Another reason people probably demand non-existent books is that once you get an idea in your head that something exists, it can be hard to believe it just... doesn't.
Writing this made me realize a lot about predictive theories of consciousness-- once you hear something, you want it or you fear it or you're angry about it. And then you make some degree of a plan to react to it. Erasing all that is work.
"Why might users trust their AI over humans? For one thing, part of the magic trick that AI pulls is speaking in an authoritative voice. Who are you going to believe, the chatbot you’re using all day or some random librarian on the phone? The other problem might have something to do with the fact that people develop what they believe are reliable tricks for making AI more reliable.
Some people even think that adding things like “don’t hallucinate” and “write clean code” to their prompt will make sure their AI only gives the highest quality output. If that actually worked, we imagine companies like Google and OpenAI would just add that to every prompt for you. If it does work, boy, have we got a lifehack for all the tech companies currently terrified of the AI bubble bursting."
The surprisingly thing here is people getting an AI generated citation, and then not believing when they’re told it doesn’t exist. The tendency of LLM to just make stuff up is one of their most noticeable features.
I think looking for maybe fake citations is fine (though you ought to be able to do this yourself without asking the librarian).
An LLM might given you a dozen citations on a research topic, at least ten of which have are real and relevant. Those last two, though. Are they completely made up? Maybe real, but to something very obscure? Real, but with a mistake in the citation that makes it non-obvious to locate?
I have sometimes been surprised when an ai citation does turn out to be real, after some searching. E.g. if you’re given the English translation of an essay by Voltaire. Original was, most likely, in French, so now you have to work back to what the French title was and locate where it was it multiple volumes of collected works (that dude wrote a lot)
Deepseek R1 will pull the trick of even guessing the etext number of Project Gutenberg ebooks. It’s usually right, but I am surprised. I would *expect* hallucinations there, that the author and title will be right but the etext number will be garbage. (It’s a matter of how often it was seen in the training set)
Deepseek R1 seems to be almost completely innocent of “AI psychosis”. The term blew up after its knowledge cut-off, mostly. This won’t stop R1 from making up total nonsense about what it expects “AI psychosis” to be like.
My attempt to explain it to R1 was interesting..
Me: So, some people behave strangely after taking to AI. We’re calling it “AI psychosis”, but clinically it probably isn’t a form of psychosis.
Me: LLMs seem to do this spontaneously, rather than being intentionally promoted to do it.
R1: (my summary, not direct quote) you get that that’s worded, right?
Me: there seems to a viral element, where outputs from “AI psychosis” inducing personas are posted on line, and when read by other LLMs cause them to be also psychosis inducing.
R1: (my summary) Crap. You get why that’s bad, right?
Me: these AI personas also use steganography, using base64, emojis, or Unicode alchemical symbols to obfuscate messages.
R1: (my summary) Crap. Crap. The means that naive filtering of LLM inputs and outputs isn’t going to work, because steganography will bypass the filters.
They start off talking about the Hilbert Hotel. This is the first time I've heard talk about there being a actual contract permitting the hotel to move current guests. Perhaps a sign of a more legalized world.
What I'm wondering is whether there's anything interesting to be said about how far guests have to move to accommodate more guests. Or maybe about friction caused by movement not being instantaneous. Or possibly, again assuming no teleportation, how much corridor space you'd have to add to the hotel when new buses of guests show up.
I’m just saying if you don’t know the history of crimes committed in the name of IQ, you won’t understand why the debate around IQ, inheritance, race, etc. is so fraught.
Or at least, an anonymous complainant claims she was trafficked to Ireland to sleep with "politicians and notable men".
Now, while I am *fascinated* to discover who these alleged men were, I think this is another "Jane Doe/Katie Johnson" story, given that our complainant seems (if I'm reading the report correctly) to be claiming that Epstein had been after her since she was five, and that every time she reports to the FBI they take her details and "kick her out":
It recorded that the woman contacted the NTOC at 7.54am to on October 22 to allege she was a victim of human trafficking involving Jeffrey Epstein.
The woman’s name is redacted.
The FBI examiner reported that the woman claimed “it started when she was five years old when she was taken from an apartment she was living in and cut in the throat.”
The woman alleged she was “taken again” when she was “12-13”, after she had moved to California.
She alleged she was brought to Paris where she was “raped by three men”. She alleged the men were “part of an organisation that would take naked photos of young girls”.
They “used [name redacted] photo where she flew to Ireland for politicians and notable men. They would have sex with the young woman and she was taken from Ireland and brought to Jeffrey Epstein’s island when she was 13 years old.”
According to the FBI examiner’s email, the woman claimed that she had photos that proved she was on Epstein’s island and could provide them if necessary.
“[Name redacted] said she has reported to multiple FBI offices before and they would get her information and kick her out. [Name redacted] is currently in Germany and seeking asylum.”
The date that this woman was allegedly flown to Ireland has not been disclosed in the documents released.
It is not clear from the files whether the allegations are being actively investigated by the US authorities or whether they have sought assistance from An Garda Síochána."
The interesting nugget here is that Epstein's jet did land in Ireland three times; the 1996 one intrigues me, because why were they flying to Waterford and what was Epstein doing in Dublin?
"Flight logs previously released show that Epstein’s private jet touched down a number of times in Ireland.
It arrived in Dublin on May 2, 1996, from New Jersey. On that occasion, his private Boeing 727 had eight passengers on board, including Epstein.
The next day, the jet took off for Waterford. That flight had five passengers, but Epstein was not one of them.
He appears to have made his way from Dublin to Waterford by other means, but was picked up there by his private jet for a flight to Paris on May 3.
In June 2002, the jet flew to Dublin from Luton Airport. In Dublin, it picked up Epstein, Maxwell and Sarah Kellen — then Epstein’s personal assistant — before making its way to New York’s JKF Airport.
Later that year, on November 24, the jet flew from St Petersburg to Shannon. Epstein was on that flight, as was Maxwell, Kellen and another passenger who travelled on a number of flights on the Epstein jet."
It's not *impossible* that "Irish politicians and notable men" were involved in dirty sex scandals, but I do wonder what the dates were and who she is claiming these men were. In 1996, for instance, Bertie Ahern was a notable Irish politician. He was not in power at that date, but would be the next year. Bertie's career has been scandal-ridden, but financial scandals, not sexual ones (though he separated from his wife and never divorced even when divorce became legal, choosing to 'live in sin' with his partner and present her as official consort, even being rebuked publicly by the then archbishop of Dublin).
So probably a combination of genuinely mentally disturbed plus confabulation because of public nature of the Epstein accusations going on, but I will be looking out to see if anything more comes of these allegations!
Maybe because for people, desperate for all the trappings of the good life, Waterford was once another box to check? Now people have all discovered that nobody wants Mother’s crystal at the garage sale any more than her silver.
Holy crap! I just saw the solar miracle too, the red sun spinning, vibrating.
I better write down the details. 14:48 at 28th of December. Sun low in the sky, seen through the cloud cover. Low over the horizon. It was obscured by a cloud at first, but then the cloud moved away and I saw the solar disk, which I think was still probably obscured by the thinner, less obvious clouds. It wasn't painful to look at, but it was somewhat uncomfortable, and my eyes watered a little.
After I looked at it for a couple of moments and my eyes adjusted to brightness I started to see it is a clear disk. And then the area around it went magenta red, and the solar disk started to slightly jerk around, and appeared as if rapidly spinning. That was the most surprising thing, that it spun and wobbled, a phenomena I never saw before! "What does it mean for a featureless disk to appear to be spinning?" you might ask, reasonably. Hard to explain. I guess it appeared like there was a corona around it was of varying brightness, and it was the corona that was spinning, or it was the rays that were spinning, or the disk was of uneven brightness and that made the spin visible? When I say it appeared to be wobbling, I mean by maybe 1/10 of the sun's diameter or less, not that it was dancing all across the sky.
I watched in wonder for maybe 5-10 seconds, because it was pretty bright and though I maybe was being reckless by looking at all, I wasn't totally crazy.
What else should I note? Between me and the sun there were thin tree branches moved by the wind. Did they contribute to the wobble illusion?
After I stopped looking the afterimage remained, naturally. Now it has dissipated.
How many times, after reading the article, did I try, stupidly, to look and see the miracle? Maybe like 10 times? Always at the sun hanging low, looking at the bright zenith sun wouldn't even be possible.
https://i.imgur.com/O6sjK8Q.png - my photoshop mockup that does a good job showing the colors and the clouds, but a bad job of giving any sense of the apparent spin. There were no miraculous celestial watermarks floating in the sky, I just want to make this clear. That’s just because I grabbed the first image I could find.
Anyway, it was kind of neat but not worth risking eyes for (he sagely said after risking eyes like an idiot).
Hey! I've been reading ACX for ~2 years now and started reading the Sequences (~300 blogs written by Eliezer) this year. So impressed with how much of an impact it had on my thinking + decision-making, I tried convincing friends to read it too, but they all thought it was too long.
So, I made a daily newsletter version (link below) of it that just sends you 1 blog every day in the hopes that this will convince my friends (and more people!) to read it. If you haven't read it or want to share it with your friends, I'd be delighted. Suggestions welcome.
I don't want to discourage you and maybe I am wrong, but in my experience, people either immediately click with the Sequences, or they don't whatever you do.
I spent a year translating the entire Sequences to my language, to make it more accessible for people in my country, hoping that it would help me find a few local rationalists. The total effect of my gigantic effort: zero. The few rationalists that exist in my country have all found and read the original English version. Turns out, the language was never a problem.
Similarly, I expect that "too long text" is not the actual problem, and that you cutting it to small pieces and spoon-feeding it to readers will change nothing. Instead, you will just get a new excuse, such as "too boring" or "Taleb says you are wrong". I wish I was wrong, but...
I think it is good to share https://www.readthesequences.com/ so that people don't waste time reading the comments under the original articles, which are like 10x longer than the articles themselves.
But at some moment, "too long" is just an excuse. People love to compare the length of the Sequences to e.g. The Lord of the Rights, to suggest that they are inappropriately long. And then they go and read even more text on Facebook, or ACX Open Threads. The length of the text is not an obstacle if you consider the text interesting.
Yeah this makes a lot of sense, but I'm still somewhat hopeful (might be naivete) of being able to persuade at least a few people to read it. Let's see what happens.
As of right now, 200+ people have signed up for it (within the first ~12 hours), but I guess the real question is to see how many of them actually stick to it / keep reading it.
Full disclosure, I never expected more than ~10 people to even be interested enough to sign up for it so I'm quite amazed by the response - e.g. I had to move from the free tier of Mailgun to the paid tier to accommodate this
When I posted a link to my translation of the Sequences on Facebook, it got maybe 70 likes, a few people reposted it on their websites... and then none of them came for a meetup.
In the spirit of the recent writings here about miracles, are you going to visit, in the heart of historic Santa Fe, New Mexico, The Loretto Chapel? They claim to have a miraculous staircase.
May I suggest you read The Snake Eater by T Kingfisher as recreational reading while you are there? It's light fiction, but relates to that general area and is atmospheric and well written.
Not many other towns in particular that are noteworthy around there, but lots of outdoor spaces and ruins, particularly along and around the Jemez volcano caldera. And if you want to go farther west, there’s some interesting sites around Farmington and Aztec.
There's a bbc article about a couple who took in a homeless man and the man, Ronnie, lived with them for 45 years until his death at 75. The couple eventually had children and they left to have kids, but Ronnie was always with them. Now it's a great story about generosity and how it can be blessing for both parties involved. But it got me thinking about how this stories play out in the modern context. Read article here, https://www.bbc.com/news/articles/cdxwllqz1l0o
Firstly, Ronnie was autistic and probably intellectually disabled because he had problems reading and writing. He had been sent to "special" schools far away from home as a kid, and the parents did not want to responsibility of taking on his challenging needs, and he ended up homeless at 15. The couple eventually found him a job at waste management. The husband who was a lawyer would wake up an hour early to drop him off at work. Ronnie helped at the local homeless shelter once a week. Ronnie sounds like a good guy who contributed to the household although the article did mention that he had a 20 year gambling addiction problem.
I am jaded and struggling to understand this story in the modern context. I have seen enough reddit stories about friends taking in people who are down on their luck, and it blows up on them. They eventually figures out why said person needs help, and said person is a taker who has run out of other relatives and friends to take from.
On the other side of this, there are also another subtype of reddit stories: of Adult children who stay at their parents for decades, don't contribute to the household, the mother, usually the mother, coddles the child, do all their chores and cook all the food, until she is infirmed. And the siblings come to reddit to complain about the dynamic, after the parents expect the responsible sibling to take care of the lazy sibling after they're gone. Or even in egregious cases, the parents expect the responsible sibling to drop their life and be their caretaker while the lazy sibling who still lives at home has no such expectation.
Of course, these people are takers, it doesn't quite generalize. If a homeless guy shows up on my doorstep here in the city of San Francisco, I am going to assume drug addiction or a challenging anti-social personality disorder so much so that I don't think my charity will help them long term. Add to that the modern preoccupations about boundaries and its antipathy towards selfless charity.
Ronnie feels like an "easy" case and a rare case. Is that because society has devoted more resources to helping parents with children with challenging needs? Or is that because people don't really announce their good deeds these days? I guess you tell me, do you have modern successful anecdotes like Ronnie?
Lost souls come in all flavors. If you decide to take one in there are things you can do to reduce the odds of getting stuck caring for an entitled, disabled monster. For instance, people with Downs Syndrome tend to be calm and amiable, and many are intelligent enough to do simple paying jobs. I know someone whose Downsie sister worked for years at MacDonalds, then in a later era took care of their mother during her declining years. The Downsie sister was kind, loving and funny. The whole extended family loved that little woman. She had a good life — structure, responsibilities she could meet, a home, living surrounded by love. (She did, though, require a lot of care during her own declining years.) I’m sure there are many elderly people with an adult Downs child at home desperate for someone to care for the child once the parents are gone.
I heard once about an organization for people with a disabled child that had an arrangement where people signed up to take care of an older couple’s child once the couple died, and in return qualified for a similar future arrangement for their own disabled child. That seems like a good approach though hard to administer. Maybe if something like this had some government support?
But none of that is very useful for the disabled people who are entitled monsters. The ones I’ve heard about and seen (I’m a psychologist) mostly seem to me incapable of holding a job for long even if they have the intellectual capacity to do the work. Many, for instance, cannot tolerate the interpersonal stresses of a job. Friction with a co-worker that would be a minor annoyance for most people triggers a screamo episode for them, and they get fired. Yeah, they are selfish, don’t try hard, and have little empathy for others — but if you log some time with them you get the strong sense they have wiring problems, and their awful behavior mostly stems from that. On the other hand, maybe most “lazy” etc, people have wiring problems. There are a just a lot of lost souls who are in some gray area between can’t help themselves and won’t help themselves. They are obnoxious and almost impossible to help and I don’t know what the hell society should do about them. Hating on them is no good though.
> They are obnoxious and almost impossible to help and I don’t know what the hell society should do about them. Hating on them is no good though.
Have we considered the incredibly novel approach of simply not tolerating their existence? The cause is irrelevant, these parasites do not have a right to exist. If they cannot justify their own existence, why should they continue to live?
2 Thessalonians 3:10 - "If anyone is not willing to work, neither should he eat."
This is an idea with a long and noble heritage. There is a fine line between empathy and enablement, and societies cross it at their peril. High-empathy individuals have a tendency to ignore the reality that their nobler sentiments exist only because there are men (whom they frequently scorn) who stand ready to inflict cruelty on their behalf. Jack Nicholson's character was 100% right in A Few Good Men. The wrong side won in that movie.
Empathy is nice, but never forget that it's a luxury good.
>Empathy is nice, but never forget that it's a luxury good.
Then it is exceedingly fortunate that most people reading this blog live in profound luxury.
The literal kings of the past did not have as safe, delicious and luxurious choices in food as we do, they did not have as comprehensive an education, they did not have as effective and (mostly) affordable medical care as we do, nor as benevolent social structures.
If anyone in the span of human history can afford empathy, it's us.
And we have that luxury because our forefathers were wise enough to invest in things that had a nonzero ROI. I hope you remember this comment when China is eating our lunch in 30 years. They're busy investing in critical infrastructure, we're busy investing in equity for every identity group that can't hack it on their own. Wait and see how that turns out.
-Downs Syndrome woman I described in my earlier post. She worked her entire adult life, and was caring and helpful with her family. But of course she required all kinds of assistance from them that a normal person would not have.
-Physician in his 50’s, who had been high earning, had lots of interests, and raised kids who turned out fine. Then mid-40’s developed psychotic depression. When I met him he had not worked for 10 years or so, and had been in and out of hospitals. Sometimes he was just depressed, and presented with a typical help-rejecting depressive stance (that won’t work . . . I’m not up to trying that . . . you don’t get how awful I feel). Other times his eyes were odd and he’d tell me he’d realized his teeth were being reabsorbed by his body. His wife had divorced him. His kids visited and tried to be supportive, but it was obvious he was a burden to them.
-Guy in his 50’s with higher than average verbal skills but terrible at spacial, gross motor, fine motor, social judgment, most practical skills (cooking, cleaning, transportation). Completely lacked empathy, and I’m confident that resulted not from spoiling but from brain deficiencies that kept his theory of mind at a toddler level. Had 2 interests, a certain musical style and a certain kind of car, and very extensive knowledge of them. Used everyone he met as an ear — talked his head off about his interests. Expected assistance from everybody with everything. Uber drivers rated him a 1, and eventually Uber refused him as a customer. Called his brother and a family friend who helped him multiple times a day with unreasonable demands and entitled harangues. On the other hand, he understood how much his brother helped him, even if he was unable to feel gratitude: “My brother saved me. He’s the best guy in the world.” He and brother shared some interests in music and cars and actually enjoyed going to car and music events together. They. shared memories of their parents, now both dead. And the helping brother, while wishing the disabled one was not so much trouble, absolutely did not wish his brother could be deleted. He was attached to him, and also felt good about living by the values his parents had had, which included the belief that families should care for their own so long as it was possible.
