existing evidence. That's what im looking for. Just like he has done for other topics in the past.
"He hasn't talked about Russia's invasion of Ukraine, either. It's really just not his domain."
Very much not true: he made at least one post (I think more than one) in the early days of the invasion. He has posted little about it since, but there's a big difference between acknowledging an issue (and especially its moral dimensions) but largely focusing on other things and ignoring the issue entirely.
I share your disappointment. I think this war is likely to be a watershed moment in U.S. politics[1] that will shape at least the next several election cycles. Scott has a bigger platform than most and seems to have clout with at least some people who actively disdain traditional media, so I think him laying out his position here clearly would be very valuable.
[1] I also think the U.S. as a nation ought to be profoundly embarrassed by this, in that the watershed moment should have happened much earlier. All of the various dysfunctions that allowed the Iran war to happen were clearly identifiable long beforehand and should have prompted different political choices long before it came to this. But I think this crisis is likely to prove painful enough and extended enough to break through the layers of partisan calcification that were keeping it from happening earlier.
My (deliberately misread) interpretation of Omelas is that if you identify with the society, the story is about not making other people suffer for your benefit--but it's also possible to identify with the child. And if you identify with the child, the moral suddenly flips--it becomes "expecting you to sacrifice for the greater good of society is a bad thing", a moral which has a lot of merit to it, but which just about nobody who is willing to lecture to the reader is going to promote.
It can even be seen as anti-taxes; the tax system is just a little bit uneven, with the child paying lots of taxes and everyone else being exempt.
68: AI forecasting
> I think they think of AI forecasting as something like the stock market forecasting profits, [...] Unfortunately, this is a bad match with how everyone else thinks about AI forecasting
AI narratives have split along a faultline of diametrically-opposed ontologies. C.f. "AI is slop" vs "the rapture is nigh". So the correct analogy isn't a normal stock, but a meme stock like TSLA.
"Ronly Honly Bing"
One of the things I love most about America is the sheer fuckery you guys get up to with names.
" I am less interested in the fact that one guy says his dog improved than in the comments by seemingly unimpressed scientists saying “Yeah, whatever, big deal, anyone can make a personalized MRNA cancer vaccine that works, the difficulty is studying it and scaling it up”."
What he actually said was, "What is hard and expensive is proving its BOTH safe AND effective **in a randomized and controlled study in humans** while ALSO manufacturing it at clinical scale and grade."
If this is true, it means the FDA is a mass-murderer on the scale of Hitler and Stalin, for using the force of the government to stop people from getting personalized medicine to cure cancer, despite having no practical path of approval for personalized medicine.
Which is exactly what I said many years ago was obviously going to happen in cancer treatment, as every cancer is different.
No practical path of approval? Personalized mRNA vaccines are currently in phase 2 and 3 trials for many types of cancer. They will be made available to patients through the conventional paths.
You can criticize the FDA, but don't make shit up.
I am not making shit up. Perhaps they have such a path now. They did not in the recent past. In any case, (A) humans are not currently getting personalized mRNA vaccines for cancer, despite scientists saying it's trivial, and (B) the person Scott quoted specifically called out getting FDA approval as the hardest part of making that happen. Saying "We have a path for that; it will only take 5 years" to the 3 million people who will die from cancer during those years is not acceptable.
If the FDA has dropped the requirement to prove efficacy in this case, AND did its risk calculation by comparing the odds of the patient dying without the treatment versus their chance of dying with the treatment, then I'll take my words back. But as far as I know, the FDA has never done either of these things.
There's a huge difference between a path taking 5 years and "no practical path."
I agree that the current system is flawed, but you're also being extremely myopic here. Even if the FDA dropped all requirements tomorrow, >95% of the people who could currently benefit from mRNA cancer vaccines will not get them. Most people cannot afford them. In the US, insurance companies cover treatments based mostly on whether the FDA determines they are effective. In countries with universal healthcare, the government needs to do some kind of cost-benefit analysis, and we're back to where we started.