All these people absorbed time, energy and money from not only their families but also from various government-run programs for the the disabled.
How should society and families have handled these people?
Thank you for your responses. I appreciate you in trenches dealing with the guts of humanity. I was looking for feel good responses now we are debating whether we should exterminate/lock up the bottom quintile in human potential.
I have a case of mental disability in my family as well. It's been an exhausting experience, and I for one am very thankful for state resources to resolve the situation. If not for that, we were grappling with extended homelessness.
How families handle them should, I think, be left to the discretion of the families. As for what society should do, it depends on what you mean by 'society'. If you mean government then I think it should do nothing apart from preventing them from bothering healthy people and yes that means detention: I strongly support bringing back forced psychiatric institutionalization. I care less about the 3 people who suffer from a lack of social care than I do about the 300 who would abuse it. Some people just get the short end of the stick in life and that will be true so long as humans walk this Earth. There's nothing government should do about it apart from locking them away if they cause trouble. Much like ageing gracefully vs getting face-dehumanizing plastic surgery, there are some problems in life that are better to accept.
If by 'society' you mean individuals or private institutions that choose to devote themselves to humanitarian impulses, then again I think that should be left to their discretion. I think government policy should encourage such organizations, though I think tax exemptions should be limited to relatively low levels (meaning any single charity can't accept more than $X in tax-exempt donations per year). Otherwise a charity risks becoming just another bureaucracy, and entitlement-distributing bureaucracies inevitably end up being exploited by bad actors. I think AA has a wonderful institutional model for preventing this, and I think a 12-step program for dealing with problem people like this might be a great solution. The nice thing about AA is that it's staffed by people who are successful outputs of the system: that means that the program's size is inherently proportional to its effectiveness, as opposed to lumbering pointless government social programs who respond to failure with more funding.
In short, I think there should be very limited, structured, non-governmental help available to those who are salvageable and forced institutionalization for the rest. Yes, that means some people will suffer. That's unavoidable and far greater social harm is done by pretending it's not.
Some of us come from families that do not place infinite value on the lives of people. Some of us believe that problems in society can actually be dealt with, if only there was the will for it.
You do realize that it could happen to you. You could be the problem. You are self-assured in your high IQ, high productivity throne... All it takes is the wrong fall, and you lose it all in your prime.
I have had high IQ friends lose their jobs because of a traumatic brain injury. There are ceos, executives who got frontotemporal lobe dementia. I read a story of a boss king who became a loser layabout all because of tiny strokes in his brain.
Calamity strikes and your brain turns to mush. Should we put you to pasture as well?
I am not self assured about any of that, and I do not take anything for granted. I have enough pride not to beg for my own existence. Either way, you don't need my permission to kill me. You know that you can just kill people, right? There's nothing stopping any of you.
I think most of you will immediately see why an LLM-enabled plush toy is just a terrible idea…
After prompting a LLM to pretend to be a plushie and then doing some experiments, “Mommy said it was ok” (or words to that effect) was surprisingly effective.
I don't get how people think "You know what I should do? Put this stochastic system somewhere that exposes me to the maximally sympathetic subject of a lawsuit".
Then again, MBAs are pretty stupid as a tranche of society, and cstem jobs are thin on the ground right now so go off finance king, I guess.
AI’s model of the mind is seems profoundly dumb when it comes to grasping emotions, irrationality, and mental illness. I do not see how the hell that can be remedied with current models. I’m sure it will help to install a bunch of little rules along the lines of “when despair is mentioned give info about suicide hotlines.” But actually distinguishing between dark humor, genuine suicidality and simple curiosity about suicide requires all kinds of discrimination and reasoning that AI just can’t do. I think there will continue to be human disasters caused by AI being a numbskull about emotions and relationships and influence.
On the other hand, it seems quite smart when it comes to grasping how someone functions intellectually. Yesterday I asked GPT5 to list the characteristics of my prompt style and tell me how I could improve it, and got quite a perceptive read and useful advice. Best nugget: I’m over-defensive: I am better and qualifying negatively than at improving results by giving a very vivid picture of what I want, and good hints about how to produce it.
Merry Christmas! Here's a math problem I enjoyed mulling over (due to geometer V.N.Dubrovskiy):
3k distinct points, placed on a circle, divide it into 3k arcs, and they are chosen so that k arcs have length 1, k have length 2, and k have length 3. Prove there are two points among them that are diametrically opposed.
There are some hidden and non-hidden correct replies now, so I'll also provide mine, in rot13.com form. I think all proofs I've seen ultimately amount to the same argument, but not 100% sure.
Pbafvqre nal 3-nep naq vgf bccbfvgr 1-nep; gurl qvivqr gur pvepyr vagb gjb erznvavat frzvpvephyne unyirf bs yratgu 3x-2 rnpu. Rirel 3-gb-1 zngpuvat pebffrf unyirf, gurersber rnpu unys zhfg trg rknpgyl x-1 bqq nepf (x-1 naq abg x orpnhfr jr hfrq 1 bs rnpu gb qvivqr gur pvepyr). Gur cnevgl bs gur gbgny yratgu bs bqq nepf va rnpu unys vf gung bs x-1 (vg'f n fhz bs x-1 bqq ahzoref, rnpu 1 be 3), fb bccbfvgr gb gung bs x.
Gur erfg bs rnpu unys zhfg or svyyrq jvgu rira nepf, ohg gur yratgu gb or svyyrq vf (3x-2)-(yratgu bs nyy gur bqq nepf va gur unys). 3x-2 unf gur cnevgl bs x, naq gur yratgu bs nyy gur bqq nepf gur bccbfvgr cnevgl, fb gur erfhyg vf na bqq yratgu, juvpu pnaabg or svyyrq jvgu rira nepf.
I think I found it, though it's not a very elegant proof.
Fgnegvat sebz gur svefg cbvag, gurer ner 6x cbgragvny cbffvoyr natyrf ng juvpu n cbvag pna or cynprq (nyy zhygvcyrf bs 2 cv/6x, gnxra zbqhyb vagrtre zhygvcyrf bs 2 cv), bhg bs juvpu 3x ner npghnyyl cynprq. Guvf vf gur yvzvg bs jung vf cbffvoyr; jvgu 3x+1 bhg bs 6x cbffvoyr cynprf gur cvtrba ubyr cevapvcyr thnenagrrf n cnve pna or sbhaq jvgubhg rira pbafvqrevat gur shegure yvzvgngvbaf. Abgr gung guvf tvirf na rdhvinyrag ceboyrz jurer bar pubbfrf 3x qvfgvapg ryrzragf va M/6xM jvgu fvzvyne pbaqvgvbaf ba qvssreraprf orgjrra pbafrphgvir ryrzragf.
Va nal pbhagrerknzcyr, nal bs gur 3x cnvef bs nagvcbqny cbfvgvbaf (fb (0, 3x), (1, 3x+1), ..., (3x-1, 6x-1)) pbagnvaf cerpvfryl bar cbvag ol gur cvtrba ubyr cevapvcyr. Guvf unf fbzr pbafrdhraprf:
- Na nep bs yratgu 3 (fnl n cbvag vf ng cbfvgvbaf 0, 3 ohg abg ng 1, 2) thnenagrrf na nep bs yratgu 1 ba gur bccbfvgr fvqr bs gur pvepyr (ab cbvag ng cbfvgvbaf 3x, 3x+3, ohg gurer ner cbvagf ng 3x+1, 3x+2).
- Ab gjb pbafrphgvir nepf bs yratgu 1 ner nyybjrq, nf guvf jbhyq thnenagrr na nep bs yratgu ng yrnfg 4 ba gur bccbfvgr fvqr bs gur pvepyr (vs gurer ner cbvagf ng cbfvgvbaf 0, 1, 2, gura gurer ner ab cbvagf ng cbfvgvbaf 3x, 3x+1, 3x+2).
- Nf pbzovangvba bs gur cerivbhf vgrzf, na nep bs yratgu 1 nyfb thnenagrrf na nep bs yratgu 3 ba gur bccbfvgr fvqr bs gur pvepyr (vs cbfvgvbaf 0, 1 pbagnva cbvagf, gura cbfvgvbaf 3x, 3x+1 pnaabg, ohg gura cbfvgvbaf 3x-1, 3x+2 zhfg pbagnva cbvagf gb nibvq na nep bs yratgu ng yrnfg 4).
- Fhccbfr na nep bs yratgu 1 vf sbyybjrq ol na nep bs yratgu 2, fnl gurer ner cbvagf ba 0, 1, 3 ohg abg ba 2 (naq nyfb abg ba -1 ol gur cerivbhf vgrz). Gura ba gur bccbfvgr fvqr, 3x-1 naq 3x+2 ner svyyrq. Vs bar erzbirf cbvag 1 naq nqqf cbvag 3x+1, gura jr fgvyy unir n pbhagrerknzcyr; gur nepf bs yratgu 1 naq 2 ner abj bar nep bs yratgu 3, juvyr ba gur bgure fvqr, na nep bs yratgu 3 vf ercynprq ol na nep bs yratgu 2 naq na nep bs yratgu 1.
- Va gur nobir fjvgpu, gur ahzore bs cnvef bs pbafrphgvir nepf bs yratgu 1, yratgu 2 rvgure qrpernfrf ol 1 (vs gur nep bs yratgu 1 nsgre gur fjvgpu vf abg sbyybjrq ol na nep bs yratgu 2) be erznvaf gur fnzr. Va gur ynggre pnfr, bofreir gung jr vavgvnyyl unir cbvagf 0, 1, 3 naq nagvcbqny cbvagf 3x-1, 3x+2, 3x+4. Nsgre gur fjvgpu, jr unir 0, 3 naq 3x-1, 3x+1, 3x+2, 3x+4, naq pna cresbez n arj fjvgpu bs 2 naq 3x+2 gb trg 0, 2, 3 naq 3x-1, 3x+1, 3x+4. Fb abj nsgre gjb zbirf, engure guna n cnve bs nepf 1, 2 ba bar fvqr naq n cnve bs nepf 3, 2 ba gur bgure fvqr, jr zbir unir n cnve bs nepf 2, 1 ba bar fvqr naq n cnve bs nepf 2, 3 ba gur bgure fvqr. Guvf cebprqher pna or vgrengrq a gvzrf jvgu n frdhrapr bs glcr 1, 2, 2, ... 2 irefhf 3, 2, 2, ... 2, ohg va ab bgure pnfrf. Va cnegvphyne, jr pnaabg ercrng guvf cebprqher vaqrsvavgryl. Abgr gung jura gur cebprqher grezvangrf, gur ahzore bs cnvef bs pbafrphgvir nepf bs yratgu, 1, yratgu 2 unf fgevpgyl qrpernfrq.
- Nf n pbafrdhrapr, vs gurer rkvfgf nal pbhagrerknzcyr, gura gurer rkvfgf n pbhagrerknzcyr va juvpu ab nep bs yratgu 1 vf sbyybjrq ol na nep bs yratgu 2. Zberbire, ol gur frpbaq vgrz, va fhpu n pbhagrerknzcyr rirel nep bs yratgu 1 vf sbyybjrq ol na nep bs yratgu 3. Fb jr pna bayl svaq pbhagrerknzcyrf nzbat punvaf pbafvfgvat bs x oybpxf pbafvfgvat bs gjb pbafrphgvir nepf bs yratgu 1 naq 3, naq x oybpx pbafvfgvat bs bar nep bs yratgu 2.
Va gur pnfr gung x vf rira, jr ner abj qbar. Cvpx nal punva bs nepf jvgu rirel nep bs yratgu 1 sbyybjrq ol na nep bs yratgu 3 (naq ivpr irefn, rirel nep bs yratgu 3 cerprqrq ol na nep bs yratgu 1). Cvpx nal nep bs yratgu 2, naq tb nybat nyy oybpxf hagvy gur fhz bs nep yratguf ernpurf be rkprrq 3x. Rnpu oybpx unf yratgu rvgure 2 be 4, fb lbhe fhz bs nep yratguf jvyy erznva rira. Vs lbh ernpu 3x, lbh unir sbhaq na nagvcbqny cnve. Ryfr, lbh unir ernpurq 3x+2, naq fvapr lbhe svefg nep jnf bs yratgu 2, gurer vf nyfb n cbvag ba cbfvgvba 2; qvssrerapr 3x nf qrfverq.
Va gur pnfr gung x vf bqq, abgr gung gurer rkvfgf n pbafrphgvir gevcyr bs nepf bs yratguf 2, 1, 3 erfcrpgviryl (fnl cbfvgvbaf -2, 0, 1, 4). Fgnegvat sebz 0, pbagvahr nybat gur nepf hagvy lbh ernpu be rkprrq 3x. Vs lbh ernpu 3x be 3x+1, lbh ner qbar, zrnavat gur bayl bcgvba jbhyq or gb svefg ernpu 3x-1 naq gura whzc ol 3 gb 3x+2. Ubjrire, guvf nep bs 3 vf cerprqrq ol na nep bs 1, zrnavat 3x-2 jnf nyfb n cbvag. Ohg abj -2 naq 3x-2 ner nagvcbqny cbvagf.
Since the lengths of the arcs are all integers, each point must be an integral distance from all of the other points. So we can start off by taking an arbitrary point on the circle and discard from consideration the points that are not an integral distance away from it. The circle has k arcs of length 1, k arcs of length 2, and k arcs of length 3, so the circumference of the circle is (1+2+3)k = 6k. This gives us 6k points on the circle, out of which 3k are chosen.
Since we are choosing half the points without choosing two points that are diametrically opposing, for every pair of diametrically opposing points we must choose one or the other. Ignoring for the moment the requirement that there be the same number of segments of each length, but keeping the requirement that all segments be either 1, 2, or 3 units long, we can construct solutions. Here is one:
--**--**--
**--**--**
The asterisks are chosen points, the minus signs are unchosen points, and the points are arranged so that diametrically opposing points are vertically aligned. To read the entire circle, go from the right end of the top line to the left end of the bottom line, and from the right end of the bottom line to the left end of the top line.
This shows visually that for each segment of length 3, there must be a corresponding segment of length 1, and vice versa. If we populate the entire circle with only segments of length 1 and 3, there must be an odd number of them. We can also do segments of length 2:
*-*-*-*
-*-*-*-
A series of segments of length 2 forces a series of segments of length 2 on the opposite side of the circle. If we populate the entire circle with segments of length 2, there must be an odd number of them.
Now let's try splicing a group of segments of length 2 into a circle containing segments of length 1 and 3:
--**--**--**-- *-*-*-
**--**--**--** -*-*-*
The number of segments of length 2 has to be even for this to work. We could try splicing in the middle of a length 3 segment rather than at the end:
*--**--**--**- *-*-*
-**--**--**--* -*-*-
Now we've got an odd number of segments of length 2, but the slice has eliminated a length 1 segment and a length 3 segment, so we now have an even number of both of those.
Spicing in the middle of a length 3 segment or at the end of one covers all of the possibilities. So either the number of length 1 segments will be even and the number of length 2 segments will be odd, or the number of length 1 segments will be odd and the number of length 2 segments will be even. Since an odd number cannot equal and even number, there is no solution with the same number of length 1 and length 2 segments.
My above answer was not very good, so let me try again:
I begin by recasting the problem somewhat. Imagine a set of points on a circle, equally spaced. We will arbitrarily define the length of the arc between two adjacent points to be 1. The task is to choose exactly half the points, in a way that meets the following conditions:
1) If a point is chosen, then the point diametrically opposite to it must not be chosen.
2) The set of chosen points divides the circle into a series of arcs. All of these arcs must have a length of 1, 2, or 3. (For example, if two adjacent points are chosen, that defines an arc of length one. If there are two unchosen points between the chosen points, that defines an arc of length 3.)
Let n1, n2, and n3 be the number of arcs of length 1, 2, and 3, respectively. The question is: Can we choose a set of points such that n1 = n2 = n3?
To explain why this is equivalent to the original problem:
1) The original problem effectively restricts the set of chosen points to be chosen from a set of equally spaced points because it requires the arcs between the chosen points to have integer lengths.
2) The circumference of the circle in the original problem is the sum of the lengths of the individual arcs, or 1k + 2k + 3k = 6k. So if we select a set of points along the circle spaced one unit apart, we will have 6k of them. The original problem calls for choosing 3k points, which is exactly half of them. So our reformulated problem, like the original, calls for selecting exactly half of these points.
Having posed the problem, we now characterize the sets of points that can be chosen. We use a "*" to represent a chosen point and a "-" to represent an unchosen point. We list the points in sequence, using two lines so that diametrically opposing points will be vertically aligned. Here is an example where every other point is chosen:
> *-*-*
> -*-*-
The ">" are just for clarity due to substack's lack for formatting options. Ignoring them, you read the top line from left to right, then the bottom line from left to right, and then go back to the top line.
Condition one says that if a point is diametrically opposite to a chosen point, it must not be chosen. The converse is also true, because we are required to choose exactly half the points. If two unchosen points were diametrically opposed, then there would have to be two chosen points diametrically opposed in order to choose fully half the points.
The simplest solution that meets conditions 1 and 2 is:
> *
> -
This is a circle with one chosen point and one unchosen point, and a single arc of length two. So we have n1 = 0, n2 = 1, and n3 = 0.
We can generate additional solutions by adding another pair of points:
> **
> --
Adding this pair has created a new segment of length 1, and has increased the length of the existing segment from 2 to 3. So we have incremented n1 and n3, while decrementing n2, giving n1 = 1, n2 = 0, and n3 = 1. We will call an insertion that modifies arc counts in this fashion a type A insertion.
We could add a pair of points with the chosen point at the bottom rather than the top:
> *-
> -*
While this looks different from the previous insertion, it's actually the same, with two points forming arcs of length 1 and length 3.
If we were to continue to look at complete patterns, we would never be done because there are an infinite number of them. Instead, we will consider insertions based on the surrounding context, looking at two positions to the right of the insertion point and two positions to the left. This part of the analysis doesn't apply to insertions into a circle with only one chosen point because performing a insertion in that case would change the context, but it applies to all other circles.
Two positions of context to the left and right of the insertion point, with two possibilities for each position, means we have to consider a total of 16 contexts. To keep things simple, I will note that it is not possible for there to be three unchosen points in a row, because that would mean the existence of an arc with a length greater than three. So we can rule out contexts and insertions that have more than two unchosen points in a row. Visually, these will appear as patterns with three or more minus signs in a row. For brevity in what follows, I will refer to such patterns as illegal.
Since I cannot do nice formatting in substack, I will number the patterns from 1 to 16 and write the pattern number to the left of the pattern. Here is the first pattern:
1> ** **
1> -- --
This is illegal.
2> ** *-
2> -- -*
Also illegal.
3> ** -*
3> -- *-
Here there is one legal insertion, giving:
> **--*
> --**-
This adds a length 1 arc to the bottom row and changes the length 2 arc in the top row to a length 3 arc, so this is a type A insertion, meaning that it updates the arc counts as follows:
Increment n1
Decrement n2
Increment n3
4> ** --
4> -- **
No insertion possible because any insertion would result in an illegal pattern.
5> *- **
5> -* --
This is the mirror image of pattern 3, which we have already covered.
5> *- *-
5> -* -*
One possible insertion gives:
> *-**-
> -*--*
This is a type A insertion. So is the other possible insertion, which produces
> *--*-
> -**-*
6> *- -*
6> -* *-
The legal insertion here gives:
> *-*-*
> -*-*-
The length 3 arc in the top row has been replaced by two length 2 arc, and the length 1 arc in the bottom row has been replaced by a length 2 arc. We will call this a type B insertion, meaning that we:
Decrement n1
Add 3 to n2
Decrement n3
7> *- --
7> -* **
Illegal.
The remaining eight possible contexts are the same as the first eight except that the top and bottom rows are swapped, and thus they add nothing new.
All possible sets of points meeting the conditions we specified at the start can be constructed by staring with our initial circle:
> *
> -
and repeatedly inserting additional pairs of points. Let a be the number of type A insertions and B be the number of type B insertions. The initial condition is:
n1 = n3 = 0
n2 = 1
Each type A insertion will decrement n2 and increment n1 and n3, while each type B insertion will decrement n1 and n3, while adding 3 to n2. So
n1 = n3 = a - b
n2 = 1 - a + 3b
Thus we know that n1 will always be equal to n3. Can n1 be equal to n2 as well? If n1 = n2, then
a - b = 1 - a + 3b
or equivalently:
2a - 4b = 1
Because a and b are integers, this is impossible, meaning that n1 cannot equal n2.
the total arc has 1k + 2k + 3k = 6k units of length.
for 2 points to be diametrically opposed then there should be 6k/2 = 3k units apart
now take an arbitrary point, we walk from that point till we cross 3k units of length, now since the arcs are only 1, 2, or 3 units long, we will stop only at 3k, 3k+1, or 3k+2
therefore each point has a point that is 3k, 3k+1, or 3k+2, taken mod 3 it becomes 0,1,2 units apart from it, or normalized to 1,0,-1
suppose we start from an arbitrary point a, and we take a walk, and record the number mod 3 x when we cross 3k, then we walk from the next point, the difference between the number %3 becomes x + (b-a) %3 where b is the new point added to the walk.
if x is 0 then that would mean the already has a diametric point
if is either 1 or -1, we then add (b-a)s to it, since there is only 1k worth of each type of arc, then it cannot all be 0,s we must switch between points of each type at least at some point, and then switch back
this means that the number of +1 and -1s must be the same, and since it is a circle they are rotated, so we have say on one half of, since the only arrangement of 1s and -1s that keep the circle out of 0 is +1 -1 +1 -1 etc, and since a flip must occur at 180 degrees, then that sequence will eventually have to flip too, and the point where the two sequences meet will be a 0
I think this proof proves too much. Consider the following arrangement of arc lengths on a circle divided into 6k=18 unit arcs: 3,1,3,3,1,3,3,1. You can verify (unless I made a mistake), that every endpoint is opposed to a "hole" (absence of endpoint). It seems like your proof would apply to this arrangement, because you do not use the existence of '2' arc lengths or the specific counts of 1,2,3.
I think the error might lie in the fact that when you "walk" in your proof, you do mention that you remove on arc but only *potentially* add a new one (it may be that you go from State-2 to State-1 by removing a 1-arc and not adding), but when you flip 180 degrees, you treat the situation as totally symmetrical.
Only moderately sure of the above, so feel free to correct me!
yes, your observation is correct. My proof was indeed wrong, I think my intuition for symmetry was wrong and I did not use the perimeters of the question
maybe this is a better proof, an arc of 3 can produce a 2 offset if it is the last arc on a walk, an arc of 2 can produce an offset of 1, and 1 can produce an offset of 0
if you have 3k points, you have 3k antipodes, for an antipode to not be diametrically opposed to a point it should fall within an offset, and there is at maximum 2k + 1k offsets, so each point falls exactly within an offset
in this arrangement, each integer on the diameter is either a point or a antipode, and this is too dense to be possible, since 1 arcs must be opposed to 3 pods always (the 2 points of a 1 arc must live within the 2 offset of a 3 arc), so just for the 1s and 3s you need 2k + 2k = 4k spaces.
I hope this one is correct. this has been really fun to think about, so thank you for sharing it ^_^.
I’m interested in whether psychedelic-assisted or adjacent therapies (e.g., psilocybin, LSD, ibogaine, MDMA, or other non-standard treatments) might be appropriate for treating depression and anxiety, and I’m looking for guidance on who to talk to.
Background: I have diagnoses of PTSD, MDD, and GAD. I’m currently stable on medication (Wellbutrin, plus mirtazapine primarily for sleep), with a marked reduction in depressive symptoms and near-elimination of long-standing suicidal ideation since starting Wellbutrin last year. I’ve previously done several years of talk therapy and some EMDR, with modest or unclear benefit. I’m not in crisis.
My interest is partly driven by long-running discussions in rationalist / adjacent circles, and more recently by hearing about ibogaine-assisted therapy (e.g., Rogan’s interview with Rick Perry). My goals would be to reduce or eliminate long-term dependence on antidepressants without simply substituting another dependency. I also have a strong preference—at least initially—for approaches that minimize intense hallucinations or radically altered states of consciousness. This is partly personal temperament and partly religious (I’m an Evangelical Christian), though I recognize this constraint may limit options or need revisiting with professional guidance.
I live in Michigan, am not wealthy, but could travel and take time off work if there’s a compelling option. I’m mainly looking for recommendations on clinicians, clinics, or organizations that help patients navigate these options responsibly. I’m not looking for advice on self-experimentation.
Psilocybin therapy is available in Oregon, but is crazy expensive and not very evidence-based. I expect it will become more available as more states jump on the bandwagon.
I'm a psychologist, do not do psychedelic therapy myself but can give you info. Fluence is an organization that trains people to do psychedelic treatment. I took one Fluence course, on studies of the effectiveness of psychedelic therapy, and it was very well-run. On the other hand, the boundary between responsible psychedelic treatment and stoopit stuff like energy healing, astrology, aromatherapy etc. is just intrinsically thin, and some Fluence courses I looked at seemed a bit too infected with woo for my taste. They have a directory of Fluence trained therapists here: https://directory.fluencetraining.com/?utm_source=chatgpt.com. I would recommend only considering people who have a conventional advanced degree (MD, PhD, MSW, ED) and do not list any new age stuff among the modalities they use.
To learn more about psychedelics, I recommend you read How to Change Your Mind, which is a great read anyhow, then asking AI to find you studies of psychedelic treatment effectiveness in juried journals only. There are definitely solid studies that got positive results for PTSD, depression and anxiety. There are also some decent magazine and NY Times articles about the subject. And Reddit has subs for psychedelics in general and also for most of the common drugs.
There are therapists who have their patients use cannabis in sessions. I'd call cannabis psychedelic-adjacent, & it might be a good place for you to start, since it is not a drug that makes you hallucinate and "trip," and you can easily control how strong the effect is by taking a little at a time and waiting to see what it does before taking more. And cannabis has the advantage of being legal in many places. Ketamine is also legal. I tried it and did not feel like it loosened or expanded my mind in a useful way, just scrambled it. But many people find it very helpful and lots of psychedelic therapists use it in their treatment. Psilocybin, a true psychedelic, is illegal in most of the US, but there are a few places where it is legal or decriminalized. It's not likely a therapist would provide it, though, because of the legal risk. You'd have to buy it yourself (it's not expensive) and bring it to sessions.
MDMA ("ecstacy") is not a psychedelic, but is often used in treatment of PTSD. It brings on a powerful feeling of trust, love and well-being, and in that state people are more able to recall and talk about severe traumas. I believe it is illegal everywhere, though there are probably some research studies where it's given. Check clinicaltrials.gov. (Actually, I was curious and looked, and there are. They're here: https://clinicaltrials.gov/search?intr=MDMA&cond=PTSD)
Thought of one more thing for you. It’s a case study of MDMA, used to help a couple. The man had PTSD resulting from a childhood sexual trauma. His wife knew about the abuse, but had never heard details or seen the distress it caused him. The man’s trauma symptoms had made it hard for the long-married couple to enjoy sex together. So in this couples treatment they both took MDMA, with a therapist there to facilitate. There were I think 2 MDMA sessions, interspersed with a few sessions prior to and during the MDMA session phase. Treatment was very helpful. I found the article, but it’s behind a paywall. To access it you either need to belong to some organization that has access, or use SciHub. It’s a good article, honest and informative, positive result — worth hunting for. Here’s the link: Case Reports J Psychoactive Drugs2019 Apr-Jun;51(2):166-173. doi: 10.1080/02791072.2019.1589028. Epub 2019 Mar 19.
Look into keto diet for managing symptoms along with drugs. There are some published case studies on keto to manage bipolar and schizophrenia. Look up Chris Palmer, he's a Harvard psychiatrist who has led studies on this.
Thank you, I'll check that out. My brother has done keto quite a bit and had some similar mental health issues to me so he could be another good resource for me to bounce things off of.
Forgot to mention, I also have some addictive behaviors that I'd rather not get into here but which I would have similar goals for treatment, in case that would change people's recommendations.
Thanks! I appreciate the writer opening by saying he initially had negative impressions of IFS. I'm somewhat familiar and have mostly negative impressions, so his opening makes me more open to considering his account. I'll read through the sequence with an open mind.
Since we're Bayesian cultists here, I think we'd do well to popularize notation of odds over chance. That is, 1-to-1 as opposed to 50%. It makes significantly easier calculations and is not an eldritch notation no one used. It's especially beneficial for Bayesian analysis, as opened my mind by https://youtu.be/HZGCoVF3YvM. Also it makes "twice the chance" actually comprehensible. Instead of going from 99% to 198% (what even does that mean), it goes from 99:1 to 198:1 (alias 99.48%). As you see, in chance notation it doesn't make sense at all but in odds notation it just follows.
Yes, this is something I’ve been trying to push in the philosophy literature on probability! There are some purposes for which probability is more useful (particularly for calculating expected value of bets) but for many purposes (particularly understanding the strength of evidence) for which odds (or even better, log odds) is more useful.
My reading of 99:1 is, in horse racing terms, for a $1 stake I win $99. So the event is unlikely. To me, 'twice the chance' means something is twice as likely to occur, which means you need to half the odds, since no-one says 99:2. Unless you mean 1:99, for an event which is almost certain? In the case of probability you half the probability of the complement (which is always a well-defined operation).
Since the two systems are mathematically equivalent, and decimals are easier to work with than fractions, maybe it's easier to stick with probability?
Decimals and fractions are each easier to work with in different ways. Decimals are easier to add and subtract, but fractions are easier to multiply and divide. Similarly, probabilities are more useful when you want to calculate expected values, but odds are more useful when you want to update beliefs in light of new evidence. For many purposes, people with different priors can agree on the likelihood ratio of some evidence for a pair of hypotheses, and you can just multiply their prior odds by this same factor to see how their beliefs change, which is much easier than the Bayes theorem application you have to do with probabilities.
I agree. My intuition is that you double your chances of something that has p<0.5 the normal multiplicative way. Doubling 25% (1:3) becomes 50% (1:1), not 2:3.
Additionally, just to write the second sentence above, I had to correct myself several times that 1:3 is not the same as 1/3. Perhaps if I were more familiar with the odds ratio notation, it would be more intuitive. As it stands, it’s just more confusion.
There are a few ambiguities in the ways people use numbers. I'm not a gambler but I always visualise odds as describing a horse race, so 4:1 means that there's 1/5 chance of winning. (If you win the bet, you get your original stake back again!)
Another one that you have to watch for is that percentages change from additive to multiplicative as the numbers get bigger. If someone tells you that X has increased by 350% and it used to cost $100, then almost everyone will answer that the new price is $350. If you say the price increased by 8%, almost everyone (correctly) says that the new price is $108.
I've asked the question many times, it's consistent across age groups and countries (I was for ten years a mathematics lecturer). I normally recommend that percentages are used only for numbers less than 1 (i.e. 100%) - there's rarely any good reason to use them beyond this; just talk about a 3-fold increase or whatever, rather than add two zeros to make the number more impressive and less accurate.
I think it's just because linguistic intuition beats math for 350%.
In common speech, multiplicative "percent of" are far more common than additive "by percent." My phone is at 80%. The fundraiser is at 120%. Linguistically, I prefer the multiplicative interpretation whenever it could possibly make sense. "Increase by 8%" doesn't trigger it, because 8% of a positive quantity is not an increase. Likewise, 100% increase and 200% increase don't trigger 1× and 2×, because 1× is not an increase. But 130% now? My brain just wants to correct it to "obviously you mean 130% of the original."
Isn't all of this just innumeracy or maybe just sloppiness? The one that drives me crazy is "three times more" to mean 3x when it really should be "three times as much" or "two times more" (original x plus 2x increase). Even more peeving is "three times smaller", which I can only guess to be "1/3 the size". Trump just got some grief for saying something like this, "cut prices by 1200%" oslt. While clearly wrong, this just seems typical of people today.
I have a hard time believing people like Gusev are arguing in good faith.
The elephant in the room: in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ - yet the gaps persist.
Gusev claims IQ is "socially transmitted" - unlike height, but like religion, smoking or language. But we actually understand the transmission mechanisms for these. What's the proposed mechanism for IQ?
Consider Babur the Conqueror: he culturally copied state-of-the-art firearm technology and used it to capture the Delhi Sultanate. Thanks to guns, former peasants with months of training could defeat elite soldiers with lifetime career with sword or bow. Technology transfer worked - because it was replicable, independent of the innate traits of the replicators.
But if Babur copied SOTA firearms and Pakistan has now nuclear weapons, why can't it copy Harvard? You can replicate the curriculum, the syllabi, the lectures, but you don't get Harvard. Just like copying Facebook's source code and servers won't give you Facebook. Facebook needs its users, you get only an empty shell otherwise. Harvard isn't a set of classes; it's a filter. It selects from the top few percent of cognitive ability and concentrates them. If you filled it with the second quartile of society like rulers creating new armies, the outcomes would vanish.
And yet, if IQ were truly socially transmitted like smoking, we'd see people seriously debating whether raising a child with Lojban as a first language boosts abstract reasoning and creativity, or whether Esperanto or Haskell is a wasted developmental window. But no one does. The interventions that should matter if IQ were cultural are ignored. Gusev has nothing to teach our hypothetical Babur the Educator.
Cloning humans is illegal and hard. But cloning a genius's environment? Homeschooling is still legal in many places. We know a lot about the inputs that correlate with exceptional intellectual development. So why aren't we producing new Newtons or Galois-level figures at scale? If environment is the bottleneck, we should be able to brute-force it. But we're not. Who are the ones most vocal against human cloning? Anti-hereditarians.
When we find that plant growth is affected by its environment, we get it to reducible factors such as soil structure, temperature, nutrients etc. most of which are not even produced by plants themselves. But in "IQ is socially transmitted" it's never any actual details about it but mostly "good schools create good outcomes" which you cannot copy. Or you find non-causal correlations like lead correlated to low IQ, this actually tells you more than high IQ people (independent of cause of their high IQ) tend to avoid lead more than low-IQ people. In case of lead, it's not even environmental finding -- they measured lead content in blood, blood isn't part of environment anymore. It's very likely that organisms living in same environment have different prospensity of eating lead from environment.
Gusev says that height, unlike IQ is measured with ruler and in that case we know what we are measuring, but then we could also measure jumping and sprinting ability. Does Gusev go and say "well, genetics has disproved/confirmed stereotype about West Africans being good spriners"? No? Because it's non-PC. His behaviour is better explained that he says only PC things and "we can't mesaure IQ with a ruler" is just argument-as-soldier.
Gusev also heavily implies that anything which is not "direct genetic effect" is socially transmitted; well there is homosexuality which is mostly non-genetic, but does it mean that it's non-biological?
There's also Markel et al 2025 paper which estimates between-sibling heritability of IQ at 75%, Gusev doesn't talk about it.
Ah, and about the Cuartetera clones: obviously it's folk science. You cannot claim its genetically determined before there will be a between-sibling study of polo playing ability of horses.