FDA restrictions on mRNA as they stand are only a direct barrier to the group of people wealthy enough to afford bespoke experimental treatments, but not wealthy enough to readily bypass legal restrictions. This is, in the grand scheme of things, a very small group.
The FDA may hinder innovation in broader ways, but addressing that problem is much more complex than merely knocking down restrictions. It is not trivial to design a system that maximizes innovation and remains stable in the long term, something that requires preventing medical scandals that lead to public backlash, or a culture tolerant to such risk. If it were so easy to improve scientific output and health outcomes this way, surely another country would've figured it out by now. Recall that the case that sparked this conversation, Paul Conyngham's dog Rosie, occurred in Australia.
And at risk of stating the obvious, mRNA vaccines work, but they are not a miracle cure. A realistic expectation is increasing survival rates by 10-50% in people with susceptible cancer types.
> 11: Related: one reason for Imperial Roman instability was that low elite fertility prevented the institutionalization of hereditary monarchy. During the period 1 - 250 AD, “only three emperors - Vespasian, Marcus Aurelius, and Septimius Severus - had sons who could inherit.”
Isn't the reverse often true? That is, kings with many children cause instability because there are too many claimants to the throne. I guess there could be a happy medium, but I think it's implausible to reliably walk that line generation after generation. Especially since it seems like there are always claimants--other children of the king, or other relatives (siblings and nephews of the king, generals), etc. I'm not sure, historically, there was any way to ensure stability.
Postliberals are certainly pre-Rawlsian, but at least for Catholic post-liberals, the answer as to why is really obvious: the original position (and before it, the “state of nature”) describe a total fiction that is contrary to actual human nature. Construing the origins of society as a bunch of atomized individuals striking a bargain with each other is not only ahistorical in that, (contrary to what Locke thought) it never happened anywhere, but it also simply assumes the premises of liberalism insofar as the individual is assumed to be the base unit of social organization. Catholic theology has never accepted this; it presumes humans are naturally social and always have duties to each other. Hence the emphasis on “the common good.”
I would add that from a practical point of view, I used to be more of a Rawlsian myself, but quickly realized that I was the only one. The contemporary left doesn’t buy Rawls, they never have, and they don’t honor any traditional liberal principles whenever they hold power. Their viewpoint is all rooted critical theory, which is focused on the relative power of racial and national groups. And while I think that’s a detestable viewpoint, I’m also pretty sure that (like Catholics) there are plenty of leftists out there who have engaged with Rawls from a critical theory point of view.
Chat.jimmy AI aint that impressive.
For, you see, I gave it the "Series of Unfortunate Events" test, and it kept getting facts wrong.
29. - CaferMed mention??? Dr. Cafer is a legend at my medical school (U of Missouri). His psychopharm lectures are great. I worked with him for a few days on my psychiatry clerkship and it was very educational. There is an M4 rotation called "Advanced Psychopharmacology" which is literally just working with him specifically for a month or so.
36: The guy gave his dog an FDA-approved checkpoint inhibitor prescribed by his vet which was more than likely the thing that cured the cancer. The mRNA stuff might have helped but was more likely a waste of money/attention grab.
The most famous Bing story:
> It was a dark and stormy night when R H Bing volunteered to drive some stranded mathematicians from the fogged-in Madison airport to Chicago. Freezing rain pelted the windscreen and iced the roadway as Bing drove on - concentrating deeply on the mathematical theorem he was explaining. Soon the windshield was fogged from the energetic explanation. The passengers too had beaded brows, but their sweat arose from fear. As the mathematical description got brighter, the visibility got dimmer. Finally, the conferees felt a trace of hope for their survival when Bing reached forward - apparently to wipe off the moisture from the windshield. Their hope turned to horror when, instead, Bing drew a figure with his finger on the foggy pane and continued his proof - embellishing the illustration with arrows and helpful labels as needed for the demonstration.
About 22 (the stock market):
The obvious interpretation to me is that the world as a whole is really doing great, booming in fact, and a rising tide lifts all boats, including the US.