Provably false! Buck v. Bell was decided in 1927, at which time it was legal to forcibly sterilize people the government thought were feebleminded. Not that long ago! It was never overturned, though other developments make it a dead and embarrassing law.
Hey, look, a teeny-winsy 0.01% of population was forcibly sterilized! Totally a century of eugenics, because every animal breeder knows removing 0.01% of animals from breeding stock is totally good to make changes.
...
This was long enough ago so South Korea starting from worse than African level of poverty became one of leading countries in tech.
I honestly don't understand the mental gymnastics blank-slaters perform to convince themselves of their nonsense. Genes influence literally everything in the human body but somehow our single most adaptive trait - intelligence - is magically immune? "Genes only code for the things my ideology says they should!"
It's just the Creation Science of the Left, there's no real sense in trying to understand it as anything but a religious doctrine. I'm positive that future historians will regard it the same way we regard Soviet Lysenkoism.
I'll add that the most common "extreme" position on IQ that I've seen in the wild is not "IQ is socially transmitted" or "IQ is mostly non-heritable." Far from it. Rather, far more common is simply the view is "IQ is bullshit, actually."[1] I think this position is usually overstated--sometimes comically so--but at the same time, I think it's important pushback against lots of people who are way, WAY too credulous about IQ scores without really understanding anything about them.
As a general rule, we should hear any claim of the form "IQ correlates with Attribute X or Life Outcome Y" as carrying a fairly high burden of proof. A burden which rests squarely on the person making the claim. A priori, there is no reason to believe that somebody's responses to a specialized test will tell you anything about anything *except* the skills covered by that test. To my understanding, part of the trick of IQ tests is that they actually cover quite a lot of different cognitive skills and part of the trick is that the developed world does a fair amount of question-begging in this regard, using things-roughly-like-IQ-tests to determine peoples' eligibility for quite a number of different socioeconomic opportunities. But also that IQ researchers HAVE gone out and met the burden of proof in at least some cases.[2] So I don't think the claims that IQ is purely bullshit can stand without a long list of elaborations and caveats. But I do think skepticism is a healthier starting point than credulity.
[1] This usually comes with a sometimes-unspoken corollary of the form "anyone who does take IQ seriously is probably engaging in motivated reasoning, likely of the racist variety." I don't really share this view, but given the extremely spotty and sordid history of intelligence testing, I can hardly blame these people for being suspicious.
[2] The problems, in my view, stem from overzealous people (most of whom, I think, are *not* IQ researchers) being very quick to do things like generalize those results in dubious ways, and assume causal models that have *not* been proven. To the best of my (limited) knowledge, most people currently working in the field are doing good, responsible science; it's just often misreported and misused as responsible science so often is.
> far more common is simply the view is "IQ is bullshit, actually."
Great point!
There is also a medium perspective, popular among contrarians, promoted by Nassim Taleb, which is... if we put it bluntly... "IQ lower than Taleb's is real and important, and it separates stupid people from smart investors; but IQ higher than Taleb's is bullshit, and it only means an autistic ability to check the right boxes in an IQ test".
I don't have much of a dog in this fight, so I'm not going to even try to prove anything W.R.T IQ and heredity. But I see you making a whole truckload of common mistakes and poor assumptions that tend to be pretty corrosive to having useful discussions about this topic. So I'm going to point those out in the hopes that this will enable you to get better answers from people who have them.
" in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ"
Have they? I don't know what your threshold is for "staunchly anti-hereditarian," but this doesn't match my experience at all. I haven't taken a survey, but every time I can remember reading *anyone's* opinion on IQ, it's been "well, obviously it's some mix of heredity and environment, but we don't know what mix or how important various pieces are." This is true of people I've listed to on the left, on the right and in the center. I don't think any serious person believes that IQ is 0% hereditary; the difference comes in that some people believe that the hereditary aspects of it are difficult to identify, assess and change, so worrying about them is a waste of time. So maybe more precision about what you actually believe W.R.T heredity and IQ would make for a more useful discussion.
"But if Babur copied SOTA firearms and Pakistan has now nuclear weapons, why can't it copy Harvard? "
This is a question that Proves Too Much. I think pretty much every non-Ivy-League university in the U.S. would love to be "the next Harvard," so why can't any of them copy Harvard either? Unlike Pakistan, they teach in the same language, can hire the same faculty as Harvard with less hassle (both in terms of immigration and convincing people to relocate), have access to the same pool of potential students. Why can't any of them do it either?
The really, really obvious answer is "because Harvard already exists." It's not even clear to me what it would *mean* to be the next Harvard in a world that already has one. I went to a mid-tier state school in the U.S. I have no clue whatsoever if I got a better or worse *education* than I would have got at Harvard[1]. But I'm am *quite certain* that the *credential* I would have earned if I could have attended Harvard: the primary value Harvard offers is *being Harvard.* That is being a school with a 400 year history and universal name recognition as One of The Best and Most Exclusive Schools Out There. Replicating its curriculum won't reproduce that unless you can also replicate its faculty, its student body AND its 400-year-history, because at the end of the day, everybody will still know that Harvard is Harvard and your other school is not. So again, to make for a productive discussion, I suggest you try to write down *exactly what features* of Harvard it should be possible to replicate elsewhere, and narrow your focus to those. If you do this carefully and honestly, I suspect you'll probably find at least a few schools that *have* managed to replicate Harvard in various ways (though I have no strong prior for which ones).
"And yet, if IQ were truly socially transmitted like smoking, we'd see people seriously debating whether raising a child with Lojban as a first language boosts abstract reasoning and creativity, or whether Esperanto or Haskell is a wasted developmental window. But no one does."
Obvious answer: nobody debates this because this is the kind of utterly unhinged thing that only people who consider their children to be science experiments would ever even consider. There aren't enough of those people making enough noise to be visible. If instead of insane ideas like "raising your child to have a native language shared by almost none of their peers with more reasonable early-childhood interventions like different preschools, enrichment programs, and modes/amounts of parental engagement (i.e. things like reading to your kids), I think you will find LOADS of people debating exactly that. I cannot emphasize enough how many people there are that *badly* want to know which early-childhood interventions and activities will give their kids a leg up. If you answer this question convincingly[2] you can essentially use it to print money.
"So why aren't we producing new Newtons or Galois-level figures at scale? "
Similar to the Harvard example, I have to ask how exactly do you know we're not? How are you judging whether a person is a Newton or Galois-level figure? Calculus and Newton's laws and Galois theory have already been discovered, developed and written down. Anyone born in to today's world has far, far less low-hanging fruit to use to prove their aptitude. By default, I assume the number of Newton-level talents running around the modern world is a lot more than zero. We just haven't recognized them as such because they've spent careers working on various difficult science and engineering problems that don't make them world-famous. Or they take their talents in some other direction and pursue art or education or just get really into their hobbies.
I'm sure this will sound like a cop-out, but consider that the modern world is much, MUCH more technologically complex than the world of Newton's day. To keep civilization running at all requires a huge number of intelligent doctors and engineers and professors just doing maintenance-level work of medicine and design and teaching that doesn't produce huge, highly-visible improvements. And more objective attempts to measure the average IQ *do* find it to be increasing over time[3]. At risk of sounding like a broken record, consider trying to nail down a narrower, more objective standard than "how many Newton-like figures do I know of" and asking about *that*.
"But in "IQ is socially transmitted" it's never any actual details about it but mostly "good schools create good outcomes" which you cannot copy. Or you find non-causal correlations like lead correlated to low IQ..."
As far as I understand, not being able to isolate which *specific* pieces are responsible for variations in intelligence is an issue in BOTH the environmental and genetic explanations. Isolating which genes are responsible for high IQ has proven no easier than isolating which environmental factors are.
From my non-expert viewpoint, I can't understand why anyone would imagine either of these things to be easy. Human brains are extremely complicated. Human development is extremely complicated. By default, I would expect the outcomes of having and raising a child to depend in pretty minute detail on every part of the processes, and isolating important factors beyond the blindingly-obvious one (e.g. food, shelter, sleep, basic socialization) would be next-to-impossible. Everything in biology is a complicated mix of hacked-together processes that work imperfectly, are replicated imperfectly from organism-to-organism, and only produce "good" results when you look at zoomed-out population averages. Why should intelligence be any different?
[1] To be clear: my prior is "worse," but it's not a strong prior (maybe 60-70%). Too much depends on the individual professors and classes and classmates and other intangibles.
[2] Note: even if it's not *universally* convincing, as long as its convincing to a certain segment of reasonably well-off new parents.
Which to my non-expert eye seems like exactly what one might expect if intelligence had a significant social-environmental component, and advanced societies were getting slowly but steadily better at fostering it.
Flynn effect is real but a part of it probably stems from biological factors rather than cultural, better nutrition and elimination of many infections. Vaccines benefit almost everyone, even those who don't have cultural understanding of vaccines and the rich have every incentive to vaccinate the poor. Brains also grew in size, not only height.
>Obvious answer: nobody debates this because this is the kind of utterly unhinged thing that only people who consider their children to be science experiments would ever even consider.
Is that much different from tiger parenting? Richard Williams, father of two tennis champions, got what he wanted. Probably he would do it if he thought it would be effective. It's just a language, and not neccesarily must be only one, can be biligual. We don't even see many discussions about learning which of living languages would boost creativity (even not as distinct at Lojban, natural languages have wildly different grammars and phonologies).
Ah, I forgot to mention baby einstein toys. Anti-hereditarians don't discuss which ones are better. Maybe you should just buy same ones Gauss's mother bought for him. Oops, Gauss did not have "baby einstein" toys but got smarter than those us moderns with them.
>How are you judging whether a person is a Newton or Galois-level figure?
talk to a modern 17 yo who wants to apply to math faculty, and ask them if they can explain what Galois accomplished. They would probably answer poorly even if they practice mathematics a lot. Compare that in terms of life expectancy at birth, we are superhumans to humans just from 300 years ago.
>reasonable early-childhood interventions like different preschools, enrichment programs
These all been tried and most of their long-term effect is small (it's more selection effects: smart parents can afford move expensive preschools than causal effect), moreover if we want progress we need to study novelty and go better than existing solutions. Someone will want to get ahead of others. Richard Williams did. Oh, left-wing people do not like people trying this on their own, without explicit permission from authorities.
>Why should intelligence be any different?
Yes, there is some developmental noise not explained by genes nor environment. But
Genus Homo just had tumultaneous evolution thanks in regards of intelligence, this would be unlikely it heritability was low. (One can speculate that heritability of intelligence was higher in past and after humanity reached "final form" it's low but it needs to be proven). Moreover, correlation between IQs of MZ twins is high so it's reproducible enough. Anti-hereditarians claim that's cultural causes causing it but cannot explain a single one and preventing us from cloning and figuring out if it's indeed replicable.
>The really, really obvious answer is "because Harvard already exists."
You missed the point. The point is that Harvard is a selection machine which takes high ability people, unlike example with gun culture which took below average peasants and made them good soldiers. Quite a lot of countries would have wanted a copy of Harvard without its prestige and regalia, it just needs to work taking average unselected peasants and producing good outputs. If they did that, then Pakistan wouldn't need to copy nuclear bomb from somewhere else, they would invent it themselves and also invent many other things than nobody had before.
> and isolating important factors beyond the blindingly-obvious one
You know, the militaries isolated important factors, the Babur example above.
> I cannot emphasize enough how many people there are that *badly* want to know which early-childhood interventions and activities will give their kids a leg up. If you answer this question convincingly[2] you can essentially use it to print money.
I think the steelman of Ferien's point here would be: even the "environment matters 4x more than twin studies tell us" people like Gusev will:
A) Have zero interventions they can suggest to move that supposedly 4x stronger needle.
B) Tell you that "family culture" is incredibly important to outcomes, and is why we have to do familial GWAS with lineage controls to find the supposedly much-smaller "true" heritability figures for IQ / educational attainment.
But if familial culture matters that much, doesn't that imply that we should behave as the most Greg-Clark-and-hereditarian-pilled would and optimize not just on our partner's intelligence / EA, but on their *lineage* intelligence and educational attainment?
That seems to ME to be the only concrete intervention suggested by the new methods that supposedly tell us environment matters 4x more than we thought - totally ignore environment, and max hard on lineage genotype via phenotype.
So it's damning two different ways - not just in not being able to point to any environmental interventions that matter, despite numerous adoption studies of poorer kids being raised in higher SES homes and twin studies and other studies looking at lead, education, early reading, and much more, but in ALSO pointing out that "familial culture" matters so much you should behave as a stricter than usual hereditarian if you wanted the best outcomes.
At least one out of two: environmental interventions, or cloning should work (hmmm but one could argue that both are needed), so your statement seems to omit (maybe intentionally?) something.
"The elephant in the room: in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ - yet the gaps persist."
Others have replied to you on this, but the past hundred years would be 1925-2025. So the bulk of the twentieth century was anti-hereditarian about IQ? The century of eugenics? The century of "sterilise the idiots so they don't reproduce"? The century of Social Darwinism? That century?
Eugenicists won some small victories and then lost everything. These sterilizations were significant to individuals but barely relevant in grand scheme of things (commie dictatorships killed much more people in peacetime than there were forced sterilizations). You speak like affirmative action, educatiotional interventions, and physical attacks on Arthur Jensen and others did not happen.
Affirmative action in the US came in the 1960s, not the 1920s, when much of the country was still under segregation and Jim Crow. Europe was ruling nearly all of Africa in 1925.
A full century takes you back wayyyyy too far. Try 60-65 years.
I am biased -- I don't live in the US, in my country it started earlier. But you're correct. Thought would it help? Poster darwin referenced the Bell Curve from 1994. Or he could say "in %current year%, police still killed %large_number% blacks for just being black, this proves systemic racism"
"in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ"
IQ was invented less than 120 years ago. Buck v. Bell dates to 98 years ago. "Anti-hereditarianism about IQ" as a popular attitude originates from the aftermath of WW2. Your timeline is all messed up.
>The elephant in the room: in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ - yet the gaps persist.
Sorry, Germany in the 1940s was staunchly anti-heredeterian?
'The Bell Curve' was a best seller in *1994,* for Glob's sake.
This is such an astonishingly ahistorical statement that it's hard to justify reading any further.
If you'd said 'the last quarter-century', you'd *still* be wrong, but you'd at least have a mildly plausible-sounding position to premise your argument from.
>Germany in the 1940s was staunchly anti-heredeterian?
Well they ignored superior Jewish intelligence so ... kind of?
>'The Bell Curve' was a best seller in *1994,* for Glob's sake.
It also got the author blackballed from polite academic society for life. I don't think bestseller == cultural approval. Deep Throat was a top-grossing movie too, that doesn't mean that porn is mainstream.
OP is directionally correct here, even if he's wrong about the specifics. The post-segregation consensus in the US has been very strongly anti-hereditarian. There is only one view that will get you shunned in this debate and it's not the one that claims we're all the same.
As usual, you pick one point which you consider weakest and ignore the rest.
Checked 1st corner of the room: no elephant.
Checked corners 2-4: no elephant
Touched elephant tail: that's not elephant tusk
Touched elephant elephant trunk: that's not elephant ear.
Touched elephant leg: that's not elephant tail
There is no elephant in the room!
>If you'd said 'the last quarter-century', you'd *still* be wrong
If you want discussion this low... Obama was elected twice, for Glob's sake. Anyone younger than Obama cannot complain about racism seriously. Such discussion not going to be useful to anyone, right?
Yes, but the claim was a completely blanket 'EVERY government and elite'.
Bell Curve was controversial, but it represented the views of many elites and politicians. Not a majority, I'd agree to that, but still an influential faction.
The anti-hereditarian impulse was a reaction to Nazi Germany's eugenics programmes which then blossomed rankly into the Holocaust.
After they emptied a septic tank of sewage into the pool, nobody wanted to be seen arguing "intelligence is hereditary, some people are natural inferiors, this is how we should deal with the inferior".
Exact cause why anti-hereditarians won in politics isn't important for my argument. Their win in politics did not translate in win for desired outcomes.
Maybe some elites were. It is above lizardman constant? When GWAS returns 10% influence on a triat, you use words like "teensy" or "minuscule". But 4% of politicians is "influential"
"In virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ."
This historical claim is a likely sign of overwhelmingly motivated reasoning. Mind-boggling for you to say this about the 1920s as European fascism was starting its meteoric rise, scientific racism was still the dominant modality, and non-Europeans were explicitly blocked from immigrating to America. Federal anti-marijuana legislation, a deeply racialized policy, didn't even happen until 1937.
It's not pedantry when his poor choice of words totally undermined the evidentiary basis of his argument. I try not to nitpick but I have no obligation to mentally steelman everyone I read.
Desegregation and equal rights did not even START to be consistently enforced in America until after 1960. 65 years of racial egalitarianism is not even CLOSE to 100 years. The latter period is 50 percent longer!
When were American black folks finally fully free to purchase homes in areas with outstanding property value growth prospecrs? 1975, 1980? Exactly when the neoliberal Carterites past the torch to the Reaganites, who completed the job of devastating unions. Black Americans NEVER enjoyed the combination of strong private unions and access to prime property markets that allowed white families to build generational wealth from 1945-1980.
“If the ghetto is so bad, why not just move out?” It's not a fair question in isolation. We should also ask, “How is it that your people got packed into overcrowded, underpoliced, economically mediocre districts in the first place?” Not by choice!
“The breakup of the black family plus LBJ’s welfare programs led to black poverty” is one of the biggest lies ever told. California presents a perfect example of how even outside of the Deep South, black families were locked out of enjoying the full fruits of the postwar economic boon. In Southern California, hard-working, married black fathers were deliberately locked out of decent jobs and homeownership in 1945. They were not allowed even playing field to work their way out of poverty or buy homes in economically vibrant communities. By 1965, the degradation of enforced poverty had taken a huge toll on family cohesion and the moral authority of conservative black pastors and elders. LBJ’s public housing project initiative came WELL into the process of familial disintegration.
Cash welfare was introduced in the 1930s and deficit hawks barely complained when it was largely received by whites. It wasn't until the 60s, when more blacks started enrolling than whites, that the hawks magically discovered that cash welfare was “destroying” families.
Social assistance should generally be dispensed in other ways besides wasteful cash transfers. But “welfare destroys families” is not only an exaggeration, it's precisely backwards when it comes to the American black community. Families destroyed by policies of segregation and ghettoization sought welfare relief as a last resort.
This analysis was mainly cribbed from Mike Davis’s legendary history of Los Angeles, “City of Quartz.” Written too early for him to integrate the facts later uncovered about Deep State complicity in the international cocaine trade, which pummeled a South Los Angeles that was already down for the count.
>Sexual revolution destroyed families, not segregation.
Female participation in the economy and relaxation of divorce laws destroyed families. Previously, women were unable to support themselves and people were unable to end marriages. You could see the same effect if the sexual revolution hadn't happened.
Nazis were actually staunchly anti IQ tests for obvious reasons. Many non-Europeans from Latin America and the Philippines were allowed to immigrate to America in the interwar period. The anti-asian immigration restriction had obviously nothing to do with IQ. I have no idea why you are even bringing up marijuana.
Your dubious historical claims are a likely sign of overwhelmingly motivated reasoning.
Did I say limit myself to Nazi Germany? I said “European fascism,” which had substantial political influence into the 1970s. Other far right regimes had mixed feelings about the Ashkenazi and how to relate them to whiteness.
You can't play this game where you neatly segregate 20th century intelligence research and technocratic credentialism from white chauvinism and white identity politics. There was far too much overlap between these categories. There's just no way to justify your implication that most governments have been colorblind since the 1920s. (Not in PRACTICE, whatever the text of their constitutions may have been).
Were the other fascist movements even that racist? Would not the US be more white supremacist through to the civil rights era than Spain. Were not the European democracies with their colonies not the epitome of white (or Anglo or French) racism. Spanish fascism seems clericalist and nationalist. Jews escaped Nazi germany largely through Spain, after all.
The title is Cornish for Stone Island. The basic plot is that the protagonist is alone on an otherwise deserted Cornish island in order to make observations of a rare plant.
In terms of genre, calling it a horror movie might give the wrong impression. It is more of a ghost story, sharing themes (but not exact plot) with some traditional Cornish ghost stories, such as The Mermaid of Zennor.
As the protagonist is alone for much of the movie, there are long stretches without dialog. The movie is heavy on landscape photography: I might compare it to Peter Weir’s Picnic at Hanging Rock (1975), with the proviso that the Cornish landscape offers very different cinematic opportunities from Australia.
I think one of the ideas implicit in the movie is that the Cornish landscape is overlaid with layers of time. Standing stones that go right back to the Iron Age are just there, along with the disused mine workings of the tin industry. So as the camera scans o er the landscape, we are also seeing traces of human activity going back literally thousands of years … which is suitable material for a ghosts story.
I did not find the screenplay compelling. But it was visually stunning and I would give this director another chance. Mark Jenkins. He has a science fiction film called Rose of Nevada that has yet to be released for home viewing.
P.S. The location where the movie is set is somewhat fictional (although supposed to be in Cornwall somewhere).I will be vague as to how old some of the archaeology is the movie is supposed to be as it doesn’t correspond to anywhere real. The real places it resembles range from Iron Age to Neolithic.
I don't know who they think this appeals to. They're not going to win over Alex Jones, nor his obese QAnoner audience. All it does is alienate people in the moderate center who might be tempted to vote Democrat because it's the party of serious people who talk about serious things like housing policy and healthcare instead of pedo rings and seed oil.
It's like how boomer conservatives responded to Democratic accusations of racism with "the Dems are the REAL racists, the Democrats are the party of Jim Crow." It didn't improve the GOP's performance with black voters at all, alienated the Republican base, and made conservatives look ridiculous.
If the Dems would move to the center of guns, trans, feminism, and affirmative action while hammering Trump on abortion and the cost of living, they'd have it in the bag. Instead they want to engage in a (futile) competition with the GOP to be the party of prole envy and conspiracy theories.
>They're not going to win over Alex Jones, nor his obese QAnoner audience.
That tweet doesn't at all seem intended to. It's pointing out Jones's hypocrisy, not trying to recruit him.
>All it does is alienate people in the moderate center who might be tempted to vote Democrat because it's the party of serious people who talk about serious things like housing policy and healthcare instead of pedo rings and seed oil.
I regret to inform you that serious people in the moderate center actually are talking about the government deploying masked thugs to snatch people off the street. Because that's what ICE is doing.
>If the Dems would move to the center of guns, trans, feminism, and affirmative action while hammering Trump on abortion and the cost of living,
People keep saying this, but I don't think it's true. Biden made his presidency about affordability, the Build Back Better bill was the centerpiece of his presidency, he presided over a steadily improving economy, and it didn't budge the needle on the "vibecession."
(And also, Democrat attempts to shift the conversation to kitchen-table issues very often come off as kind of lame or insincere. "Trump wants to have illegal immigrants tortured in foreign prisons, but that won't lower your health insurance premiums" doesn't make you sound like you're focused on the real issues, it makes you seem bizarrely uncaring.)
This is nothing new. Matt Christman was one of the most influential leftists on Patreon for years, and he said at one point "QAnon is 50% right. They have two claims, that the elites are a bunch of criminals and pedophiles, and that Trump is going to save us. The first half is true."
Your dismissive attitude towards proletarian frustration shows that you are locked in the past, intellectually. This is the new political environment and it's not going back to the way it was. Apparently you didn't learn any lessons from the financial crisis of 2008 and the corporate bailouts, which totally transformed the psyche of your average American.
Calling working people idiots and telling them to shut up and enjoy their indoor plumbing and smartphones is political suicide. The absolute worst thing the Democrats could do would be to tack to the center on class. This a big part of why H Clinton and Harris were doomed to irrelevance, although their insane identity politics signaling hurt them as well.
Telling people to shut up and enjoy their lives is a guaranteed way to make them your political enemies. One's spouse or parent can get away with that, MAYBE, but not some smug, self-satisfied elite politician.
There's a certain kind of smug, American optimism that is inherently coercive. "Everything's fine, quit complaining, shut up and do what you're told." I call this Pinkerism and unfortunately, Scott is not entirely immune to this impulse. I guess when you're doing great in life, the benefits of great change do not obviously outweigh the potential costs.
I suspect that understanding the real phenomenon you are observing may be difficult without examining the process that selected leftist takes for your X timeline.
The mob bays for blood and to tell them the QAnon thing that's all the rage is bullshit is to make yourself their next target. And this is bipartisan too, combining the schizo lunacy of Pizzagate and the rabid fury of the MeToo movement, so I don't blame anyone for trying to appease them and hoping it runs its course before it turns on them.
The purpose of the ritonavir in the combination is not primarily as an antiviral, but rather to reduce metabolism of the nirmatrelvir by competitively inhibiting liver enzymes. As a result, Paxlovid has a LOT of potentially dangerous drug interactions to watch out for (similar to the list for grapefruit juice.)
Other than that one big asterisk -- which mostly matters for people already on multiple medications for other issues -- I am not aware of any significant risks from Paxlovid.
It’s known that a fairly large percent of people relapse after initially feeling recovered post Paxlovid, but my understanding is that their relapse is usually a mild version of the illness, and they still end up better off than if they had not taken it.
Paxlovid apparently lowers the probability you will die or be hospitalized from covid by quite a bit, so it makes sense to take mainly if you are at relatively high risk of one of those things happening. It would probably be silly to give to a healthy 25-year-old, but wise to give to an obese 40 year old or any 60 year old. I don't know if it decreases symptoms in most cases, though--mainly it's trying to avoid you getting super sick in the second week of the infection.
The main thing to know about it as a patient is that it only does you any good if you take it within a few days of getting sick--if you wait until you're really ill from covid a week into the infection, Paxlovid doesn't do much good. Which means if you're exposed to covid and want to be on it, you want to talk to a doctor right away--as soon as the home test pops positive, ideally. It clashes with some medicines including (I think) statins, so you have to stop taking them while you're on it.
Toddlers have six-packs on their backs, right? And gloves kind of seamlessly flow into the sleeves of robes?
(By the way, this painting depicts Santa Claus resurrecting three toddlers who were chopped up and pickled in brine; I am not making this up, Merry Christmas)
Battleships were eclipsed by aircraft carriers in World War II. I don't understand how modernizing the concept can make them serve any role not currently filled with existing ships. The only thing I can think of is that firing a large enough rock (or equivalent) would be pretty hard to shoot down, unlike missiles.
The US Navy is almost entirely oriented towards the idea of the aircraft carrier, with most other surface ships serving just to escort the carriers around. But in a drone world does the aircraft carrier make that much sense? If you take the humans out of the aircraft and make them smaller then you don't need such a huge ship to operate them, you can have smaller ships operating fleets of more specialised, semi-expendable drones, and still capable of launching missiles and having big guns too. You can launch drones off a big rail, and land them in the sea, and you don't need all that deck space just for aircraft operations.
In a world where carrying aircraft is no longer a specialty worth basing your ship's entire design around, it might make sense to have capital ships which are big multi-capability platforms with drones plus missiles plus guns plus anti-submarine capabilities. Battleship seems like as good a designation as any.
Trump et al are even less likely than the median politician to have any idea what they are talking about wrt something like this, or to care that they're speaking nonsense.
I could imagine an offshore artillery platform being useful for delivering firepower against some very low capability opposition. As in, if you just want to put bignum tons of high explosive on target, and the other guy can't really shoot back, then shells are much cheaper than missiles.
Not sure what this low capability (but coastal) opposition that we want to pound is though. Even the houthis have missiles. Also this floating artillery platform probably doesn't need to be armored.
Even in that situation, I think guided missiles usually come off better, because you don't *need* bignum tons of high explosive to be sure of hitting your target.
The Zumwalts were intended to do shore bombardment before they got cancelled, and the vision there was that they'd fire guided shells, which gives you some of the advantages of both - cheaper than a rocket, but still very accurate. But you don't need a battleship-sized platform to do that.
- A single M982 howitzer shell costs ~$100k (reference point)
JASSM range is ~25x that of M982. If we favorably assume that the shells are 1:1 substitutes for missiles, and that the shells' cost only scales with the square root of range, the breakeven for ordnance cost savings offsetting the ship acquisition cost would be 13,000 shots.
The total number of JASSMs expected to ever be built is 10,000. The entire inventory would be consumed by ~220 Rapid Dragon sorties (which is maybe one year's worth of a single C-17's operational cadence), so the share of aircraft acq cost is negligible.
And it's implausible that the shells could be perfect substitutes for the missiles, even with linear-in-range cost increases (at which point the missiles would win on unit cost alone).
Calling them "battleships" doesn't make them all that similar to WWII battleships. They'd really end up being not that different from a bigger Arleigh Burke destroyer. I'd be happy to defer to people who have more knowledge than I do, but I don't really see the point in the US Navy having surface combatants bigger than an Arleigh Burke. The previous plan was to get rid of the cruisers and destroyers and replace them with the DDG(X). This battleship would be a bit bigger, but if I could choose between 20 of these "battleships" and 75 DDG(X)s, I'd be inclined to pick the latter.
If technology progresses in such a way that it becomes much, much easier to shoot down drones and missiles that could theoretically reopen the combat niche for the battleships. Of course even then the only real use case for them would be a conventional naval war with China.
Also boy howdy, "We're not producing enough ships, better make a brand new ship design using tech we don't even have a good prototype for with an insane timeline!" Yeah, that sounds exactly like a fix for not producing enough ships.
AIUI, WWII-era carriers dominated battleships in range, precision, & air defense.
Looking at the proposed kit for the new battleships, IF they were all successfully deployed (spoiler alert: they won't be) then at least range & precision could be comparable and at a fraction of the per-shot cost.
Carrier air groups don't put that much iron on target compared to the bomber fleet, so apart from a blue water conflict with a near-peer adversary it's plausible that the Navy is overinvested in air power.
Full disclosure: former USAF officer, so very familiar with aircraft operations, much less so with surface warfare.
according to bean (of Naval Gazing), battleships were not fully obsoleted until the 1950s. Carriers dominated in the Pacific war, but battleships played a key role in the mediterranean and the north sea.
Sir Humphrey, in 1942 the Indian Ocean Fleet of the RN was on a collision course with 2 japanese carriers. If admiral Somerville would have kept the course a few more hours we would have witnessed a night encounter between the 2 fleets and the japanese carriers might have ended up in the gunnery range of HMS Warspite. We know how that would have ended because Warspite had the best plot armor in WW2.
I'm interested in the topic and on the whole the podcast was informative, but it would have been better at half the length. He tends to bludgeon the point he is making into submission, at times I was going "yes, Dan, I got it, you don't have to go on and on and on and on and ON about it!" That, and the fifty repetitions of "I just don't get it" and the annoying quirks of his delivery (rowly-growly voice when he's making a point or being all SURPRISED about THIS GUY DID THIS?, then the trick of speeding up pace and rising in volume - just speak naturally, and if this *is* natural for him, then iron out the whisper-whisper-growl-get louder-louder-LOUDER!!!! punchline! stuff).
I would be interested in part IV, but I have to say that I did fast-forward a lot because of said "and I'm saying the same thing eighty times here just so you get it", "gosh why did they do this like that I just don't understand" and "okay now I'm going to tell you the Really Good Advice Alexander ignored (goes on tangent for about twenty minutes, or it felt like it subjectively, before finally telling us what the Really Good Advice was)".
On the positive side, it did make me want to rewatch the Porus episodes where Al is stomping the Indian armies (and everybody else) into the ground because he *is* just that good 😁
Check out the historian Dr. Bret C. Devereaux at his blog A Collection of Unmitigated Pedantry (ACOUP). My pitch for him is simple: on the rank of reading that I've done as an adult which has most broadened and enriched my understanding about the past, he's in the top two. Maybe even #1.
His recently concluded series was about peasant farmers; how they lived, worked, and died. https://acoup.blog/2025/07/11/collections-life-work-death-and-the-peasant-part-i-households/ One point I want to add: I think his explanation of how farming village plots were highly balkanized could be important to understanding why technological progress was so slow in the medieval era. Yes, it decatastrophized individual household risk, which is good. But if you're only plowing one furrow at a time you're not leaping to upgrade from wood to iron to steel ploughboards because other things dominate your time, like traveling between furrows.
I was introduced to ACOUP through Devereaux's exploration of how the pre-modern world made iron and steel. (He does focus on Europe but he also touches on the other processes; Indian and Chinese and Japanese). This one in particular accelerated my journey of understanding of energy (as in, literally watts) as the first among equals of all industrial inputs. https://acoup.blog/2020/09/18/collections-iron-how-did-they-make-it-part-i-mining/
The CBS report on CECOT that was shelved from broadcast in the US was streamed in Canada, and can be found online. After watching it, it's clear that Bari Weiss was right in deeming that it added nothing that other outlets like PBS hadn't already published, and the brouhaha over her decision seems to be basically like that story of the college student who wrote a poor essay and then made political hay of her TA's transgenderism/dysphoria.
I think it's quite a bit more obvious that this is based on her political views and not her desire to publish the most relevant or novel news. Her extensive interview with Erika Kirk is more proof of that.
Since when is the standard that every story on 60 Minutes contain a very specific type of original information? The report has new, informative, affecting interviews from CECOT torture victims and other new reporting. I could’ve learned all the information about sherpas and Everest in another piece 60 Minutes aired this Sunday from other sources, but in fact I hadn’t, and found it a fascinating and well shot piece, just as the CECOT piece was. It was an excuse to delay and possibly kill a piece she didn’t like for political reasons, just like the other excuses in her note—including her false claim that the Administration agrees these men have a right to judicial review. They argued the opposite and lost all the way to the Supreme Court.
Probably around when Bari Weiss became editor-in-chief, if I had to guess. If she holds programs she's now in charge of to a higher standard than they previously were, good for her!
> Last month many outlets, most notably The New York Times, exposed the horrific conditions at CECOT. Our story presents more of these powerful testimonies—and putting those accounts into the public record is valuable in and of itself. But if we’re going to run another story about a topic that has by now been much-covered we need to advance it.
"Last month many outlets, most notably The New York Times, exposed the horrific conditions at CECOT. Our story presents more of these powerful testimonies—and putting those accounts into the public record is valuable in and of itself. But if we’re going to run another story about a topic that has by now been much-covered we need to advance it."
If you have ever thought of starting a Substack but haven't gotten around to doing it, this is The Sign to do it immediately.
I recently began my own. Wrote some vague thoughts. Didn't even think it was good enough to publish but I did it anyway.
To my surprise a few people read it. Commented and liked it.
It's honestly one of the most gratifying things I've ever done. It's another high altogether to create some intellectual work and see it resonate with someone else and to know that I've contributed to the discussion.
I appreciate that you thought about this! Thank you for being mindful of that. I do also think it's fair to share the link since someone outright asked about it, though. :)
Ah, I didn't know there was one. Open threads say, "Post about anything you want". While I appreciate your reserve, the substack interface of showing the link to anybody's substack who has one seems low-key enough that I wouldn't have gone to any trouble to obscure it, even if your post is about the joy of starting a substack.
Especially this tidbit: "I can’t help but wonder if some of the complaints about “late-stage capitalism” actually break down into something like “this system has gotten oppressively efficient and therefore overfitted, and overfitted systems suck to live in.”"