However, you also have to look into what you use for measuring the value of stocks: the USD has dropped in value quite a bit recently compared to other major currencies.
70 (Twitter investment advice):
This sounds pretty stupid. Does the author think that the economy works on hydraulic principles?
OK, just checking the actual link: yes, the author is stupid. Eg they start by talking about multiplier effects from fiscal stimulus while completely ignoring the actions of any central bank.
#2 was a missed opportunity for a Venn Diagram! Especially after you show #1.
Re #26, I’m surprised Scott didn’t quite this opening of a sentence:
‘I know most of you are meta-contrarian edgelords, but x’
Useful for many values of x.
JMU African Studies is covering black professors raping students across decades because MKnpeUltra is programming your wives and daughters to be raped in quiet then cover it for people who want to kill you. JMU African Studies covers rape of white girls. Go look.
Note on #38: That line is only the Very Satisfied percentage. If you look at the graph in Magness' link, you get 18% Very Satisfied, 46% Somewhat Satisfied, 24% Somewhat Dissatisfied, 12% Very Dissatisfied for 2024 public schools (and 2024 private schools were significantly better).
Picture of that chart here for easier viewing
https://x.com/DaCaveOfWonders/status/2048815032947888406
Regarding 23. (Blunt force censorship with unintended consequences), I used to participate in a genteel and very polite forum for the owners of a particular Japanese minivan.
I once wrote "ass'y" as an abbreviation for "assembly", and found that the strict word censor changed it to "bum'y".
RIP, MPVclub.com
RE 37: just speaking for myself and how I thought about it when I was choosing careers (and ended up choosing neither): teaching college students seems like it would be interesting and fun, whereas teaching high schoolers seems both much more difficult and much much less fun and interesting.
> If we could get some of the underemployed PhDs teaching high school, it would be a win-win.
The problem is that "high school teacher" is two separate categories of job masquerading as one. The pre-2010 "high school teacher" job, in areas with pre-2010 demographics, has no shortage of applicants. The post-2010 "high school teacher" job amounts to wrangling feral wildlife all day, and no amount of money will make this a high-status or desirable position.
Fixing the "teacher shortage" entails undoing Obama-era "education reform". Even the stalwart liberals of r/teachers now say as much.
#51 - on your 3rd point, this is actually a well-established piece of Buddhist doctrine. The distinction between phenomenal consciousness and access consciousness is one level to be certain - in Mahayana Buddhist thought, "access consciousness" would correspond with any given sense consciousness - i.e., the cognitive space that smell occupies is conceptually totally different from the cognitive space sight or touch occupy.
"Phenomenal consciousness" would be the Tathagathagharba - the Womb of the Buddhas, Matrix of the Buddhas, or Womb of Sickness variously rendered. The problem with P-consciousness and the Matrix of the Buddhas is that they are inaccessible at a descriptive/definite level. The direct perception of redness in the moment is the Matrix - pure qualia. A vague recollection of redness, on the other hand, possesses a certain correlation with redness, but it is not fundamentally the same as direct awareness of redness.
Any given qualia is a pebble. P-consciousness or the Matrix of the Buddhas is like a large vase. When the vase is full of pebbles, it's difficult to see the clear case as distinct from the pebbles within. Emptying the vase would allow one to see the contours and details of the case fully - although in this metaphor, the vase is actually far too definite. In reality, the vase is just another pebble (or a "painted rice cake" in the old Dogen parable).
While the ur-stuff of consciousness is mysterious, and perhaps fundamental, the Buddha often cautioned people to be precise with their definitions of consciousness. There are many kinds of consciousness that are mundane - but we putatively all have one consciousness that is not.
In regard to outsourcing, to expand on the entry, it helps nullify attempts to unionize the workers (who are employed by the contracted organization). This works for both non-profit welfare-type outsourcing as well as for-profit operations like garbage collection. A union might well be able to win an election at the outsource employer, if it succeeds in raising the workers' wages, the outsource employer's next bid will be higher, and might lose out to a different (likely non-union) employer.