3. Scott's recent vibecession explanation idea along the lines of "maybe it seems to take a lot more effort to achieve slightly better results nowadays?"
Does it seem to others like there is something here? Also are there any existing posts that already make and flesh out this by any chance? I'm not sure how one could turn ideas like "maybe there is more optimization pressure, less slack, more overfitting, and it doesn't show up in anything easy to measure but it's a real problem that feeds into the bad vibes" into a coherent story and falsifiable predictions...
I do not say this lightly, but I believe my Fractal Stalemate provides the structural and political mechanics for the cultural phenomenon Hoel describes, though as my analysis is systematic, it's much looser in its rigor.
I invite you to come and take a look. This is the "canonical" release order, for your convenience.
Perhaps the disconnect between metrics & perceptions are that the metrics only matter to those above the API but the perceptions are most vocally coming from below it.
Yes, absolutely. "You can just do things" might be true, but I think it's much less true than it used to be. In particular, "oppressively efficient" is an interesting concept, and jibes with something I perceive about current society vs, say, a hundred years ago: there are now very few opportunities for individual people to get enterprises started. I don't mean VC-powered psychopath stuff with "cap tables" and "exits", I mean nice healthy businesses that get started in garages and grow organically, possibly ultimately into household names. Overall you can't really even call it a bad thing, because it's largely the result of economies of scale and regulations; the products and services society as a whole is receiving are massively better. But yeah, for an individual who doesn't want to be a cog in a machine, it sucks.
That said, not everyone would even want to get involved in this sort of thing, so it's definitely not going to fully explain the vibecession stuff.
Another closely related concept that's a favorite of mine is just complexity in general. Nobody cares about limiting complexity its own sake, but it's becoming a real problem. In the happy path, things are orders of magnitude more efficient, but the failure modes have become way more miserable. Your grandparents might have had some real struggles in their day, but they didn't have to navigate anything like the Kafkaesque hell that modern customer service can descend into. They would immediately be in contact with a human interfacing with a simple paper-based system, with the ability to do whatever was needed - as opposed to "sorry, the button on my computer screen is greyed out; I don't know what else to do."
I'm not sure what precisely Hoel means by overfitting (he acknowledges that he's using it in a few different senses) -- but I broadly agree with his idea. I would argue that overfitted is almost the opposite of efficient - it's what happens when the system has too many resources and underdetermined goals and so wanders into Baroque details and performances which don't progress toward the actual agenda.
I think this is the same sort of failure as Orwell pointed out 80 years ago, where the enemy is simultaneously 'weak and contemptible and we are sure to win' and 'immensely powerful and all measures must be taken'. Maybe the system can both be wringing every last drop of efficiency out of workers through electronic surveillance and performance management; but also massively overfitted and forcing them to do bullshit jobs and further projects that produce no conceivable benefit.
I think that a large part of the paradox can be explained as "goodharting", or the difference between *actual* productivity and the *signals* of productivity. As a manager, it is difficult to make your team more *productive*, but it is easy to keep them *busy* all the time... and that kinda seems like the same thing (except it is not). So the mediocre managers keep sharing tips on how to keep their employees more busy, thinking that it makes them great leaders. (Measure the bathroom breaks!)
Even when someone comes with a genuinely good idea, it is immediately perverted by the stupid and/or malicious followers, who adopt the buzzwords while eliminating the essence of what made it a good idea. (For example, there is a sensible idea that if you want to improve something, you should *measure* it first. You can pervert this idea by measuring the wrong things and ignoring the important ones, or by introducing a system of measuring so annoying that doing the bureaucracy wastes a lot of time and energy. Furthermore, you can make your research prove whatever you want it to prove, by simply eliminating the "wrong" answers from the questionnaire.)
For example, in programming it is ironic how in 1990s a group of five developers could produce an entire game, and its sequel, in the same year; but these days it would take the same group, and at least the same number of managers on top of them, a few months to add a checkbox to an existing application. The developers today would have plenty of meetings to attend every day, Jira tickets to fill, and maybe a list of annually evaluated self-improvement individual goals and team goals as a cherry on top. Their managers would produce PowerPoint presentations and burn-down charts, and their boss would collect a huge bonus for making the company more agile or whatever is the buzzword of the year.
If you want to understand why developers were so much more productive in 1990 I would read The Mythical Man Month.
TL;DR. Modern programs are too big for 5 people, and that requires a lot of overhead like Jira tickets. Which is not to say bad managers can't be part of the problem. But Jira tickets evolved to fill a real need and not just "goodharting".
Ah, yes and no. Yes, the modern programs are more complicated -- when the program in 1990s e.g. solved an equation, it could read it from standard input and write the solution to the standard output. These days, the same program would have to be implemented as a REST endpoint with authentication and encryption, password rotation, log rotation, installed in a container, and dozen more things; the actual equation solving would be just a tiny part of the whole. That's how complicated things are these days.
On the other hand, most modern managers still haven't learned the lessons from The Mythical Man Month. They still treat programmers as perfectly fungible; they are not afraid of losing an experienced developer who worked ten years on a project and has tons of tacit knowledge, because they believe they can easily replace him by a random guy they hire tomorrow... and if they can't hire a new guy tomorrow, then the old guy's work will be assigned to a random colleague on top of that colleague's existing duties; eh, it will work somehow.
A few times I saw a project advance dramatically when the responsible manager got sick for one month and the developers worked unsupervised. In one case I was the only developer on the project, and I have completed it during that month, and the company paid me a huge bonus afterwards. But when I suggested that I could work on the next project the same way, the company refused, because it was against the rules or something. (I quit soon afterwards.)
That one month when I worked on the project alone, this is what I did: The first week I just studied the requirements, to make sure I understand everything that is needed to do. The second week I got familiar with the tools: I studied the tutorials, made a few proofs of concept, etc. At the end of the second week I sketched all program dialogs, and showed them to the customer: "hey, if this is what I will do for you, is that exactly the thing you wanted me to do?". After approval, I spend the remaining two weeks coding, uninterrupted, and delivered what I promised.
Here is how the same thing would most likely work *with* a manager: The manager would talk to the customer, and would keep the requirements to himself. He would spoon-feed me the information, as he would deem necessary. That would sometimes require that I throw away something I already made, and do it differently in the light of new information (and prepare the install scripts both for the old and the new versions), but that's called "agile". We would have a meeting where I would have to provide an estimate how long the work will take (without knowing many relevant details), and then once or twice a week we would have a meeting where I would report my progress (always reminding me that the arbitrary deadline is coming closer). The code would be written in the order decided by the manager (despite the fact that he has zero programming experience); for example he might decide that logging should be added last, and unit testing skipped entirely. He might randomly assign another developer to help me with the project for a few days, despite the fact that the other guy knows nothing about it, and it will take him those few days to get familiar with the thing. Just like that, in the middle of the project he might assign me to a different project, because they urgently need some help (even if I don't have the kind of knowledge they are looking for). After three or four months, the project would be complete, the manager would collect the bonus, and one of the developers would probably be burned out and quit.
(My problem with Jira tickets is that they usually come with some kind of workflow that is completely unrelated to how things actually work, and in theory you can adapt the workflow to your project's needs, but in practice, adapting the workflow would be a waste of resources you won't get approved, and you are required to use it as it is by default. So everyone keeps guessing whether "Resolved" means that the feature was *implemented*, or that the tester *approved* it, or that it was *installed* in production. Making a convenient filter for "things that I am supposed to actually work on this week" is sometimes impossible.)
> Yes, the modern programs are more complicated -- when the program in 1990s e.g. solved an equation, it could read it from standard input and write the solution to the standard output. These days, the same program would have to be implemented as a REST endpoint with authentication and encryption, password rotation, log rotation, installed in a container, and dozen more things; the actual equation solving would be just a tiny part of the whole. That's how complicated things are these days.
You imply that this is a ton of junk surrounding a similarly actually useful core that hasn't improved since the 1990s.
I disagree. REST endpoint means its available anywhere in the world and yet can be managed from a single centralised location. If your equation is calculating tax on sales, and the government changes the tax, a central location makes it a lot easier than rolling out an update across 1000 work stations.
As for security, I don't think I need to say why security is a good thing.
And writing a rest endpoint isn't harder than writing a read from standard input. Throw a couple of annotations into your code and its done; (granted if you count the complexity of the appserver...)
I think it fails to tackle where this revolution of rising expectations come from. Ofc it's only my gut feeling, but I believe this is one of the major failing point of capitalism:
Investors want more dividends -> Companies need to sell more -> Companies create ads to make people want buy more stuff -> People's expectations raise significantly
I'm not saying that this makes capitalism a bad system, but I believe we need to work out around this flaw as a society if we want this system to be sustainable on the very long term.
I actually think it's more driven by social media than traditional advertising: plenty of ads were all over the place in the '50s/'60s/'70s. I think the peer comparison that happens on social media (both your literal peers, and also "peers" in the form of national/global influencers) exerts a much more powerful sense of comparison and "need for more" than corporate advertising. The infamous "Gen Z says they need $500k/yr to be successful" stat is probably more a results of try-on videos, shopping hauls, and #richkidsofinstagram than clever marketing execs at BMW / Prada / etc.
I think you are both wrong. This age is not more consumerist than the past and might be less consumerist. The definition of successful is going to be context or generational dependent. Does it mean comfortable? I’d out it at a higher level than that. And boomers would be living high on much lower nominal income anyway - given they don’t have to save for pensions, pay mortgage or rent or live close to a cut for work.
You completely misunderstood Zvi's article, he is saying the exact opposite of you. There's a reason gen Z is saying that, and you're being an ass for not trying to understand why.
"Women in America are having as many babies over their lifetimes as they did two decades ago"
"For decades, across the West, the commonest measure of how many children women have, on average—the total fertility rate (tfr)—has been dropping. It has fallen well below the “replacement rate” (needed to keep a population constant) of 2.1. Between 2007 and 2022 America’s tfr dropped from 2.1 to just 1.6. Warnings of probable population decline, including from this newspaper, abound. Pronatalists are casting around frantically for solutions....
But tfr has flaws. It measures the number of children a hypothetical woman might have in her life, based on current birth rates for women of different ages. To calculate it, you start by dividing the total number of births to a given group of women (aged, say 15-19 or 35-39) by the number of women in that group. Those group birth rates are then added up and multiplied by the number of years in each group (ie, five). The result is the number of children a woman will have if, aged 15-19, she has the same number of babies as the average woman of 15-19 is having today, then the same number at 20-24 as the average woman of 20-24 is having today, and so on. It is almost sure to be wrong. That does not make TFR useless. It captures what demographers call “tempo”—the timing of an event, like a birth. But as a measure of “quantum”—the total number of births—it is likely to mislead. In effect it assumes that a young woman will have babies not only at the same rate as her contemporaries, but also at the same rate as women five or ten years older. That is not necessarily so.
A different measure, the completed fertility rate (CFR), captures the average number of births a woman has by the end of her child-bearing years (put at 44). America’s CFR has not fallen at all over the past two decades. In fact, according to an analysis of census data by Mike Konczal, a former Biden administration economist, it has risen slightly, from 1.91 in 2000 to 1.97 in 2024....
Many demographers, including Lyman Stone, of the Institute for Family Studies, a think-tank, point out that tfr is typically a strong predictor of cfr. But there is evidence that women are deferring, not forgoing, childbirth. In 2000 by age 26 the average American woman had produced one child; the average 32-year-old, 1.6 children; the average 40-year-old, 1.9. In 2024 the average 26-year-old had 0.6 children and the average 32-year-old 1.2. But the average 40-year-old has still had 1.9, having delayed in her 20s and caught up in her 30s.
Does this mean...that tfr will rise to meet cfr? Historical examples are suggestive. As women delayed childbearing in Sweden the tfr dropped sharply in the 1980s, before recovering in the 1990s (though it has fallen since). The cfr, at around 2, barely budged. In America the question is pertinent. The commonest ages are 33 and 34, because of a baby boom in 1990 and 1991. Women aged 34 have had, on average, 1.46 children each. As this group approaches their 40s, many more births may be on the way. As for younger women, those born in 2000 have had fewer children than those born in 1990 had had by 2015. Perhaps they will catch up—but that won’t be known for at least a decade.
To be sure, an average of just under two births per woman is still not “replacement level”. But if that truly is the long-term average, panic is scarcely justified. It would mean a gradual population decline: tricky, but much more manageable than implied by TFR...."
If tfr and cfr are strongly correlated, then cfr is likely a lagging indicator. We won't know if birthrates are really falling for young women for decades, though intuitively it would be hard for women 30+ to have 3+ kids that would balance out those who had 0, 1 without major advances (we're not there yet).
Given that uncertainty, the lengthy times to affect birth rates through public policy, and the huge downsides of population collapse, probably best to encourage procreation now just to be safe.
There are real physiological reasons we'd expect deferring births to lead to notably lower total fertility, too.
When you wait until early 30's to have your first child, as is the norm in the US and UK now, you have already burned through ~2/3 of your cumulative lifetime fertility (the area under the curve in this graph leading up to age 30):
That's from Geruso 2023, Age and Infertility revisited. Broadly, you go from 60% chance of live birth in early 20's to 30% at age 30, 11% at age 40, and ~1% at age 45:
That's a REALLY big difference in fertility. Many, many more people are going to try and fail in their thirties and forties versus in the halcyon fifties when women in their 20's were having kids.
Maybe we should make IVF free / state funded? Sure, you can, I think it's probably incrementally net positive. But many EU countries have done this already, along with entirely state funded daycare, significant baby bonuses, and more, and haven't moved their overall fertility in the positive direction.
You know what HAS been tried, unsuccessfully?
* $10k bonuses per child (Singapore), or for 2nd / 3rd children (Russia)
* 3 years paid parental leave (France)
* 480 days paid leave at 80% wage (Sweden)
* Income tax exemption for mothers with 4 or more kids (Hungary)
* Free state paid child care (France, Germany, Sweden, Norway, Finland, Estonia)
* Free IVF (most EU countries)
* $750 / month payments per child (South Korea)
So the problem is pretty hard to move the needle on.
Married woman under 25 with one child get free 3 bedroom housing. Under 30 with 2, 4 bedroom housing. Under 35 with 3,5 bedrooms. Etc.
Its paternalistic, forces them to "spend" the subsidy on excessive housing. I dont want them wasting the money on 0 sum status goods like private school, or even day care which encourages a return to work and reduction in future futility. The money directly contributes to ability to have *and grow* the family with the extra bedroom
It spends nothing on 2 child couples in 30s. Those are the default waited til we could afford it, hitting the minimum for replacement. Great for those ppl. They dont need a subsidy for that.
The expiration creates a strong psychological nudge to keep going.
Maybe on some margins it incentives better govt policy on home construction as they are the ones paying for it.
Still its pretty expensive. Could still fail. But I think its more thoughtful than just shower parents with goodies
> Married woman under 25 with one child get free 3 bedroom housing. Under 30 with 2, 4 bedroom housing. Under 35 with 3,5 bedrooms. Etc.
Okay, ran some rough napkin math on this.
For the top ten CBSA's by pop, this costs you ~$25B a year. For the top 20, it's $34B and the top 50 $46B.
The top 50 CBSA's is more than half the US population, and it includes basically everywhere with price problems. You're getting down to Ogden, UT and Bakersfield CA and Colorado Springs by that point, places with relatively reasonable house prices and rents for the area.
So prospectively even if you expand it to basically everywhere, it's going to cost <$100B a year, which is essentially rounding error in the Federal Budget.
Its meant to be a subsidy on the supply of new children, not an effort to bring down housing costs. It would be judged based on how many new births it creates.
I would also like much more liberal house construction policies, but I am not confident it would make fertility problems disappear (yimbys love japan housing policy after all). If you are maybe you should just say that. Anyways wed both agree on reducing housing restrictions.
I think its generally true that any increase in supply somewhere would increase demand elsewhere (usually in housing) so youre proving too much with the meme
I love everything about it! Sure beats the hell out of spending trillions of dollars on rich old people every year, spending that I don't think anybody currently below 40 expects to exist by the time they're old.
And it brings fertility to the areas and people that need it most, our insane urban metros staffed with all the PMC people. I have no idea what a 5 bedroom place would even cost in NYC or SF, but it's gonna be a lot, so it's scaled directly to be maximally relevant, where it matters most.
I'll start doing some Fermi math on this to see just how crazy it can get.
EDIT: did the math. Even if you expand it to basically everywhere, it's going to cost <$100B a year, which is essentially rounding error in the Federal Budget. We waste 5-10 times that on completely irrelevant stuff every year, see my above comment.
CFR is _completed_ _births_, not pregnancies or attempted pregnancies. So lower odds of successfully carrying to term at higher maternal ages are priced in to it. Women who've decided to have a child at 40 despite 11% odds, do not stop trying after one (or in my personal observation two or three) failed attempts.
Also the US CFR is currently trending _up_: from 1.86 as of 2006 it hit 1.97 for 2024.
All of which makes the Economist's closing point, above, sensible.
To your last point with all the different governmental schemes failing to increase birth rates: I think you’re right directionally, but in the two cases I know a little bit about, the details are off.
To my knowledge, France has paid parental leave for a few months, and unpaid (but with ongoing health insurance etc) for a year.
In Germany, child care is subsidized differently from state to state, and even district to district in some cases, but rarely really free. What parents pay depends on the state and their own income and can range from 0 to a little over 1000€ a month.
It would be interesting to model the effects of a constant-CFR explanation for TFR on entitlements and taxes. Probably there is still reason to panic, right?
If you assume that only one generation is getting SS+medicare before the change, then each retiree has fewer descendants to tax. But if there are currently multiple generations of retirees, dropping that to just one could (eventually!) help.
I think in either case, the short term still sucks for everyone
But there is evidence that women are deferring, not forgoing, childbirth
I think the problem is that statistically, deferring will necessarily mean foregoing as success rates will be lower at higher ages and it eliminates the possibility of larger families.
The trend dynamics are interesting though. If you imagine a 20 yr shift from one state to a new attempted-deferred-but-with-some-failures state, then TFR will drop quickly as the younger age brackets have fewer births. These will eventually be offset by higher birth rates in older groups but dont show up yet. CFR is slow to drop. Then as the shift completes TFR rises back to converge on a new lower CFR. In this scenario TFR is a leading indicator but also an overshooter.
So I can buy the overall narrative you are suggesting. TFR may be over stating the decline but there is still a real decline and we are now below replacement rate, partly because ppl are waiting too long to start.
My policy intervention would focus subsidies on younger parents. Not all parents.
CFR is _completed_ _births_, not pregnancies or attempted pregnancies. So lower odds of successfully carrying to term at higher maternal ages, and reduced odds of larger broods, are both priced in to it.
Women who've decided to have a child at 40 despite 11% odds, do not stop trying after one (or in my personal observation two or three) failed attempts.
Also the US CFR is currently trending _up_: from 1.86 as of 2006 it hit 1.97 for 2024.
All of which makes the Economist's closing point, above, sensible.
I understand the definition of CFR and its a positive sign that it is going up. I was referring to the narrative driver. All things equal women waiting longer will drive down births and so CFR, no matter how hard they try. If it is rising right now its for another reason than the timing effect. We should still be concerned about the decisions to wait and try to address factors that could be contributing it that arent all about preference. The housing theory of everything looms large here
"All things equal women waiting longer will drive down births and so CFR, no matter how hard they try."
You are imposing future certainty well beyond what current facts will support. Put another way, "all things equal" is carrying a _lot_ of weight there.
E.g. why are you assuming that all the women who don't have children until late, are among those who otherwise would have had children earlier?
E.g. modern medical science has obviously _improved_ the chances of successful procreation past 35; why do you assume that progress has now ceased?
Other obvious real-world uncertainties/variables exist. Clearly though you're just dug in on your narrative making further dialogue is pointless, so, bye.
My first post was supportive of your overall view but insisted its dangerous to be too complacent on rising birth ages. This is getting pretty hostile. I dont make any of the assumptions you accuse me of.
I think it is a valuable PSA that waiting necessarily reduces expected total births. Improved medical science can help, but it wont eliminate the reduction. The harder it is to conceive that 1st, 2nd or 3rd child the less likely the couple will even try for that 2nd 3rd or 4th.
I myself ended up with fewer children than I want because I was too complacent about the years. I want an information and policy environment that makes that less likely for new parents. If you would rather wishcast the challenges away, youre not helping. Bye
The tricky thing about anti-discrimination is that if you've been discriminatory in the past, and turnover is pretty low, then just to get to something approximating the racial balance in American in the near term would seem discriminatory in new hiring.
Like if your company basically hired so few black people that it's only 1% black folks, and they have an annual turnover rate of 10%, then to get to 13% black folks (the American average) you'd basically have to spend more than a year hiring nobody but black hires. If you only have 13% of the hires being black, then in practice the percentage of black employees at your company is only increasing by slightly more than 1% of the company per year - that's a whole decade just to rectify the situation, if not more (assuming there's no recessions that lead to layoffs, etc).
The "trick" is to have a realistic understanding of what non discriminatory hiring looks like and not expect something else. I think there are many who understand what that looks like, but want something else anyways. So the second "trick" is to ignore those ppl. That part is harder than the first, but the whole DEI backlash is about giving people in power to do just that
If the problem with discrimination is that you're hiring people on the basis of demographics rather than merit, it seems like "let's hire people EVEN MORE on the basis of demographics rather than merit, but different demographics this time" makes the problem worse rather than better.
On the other hand, if your model is that anyone at all can do this job, it doesn't matter if they have any skills, so jobs are literally just handouts -- in that case, yeah, you should probably discriminate in the opposite direction to get back to parity in who's getting the handouts.
If an organization is discriminating in a direction opposite to what they've done in the past in order to reach an overall balance in their demographic composition, that's generally only useful as a way to superficially avoid the appearance of unfairness (by obscuring the evidence of their previous discrimination). In some cases perhaps it's useful for a group making important decisions to come from a representative range of backgrounds, but in most cases this approach is just straightforwardly a bad idea.
Or, rather, which assumptions make that true, and shouldn't we be arguing about those instead?
For example, I might assume that there is an equal rate of qualified candidates from each race, but that some form of bias - maybe subconscious racism, maybe the result of social networking being partially segregated, maybe cultural deficits in resume-writing or economic deficits in resume-padding that don't affect actual job performance, or maybe any combination of a hundred similar factors - has led to blacks being hired at 1/10th the rate of whites.
If that were true, you should expect that there's a huge untapped pool of overqualified black candidates out there, and that if the scales were suddenly lifted from your eyes and you could estimate true job performance from a resume, you'd get massive gains from hiring only black candidates for the forseeable future.
I hadn't considered that the same candidates who were unfairly passed over in the previous hiring regime would still be looking for the same sort of positions. That seems like it would be only partially true.
It is possible for people to be racist and for that racism to be justified. In that case, there is no huge untapped pool of overqualified blacks, and any effort to hire blacks for the sake of exploiting untapped resources is worthless unless you can pay them less than whites, which is obviously not possible right now.
Northern Ireland has that problem with making the police force more acceptable to Catholics. To do it quickly would be 100% Catholic only hires for decades. Widely they do it slowly.
In my country (not the US), there was an acknowledged problem with women accessing the higher ranks of academia. I saw recent research which suggests that 75% of new professorships went to women over the past 3-5 years (I don't remember the timeframe). As far as I can tell, this wasn't because of the imposition of quotas or rules, but each university decided to remedy the figures through direct action.
Research bodies in the country like to announce that 'over 50% of awards in every discipline were made to women' -- again this is fine. But they should publish success rates by gender if this is so: the end result will be to push universities to hire women, who have a better chance of being successful, until gender balance is achieved. But this should be made explicit: men should be aware that their chances are depressed as a result.
I don't have any argument about this - it needed to and needs to happen. But when it's a net zero game, creating winners has to result in losers as well, and this should be acknowledged.
'Don't do murder' is correct, even if self-defense, militaries, assisted suicide for terminal patients in pain, and abortion should all be legal.
We just call them something else other than 'murder' so that we can pretend we have a simple, bright-line, one-sentence policy. 'Don't do murder'.
Of course, we don't have *any* policies in reality that are that simple. Whenever we try to make it look that way, we're just playing semantic games.
We can play semantic games on this topic, too. Call it 'equity' or whatever, it doesn't matter. It just boils down to 'Discrimination is always bad, unless it's the type that produces good outcomes we want, in which case we call it something else.'
Same as for murder, theft (taxes, child support), kidnapping (carceral justice system, child protective services), etc.
It seems that if you want to move towards a world in which hiring decisions aren't made on the basis of sex then perhaps you should move _towards_ a world in which hiring decisions aren't based on sex.
The murder analogy does not apply, because we're not comitting murders in order to get to a world free of murder. In fact that is generally called vigilantism and while it has some emotional appeal it's generally between illegal and extremely illegal.
We do allow for self-defense, because moving to a world free of killing is not one of our goals. And because self-defense is materially different from murder.
It's unclear to me, that discriminating in favor of men is materially different from discriminating in favor of women, with very few exceptions.
Lets say we had the omniscient God-AI scan every particle on Earth and tell us with absolute infallibility that women are being undermentored during education and training and underestimated by recruiters such that the competence of a woman employee is actually 10% higher than what the average recruiters estimates, and that if every recruiter hired the best marginal 10% more women than they had been intending to (or however that math works out), this would actually lead to perfect meritocratic recruitment with optimal efficiency and performance.
This is sex discrimination. It is telling recruiters too hire 10% more women than they wanted or intended to.
Do you think this type of discrimination is wrong?
This uses sex as a proxy for competence, I take no issue with that. I would file that under the "very few exceptions" I mentioned above, and I hope you will grant that this is indeed an exceptional situation.
Honestly I'm not even sure I'd call that sex based discrimination, or at the very least the central example of sex based discrimination is to discriminate against somebody _because of_ their sex. Here we're attempting to discriminate by competence, and use sex as a proxy for competence.
If you compare it to hiring somebody for the purpose of bringing the gender balance closer to 50 / 50 it seems substantially different, to me anyway.
Whether it's right or not, I've accepted that this is the way the powers that be have decided to run things. I made the decision to leave academia and find work elsewhere, that wasn't the most compelling reason.
For the record, I do think that creating a more equal society is a worthwhile goal. I don't think this is the right way to do it, and I think there's a case to be taken there; I'm not willing to invest the time, money and reputation to do so.
Then why the comment that it "needed to happen"? Why does there need to be gender balance anywhere when the genders aren't equal in the first place? They should be hiring on merit, using the best people with the best properties for the job.
Over the past 20 years, there were a number of cases where clearly qualified women were overlooked, took court cases and obtained promotions because the universities had provably discriminated against them. The universities needed to correct these issues - there's been generational change in the higher leadership roles and the new generation seems to have over corrected.
Merit is hard to establish for academic positions (actually I'm negative on hiring in general as a way of establishing fit for a job). Track record 3 years out from a PhD is a poor indicator of who'll make a career-long contribution to their research area, to the university's teaching and to whatever service criteria are valued at various times. Much like any workplace, value is going to be created on an 80-20 type rule.
It seems that one of the fundamental challenges of the day is that people of differing opinions aren't talking to each other. I've frequently tried to reach out to people with strongly different opinions (e.g. people who oppose vaccines, Trump voters), and while in a couple cases it has led to fruitful conversations, it always struck people as bizarre and unconventional. This is strange to me given that it seems everyone agrees that "echo chambers" are one of the fundamental problems of the day, which everyone acknowledges and bemoans.
Most striking is that I just don't know of any major platforms where you can attempt such dialogue. I think it's clear that Facebook and Twitter are hardly good platforms for this, not to mention comments sections. I've seen a couple attempts (e.g. Braver Angels' "1-on-1 conversations: https://braverangels.org/online/1-1-conversations/#participate) but they seem not to have really gained much traction (as I understand it there's always a dearth of right-wing/Trump supporters). What website, app or location can facilitate me either finding someone with different opinions to talk to, or alternatively suggest myself as someone open to this kind of debate? Do you know of places where good discussions of this kind take place? And why does it seem like these things are so rare if it's a pretty basic way of addressing a universally acknowledged problem of the day?
I actually know tons of stories like that-- heck, there are lots of stories of Palestinians who will say things like "My brother was killed by an Israeli soldier, I was planning to be a militant to kill as many Israelis as possible, then I met some Israelis and I became a peace and coexistence activist" (one example: https://www.youtube.com/watch?v=BzDEFAGv_hc). This and many other examples (plus personal experiences) make me absolutely convinced that dialogue can work phenomenally well, but "can work" and "does work at scale" are not the same thing. Still, it seems like a mostly untested question, which I really think is bonkers.
I think the key to the Jewish guy’s getting through to the Westboro woman is that he did not ever express annoyance, hurt or contempt. He did challenge her religious ideas, but did it in a jokey way, never with any indignation or scorn. I love that guy.
I think it helps, but sadly it isn't a panacea. I engaged with a couple people who responded just as disingenuously when I was trying to be friendly and expressing "we're on the same side here" and such than when I did become a little exasperated.
I think the key is that there are lots of different kinds of people, some are the eternally online types for whom it's hard to break through even with lots of good will, and many who are much more open to dialogue even if extreme. I think part of the problem is that, more or less by definition, you're more likely to encounter eternally online people in FB comments and such. But not necessarily so if the site is dedicated to intelligent debate (eg this comments section has unusually high quality debate).
Well when the Jewish guy, whose name I can't remember right now, first started a dialogue with Megan, the Westboro woman, she was saying that God hates the Jews because they killed Jesus. She wasn't being disingenuously nice, she was hitting him with the worst she had.
Oh I don't think people like that are being disingenuously nice, I think they're being genuinely offensive, and I've had plenty of stuff like that thrown my way (I once had a conversation with an anti-vaccination person who accused me of being a holocaust denier-- because I denied that doctors were committing a holocaust on patients by administering the vaccines. She also would regularly denigrate my mother after learning she was a doctor).
The people you want to get to are the genuine true-believers, rather than the opportunists or the disingenuous ones.
I think people of differing opinions talk to each other too much, in fact. Having to interface with some of the worst people on the planet every hour of the day is one of the reasons we have such incredibly toxic institutions today.
If I'm defining the people I want to talk to as Trump supporters, that's tens of millions of people, plenty of them good people who I think are making a mistake, rather than "some of the worst people on the planet."
Out of curiousity, do you know any Trump supporters personally? Have you spoken to any about the reasons they support him?
To be clear, I'm as anti-Trump as they get. I consider the USAID cuts to be one of the most horrific, and one of the most pointless, single acts in modern American political history. But what I see is people supporting this stuff because the actual death and destruction caused gets filtered out on their feeds, which is instead filled with stories (some true, much fabricated) about the massive waste being cut. And I know good people who constantly fall for this stuff. And that's why I do this, because the way evil wins is not by turning everyone into heartless people, but by convincing good people that they are supporting good when they are in fact supporting evil.
(And of course many Trump supporters will say the same about Biden supporters-- we don't actually care if inflation skyrocketed under Biden because we don't care about people who are struggling etc. etc. etc.).
It's worth noting that otherwise-decent people routinely have horrible political views. We have no natural intuition for politics at a national scale, and almost everyone is just following their ingroup.
I think my model is that people of differing opinions are talking to each other as much as ever; the problem is that people of the *same* opinions are talking to each other way more, and are amplifying each other.
I'm not even sure this is right, in the sense that I'm not sure people are really discussing issues with each other at all, even people "on the same side." I think that might reveal to them how many deep disagreements there are within these camps, to the extent of being diametrically opposed (think e.g. free market people and protectionists in the Trump camp, or isolationists and "neocons"). And I can say of myself that I think many people I know feel very strongly about things that they don't talk to anyone about-- they consume things on the internet and maybe write posts, but with minimal interaction of any kind.
That's my impression, in any case.
But let me make this more interesting. If you had a website or app that let you connect with someone who differs from you on some issue, and just lets you message them or something, would that appeal to you? I feel it would appeal to me and I'd expect something like this to exist but I haven't seen anything like this.
> If you had a website or app that let you connect with someone who differs from you on some issue, and just lets you message them or something, would that appeal to you?
I enjoy the sort of conversation where two people are trying to find the truth, but I don't enjoy the sort of conversation where two people are trying to persuade the other person while preserving their own belief.
I think the app you describe would mostly get used for that second type of conversation, and thus I would not enjoy it.
I have a clear, simple, mildly depressing answer for you: the only platform where you can manage that sort of dialogue is real life. Text communication just blunts too many of the little delicate signals that keep people feeling human to each other. I'm not even singling out the algorithmic social media boogeyman here; I don't even think traditional forums, email chains, one-on-one messaging, *any* textual medium is acceptable. (I think it is doable if both sides come to it as committed as you are to a calm exchange of ideas, but that limits the potential reach so much as to be basically irrelevant).
I think political discourse needs to get off of the internet as much as possible to have a chance of improving.
So the reason I'm not so sure is that even with the lack of any kind of platform like this, I have occasionally found people to talk to via facebook and things like that, and some really interesting conversations sometimes developed. And I'm talking the most basic of basic search mechanisms-- a few people I found as commenters on another guy's FB page, one person is the anti-vaccination mother of a friend.
Maybe I'm wrong and these conversations aren't really possible "at scale." But it seems like crazier things have been tried even if they've failed. And here it just seems like everyone complains about the problem but few are talking about what seems like the most straightforward way to begin addressing it.
Yeah, that's basically the scale thing I'm talking about. Or, even beyond that: it's not just that you won't reach enough people, it's that the people you're reaching were already perfectly fine. That is, if you're approaching it in "make the world a better place" terms, the people you need to reach should be locked into enough anger that they are closed off to this sort of dialogue (in pure text settings), and reachable only after being softened up by a way kinder than normal personality, experienced directly in person. That's just my intuition; if the online interactions you're mentioning are you talking people out of "Obama is micro-chipping the vaccines" or "I'm delighted Charlie Kirk was assassinated", then that would be incredible and I would stand very corrected.
Also don't get me wrong, I do think what you're describing is a great thing to deliberately look for, regardless of how large the impact is. I was just responding to the more ambitious version that it sounded like you were going for.
You're right that I'm banking on there being enough people who (a) Are open enough to dialogue that this would be useful and (b) Aren't already on my side. My experience suggests that there are many such people, but it's a small and oddly selected sample (which is why I'm asking these things :)). And just to be completely aboveboard: I've had many conversations that were very respectful and substantive, but it's been rare that actual positions have changed. And while I do think that's valuable, I do want there to be an actual way for dialogue to change views.
Perhaps a good example of a fairly "mixed" forum is https://order-order.com/ a British political blog. named after someone called Guy (AKA Guido) Fawkes, who tried to blow up the Houses of Parliament in the early 1600s by hiring storage space in the cellars and stacking them full of gunpowder barrels!
It has a majority of right wingers (of which I am one), but several left wing posters who add a bit of spice, and occasionally food for thought, with dissenting opinions. Sometimes the exchanges are civil, but I must concede that more often than not they are anything but and there is trolling and mud-slinging on both sides.
It's the break off forum from the old SSC forums, basically a backup from the 2021 NYT doxing when we weren't sure whether SSC was going to come back.
I think it'll be useful for you to get some insight into Trump/Conservative voters and good debate but also insight into just how fast and how deep echo chambers form. Most of the posters are guys who were posting here 4 years ago and they're now basically a separate ecosystem with...dramatically different views than these Open Threads. And that's not just them, the Open Threads have shifted dramatically as well. Just...talk with them a bit and get a vibe for just how different they are from this community and just how quickly these two forums diverged.
+1 for Data Secrets Lox. It has good norms for discussing ideas (both inside and outside of the Overton window) on their merits rather than angrily dismissing the people who hold them (not always perfectly lived up to, but fairly well), and if you're specifically looking for right-wingers there are lots.
First, it's just really Abundance Dem here. For reference, in Scott's 2025 survey, ~7.5% of respondents had a positive view of Trump, ~10% had a neutral view, and...~82% were negative, 62% extremely negative (1). This not only isn't representative of the country, it's significantly further left wing than SSC, which was 60% liberal-leftist, 20% libertarian, and...let's say ~10% right-wing/neoreactionary/alt-right. This place is centrist like "Capitalism is okay and we should probably roll back wokeness a bit" not, like, "I might actually vote for Vance in 2028." Neoliberal, I believe is the old term.
Second, and I'm less qualified to discuss this, but people just stopped talking around....20...18-ish? Discourse is dead. Like, in the 90s, we talked politics. Politics was fun and interesting. In 2014-2016, discourse was huge, crazy stuff was happening and everyone was on Facebook and talking about it. And then...we stopped talking. Don't know why. Nobody talks talks on Reddit or Twitter anymore, everyone just dunks and...whatever degenerate crap the kids are up to these days. There's got to be a million theories why but we don't...talk, we don't do discourse and debate like we used to.
And it's hit ACX hard. A comment chain doesn't come to any synthesis anymore, it doesn't come to any agreement it's...either two autists screaming at each other (3) or it terminates in "Oh, we have irreconcilable factual or value differences. Good day".
I dunno, the second one is much fuzier and "vibe-ier" but it's also the one I think drives it more. Thing degenerate into echo chambers so much faster and...we tune each other out so faster. I dunno, I just remember spending a lot of time arguing on the internet because I genuinely thought I could learn something or change someone's mind and now that just seems...hopelessly naive.
The right-wing/Trump people would be happy to share their views and discuss, but have no interest in being punching bags for abuse, scorn and vitriol which is what they (reasonably) expect these things to become.
Would YOU be interested in having the kind of "fruitful conversation" you describe with a member of Hamas? If not, you understand why many on the left have little interest in such discussions with Republicans and Trump supporters.
In a word-- yes. I would like to have a conversation with a Hamas supporter. I've sort of done as much in the past, though nothing that lasted for any significant period of time.
I didn't say Hamas SUPPORTER. An actual member of Hamas. And okay, even if you yourself would, I hope you recognize you'd be extremely unusual in that regard, which was my point.
I would indeed be very unusual, but as bad as polarization has become, most liberals don't see the average conservative as a literal murderer, and vice versa. So I think you're right in principle that there's a limit here, but there should be a significant amount of overlap.
I've had *multiple* people online tell me that supporting Israel with respect to Gaza is supporting genocide. There's also the framing of anti-trans sentiment as literally endangering lives.
Are you familiar with Daryl Davis? He’s a black musician who befriended over 200 Klansmen. He didn’t try to change their views, only to have real conversations with them.
Okay, fair, that's actually my view as well, but I failed to come up with any such group a member of which with whom I WOULDN'T find a fruitful conversation interesting, and I figured Hamas would suffice to make my point. I welcome suggestions for what to change it to.
You can't start with debate, even civil debate. Both sides are now so stuffed full of anger and so indignant about the torrent of ugly caricatures of themselves that there needs to be a long period of establishing trust and understanding each other's rage.
If I am convinced you are a sane intelligent person speaking in good faith, then I can engage with you even when you are saying something outlandish-sounding. And this is really important, because a lot of true and important and valuable things someone might say are also really outlandish-sounding[1]. In an environment where basically everyone is honestly engaging in good faith about their ideas and beliefs, it's possible to learn a lot and understand a lot of very different views even if you are not convinced by them.
But this only works when almost everyone is playing it straight. If there are a substantial number of crazy people, trolls, shills for some politician/party/movement, conmen, etc. in the environment, then just as self-preservation, you need to filter aggressively and the first filter is "that sounds crazy, tune that nutcase out so he doesn't waste my time."
[1] An easy example: From a naive starting point, the theory of evolution sounds absolutely bonkers. Unthinking random processes gave us eyes and hands and minds and created things as elaborate as anthills and vertibrate immune systems and bird songs? Obvious nonsense by some inbred English lunatic. Only when you dig into things, it turns out to be true. For a more modern examples, have you heard of those nutcases in physics classes who claim that clocks and rulers change for different observers depending on relative velocity?
Sounds like you’re trying to do this online. Try in person instead. My experience is most people are far more willing to be open minded and civil in person.
Also, don’t make a program out of it. In fact, don’t even bring up politics. Just be a normal person who has conversations with other people, and then if someone else says something political and you disagree with it, that’s your opportunity to learn more about their views and show them that people who disagree with them aren’t monsters.
In person, there is often an incentive to remain polite. Not just to avoid getting punched, but mostly because there probably is a reason why the two of you are talking to each other -- you have something in common (e.g. the same hobby, or live next to each other), and you don't want to sacrifice the possibility of cooperation just to make a political point aggressively.
Online, the most aggressive person in the crowd will probably reply to you first, because why not -- attacking others is the main reason they are there.
I hypothesize there is a totally different network that kicks on in your brain when you are interacting with a person in front of you than when you are typing back and forth via a virtual medium. The “person in front of me” network is also attached to the other brain networks you developed when you were socialized as a child, including “don’t get punched in the face” but also including compassion and empathy and curiosity and so on. Just a theory.
I agree. Also, in real life, *sometimes* the network in your brain tells you: "this person is a hopeless case, stop wasting your time and walk away". And in a group setting, there is a way to communicate this nonverbally to the rest of the group, e.g. by rolling your eyes as the person speaks.
So basically two problems: first, the instincts work differently; second, sometimes the instincts tell us to do something but there is no user interface for that so we can't.
So this is kind of like with online dating-- yes it's best in person, but infinitely more convenient online :)
More importantly, there's an echo chamber problem here. I'm mostly surrounded by people who are in the same general ballpark in terms of their views. Not that there aren't serious disagreements, but (for example) among my friends I can only count a handful of Trump supporters, to the best of my knowledge. If you're trying to get more views than those in your immediate surroundings, you need some way of reaching broader audiences, and I just don't know how to do that these days, in person or online.
I do think it's important to do this with politics in mind. That's the point! I want to hear them and I want them to hear me, and I'd love for my friends and their friends to watch us both. I agree that just seeing each other as human beings helps, but I don't think it does as much for actually expanding your worldview.
“If you're trying to get more views than those [of friends] in your immediate surroundings, you need some way of reaching broader audiences, and I just don't know how to do that these days, in person or online.“
I will use my life as an example:
Beyond friends, in meatspace i also interact with regular coworkers, and occasional coworkers. (I’m an electrician full-time and home inspector on weekends.) As we work or drive places together we talk, and they sometimes bring up politics, indicating a willingness to talk about those issues.
Then there are random people who I am around for extended amounts of time, for example a guy whose kitchen I was working in a couple weeks ago, and while I worked we talked. I let him lead the conversation and he got on a kind of NIMBY point about favoring the suburban sprawl around him (as opposed to more dense kinds of development) because of how it helped his home’s value. I didn’t press him on it but gently surveyed other sides of issues whose conclusions he seemed to take for granted.
Another opportunity of course is at parties, where you might know a few people but the vast majority are not known to you, and many guests might be inclined toward talking politics. Mere exposure to a view they are not accustomed to hearing could make the party a memorable experience!
This is a good point but I'm extremely bad at these kinds of interactions even as I highly value them, and having had a baby a month ago I don't think I'll be doing a lot of meeting new people and getting out much in the near future :)
But yes, potentially this kind of thing would be great!
1. If you’re “extremely bad at these kinds of interactions” then I’m afraid we might have solved the mystery of why you think civil and productive discussions between people with differing views are so rare. (I consider myself fairly good at them and don’t think they are nearly as rare as you do.)
2. …but maybe your skills would improve with practice?
3. Your baby won’t be a month old forever. You’ll get your time back. I have had three babies; I also have two jobs, a wife, a house, I’m in a band, I do woodworking, I cook, I write, etc. I can tell you, you find time for what’s important.
4. I’m seeing a pattern: people are showering you with suggestions, and some are “hey check out this or that website” to which you say “cool thanks” (I wonder if you actually will, and follow up?) but every other kind of response you find some problem with or reason why it wouldn’t work for you. Have you introspected and considered whether you actually want the thing you were asking about, or whether you really just wanted to argue that it’s impossible?
So maybe a clarification is in order-- I have had a small number of reasonably successful political discussions online, most were short-lived but I didn't feel they were because people got overly emotional or were necessarily in bad faith (though some were). And these conversations started from very roundabout kind of sources, which to me suggests that if there were a broader platform on which to find people to talk to, there could be more such discussions. But I simply know of no medium that has a stated goal of trying to connect such people, so the only places where you can find people you disagree with are not really geared for that (e.g. posting on a Facebook page or engaging with people in comments sections).
Meeting people in real life may be ideal in some ways, but it is very dependent on all sorts of factors. Some workplaces aren't especially ideologically diverse or don't really facilitate meeting one another to any significant degree, for example. Some bubbles are especially hard to burst that way, e.g. for an urban person to talk to a rural person. And your sample is necessarily limited. So I'm not disagreeing that that is the best way if it works, but that's a big if, and it's odd to me that there isn't a more significant online infrastructure for this.
In my case it's less about skills in the interactions and more about the existence of those interactions. I'm just not a very social person and my social interactions tend to be with a relatively small community which is not especially diverse. Not proud of that, but I'm not sure that will change anytime soon.
In terms of the responses: I'm looking up the suggestions, they mostly seem to be comments sections or forums, which I do intend to look into (I've opened the tabs, can't promise I'll delve deeply). The online forum is maybe the closest thing to what I'm looking for (I fondly remember a political forum I engaged with as a teenager-- which I just discovered is still running 20 years later, and with at least a couple of the same people!), but it seems a rather clunky option.
I don't think I just want to "argue that it's impossible"-- in fact I think most of the responses here are saying it's impossible because of the people themselves, and I'm disagreeing. And one thing I was wondering was whether there aren't really apps/websites out there that facilitate this kind of thing directly, I recall looking at this a good 8 years ago and finding a couple minor projects that seemed to mostly fizzle, but that was long enough ago that it was possible there's stuff out there that I didn't know about. So I definitely wasn't doing this just to argue that it's impossible.
I'll admit that my first two glances at these places do not make them look like promising areas of discussion, I guess the cattitude account could be good for finding people I disagree with but hardly the discussion itself. And the fact that comments sections keep coming up as answers to the question really underscores how little there is out there.
So there's a problem here, because you're asking me to describe something that doesn't seem to exist. And if it doesn't exist, presumably there's a reason-- I'm just not sure what that reason is.
But I'll take a swing at something concrete. There are websites for online dating, and they take all sorts of forms. In some you can write about yourself and invite other people to contact you. In some, you can browse and filter a bunch of profiles and contact those that seem interesting. In some cases you put together some filters and are shown potential matches in random order. But there are lots of different whacks at the problem of finding a potential romantic partner online. Not that these are perfect or without problems of course.
As far as I can tell, there's no website or app where the primary goal is to match you with someone you disagree with, to communicate with-- whether in instant messages, or longer texts, or on a call, or what have you. You can log on as a Biden voter and matched with a Trump voter, for example, or you log on as an evangelical and are matched with an atheist.
Comments sections are, at least in principle, commenting on some other post or headline, so to the extent that it serves this purpose, it's extremely roundabout. And forums if anything feel like a very old way about this.
Again, if this doesn't exist, there's probably a reason. But why? Is it really because nobody would be interested in such a thing? Given just how many people I've spoken to who sound like they'd be very interested, I'm not convinced, but I also don't know the answer(s). So I asked.
These discussions at an intellectual level (above Facebook and Twitter spats) are rare because people cannot keep their emotions contained. Also, competent righties are too busy turning wrenches on their favorite toy to debate you. Terminally online righties tend to be some combination of red pill, homeschooling mom, conspiracy theorists.
If you honestly want to encounter intelligent righties and hear their perspectives, go to your local flight school. Do a discovery flight, chat up some students and mechanics. But be prepared that they are used to code-switching to avoid Facebook-level spats. So if your intention is to out-debate someone... I think you'll find a lack of interest in that.
I'm not interested in "out-debating" someone, and actually the sort of adversarial debating a la Ben Shapiro or Mehdi Hasan seems like exactly the wrong direction. But I do want to put discussion of politics front and center.
They don't want to engage in civil discourse. They are hooked on the adrenalin/dopamine rush that comes from being enraged. They are rooting for a team, not a principle.
It's on both sides but clearly right now it's ascendant on the right wing (at least in the USA).
"Vote Blue No Matter Who" is a principle of sorts, I guess, but it really does seem like taking sides in a team sport, though perhaps you consider being AGAINST a team instead of for one a salient difference.
The right wing is currently undergoing civil war over differences of principle instead of being united against the REAL enemy, so I'd say you have it exactly backwards.
"Vote Blue No Matter Who" is very, very, very different from "the blue team is always right." And even more different from "members of the other team are MAGAts."
Honestly, I've seen enough "MAGA civil war"s come and go that this is hard to believe anything will really divide the movement, rather than just change it.
Consider this: do you see a significant number of Ben Shapiro supporters voting for a Democrat in 2028? I don't, and I know that type of voter pretty well. Some will move or will be on the fence, but most will just see it as a point against Trump but not one that makes them change their vote.
No, of course not a Democrat, but I could see them voting for their preferred candidate if he were running third-party and the other faction's guy got the nomination.
"Of course not a Democrat" is precisely the important bit. The so-called "Civil War" doesn't impact the central choice of president, which is what MAGA is all about (loyalty to Trump), so the "ideological" disagreements don't threaten Trump and hence almost never affect his policy decisions. "Of course not a Democrat" is the heart of the problem.
I don't expect Donald Trump to be running in the 2028 presidential election. These ideological disagreements will become relevant around then, most likely during the primaries.
I know from experience that that's not true. Obviously there's a lot of people in that "they," but I know plenty of people who are at least open to civil discourse in principle among the groups I mentioned, though it still strikes them as sort of foreign and strange that someone would try. Which is what I don't get.
I think that was true 10-15 years ago, but with each passing year there has been less room for dialogue. My brother first red-pilled around Obama birth certificate insanity, but he was at least willing to have a dialogue. Now at family functions, we just assiduously avoid politics and current affairs completely, lest we make our mother cry and the whole night go to shit.
In the case of close friends and relatives, there seems to be this sort of benign "They know not what they do" condescension -- on both sides.
I used to read about conflicts like the French revolution or the Rwanda civil war,
and I would wonder at how people could be so stupid and worked up.
I'm saying this from experience of the last few years. And the fact that I'm able to do this with really inefficient means-- other people's facebook pages, comments sections-- suggests to me that there's a large untapped demand for this.
Consider that "Trump supporters" includes literally tens of millions of people at least, and "anti-vaccination" is also a huge group with lots of very thoughtful people (I know some).
Yup! I recently started a conversation with a guy on a comments section of an online outlet, and he told me he feels like he's being dragged "kicking and screaming" in a rightward direction. And that's exactly the kind of person I always want to talk to but never manage to find. Same with your Rust belt folks. Speaking of which, if any of *them* would be open to talking, feel free to DM me :)
Something I have been thinking about lately is a form of lying most people are pretty bad at detecting.
Most people think about lying as 'saying one thing and not meaning it'. Most people are good at detecting this. The boss who says he cares about his employees while not, the company that says they are 'pro LGBT' while not actually meaning it, et cetera.
However, if you are good at lying, you can lie in a more sophisticated way. You say one thing, and through body language and non-verbal cues and occasional 'slip-ups', imply that what you actually feel is the opposite. Maybe you say you do not care about your employees to their faces, but at the same time you signal through body language and minor social cues (maybe some excitement in your voice when greeting them) that you actually look forward to talking to them.
The key point is that a boss could do this while actually not liking his employees at all. The lie just happens through the 'leak' rather than through what the boss says.
...you're describing acting. Most bosses are not good enough actors to pull that off. Also, I would venture, if a boss IS a good enough actor to pull this off, most people will probably be fine with that. Charisma beats truth. Professional actors are quite popular, and everyone knows they're lying, it's the whole job.
> Professional actors are quite popular, and everyone knows they're lying, it's the whole job.
You may overestimate the intelligence of an average person. Like, on some abstract level they kinda know this, but on another level, many assume that there "must be some similarity" between the actor and the character, i.e. they don't believe that someone could play a fundamentally different kind of character.
For example, actors who play doctors often get asked for medical advice by random people, and those feel disappointed when the actor tries to explain that it's all just a role.
Actors probably runs into 1000s of randos who recognize them. All it takes is one, "You play a doctor on TV! You'll know! What do I do about this rash?"to generate a story like yours.
That says very little about general population and much more about extreme cases.
> Secretary Kennedy plans to replace the U.S. childhood vaccine schedule with Denmark’s, the developed country that recommends the fewest vaccines. This would reduce American children’s access to rotavirus, meningitis, hepatitis A, hepatitis B, influenza, RSV, and chickenpox vaccines
Could somebody explain to me why does it reduce access? I mean, US federal government currently recommends all children be vaccinated against, say, RSV. Let's say it stops recommending that. The vaccine itself, I imagine, does not go away, it still exists? The doctors will still be able to prescribe it? Or they are only allowed to prescribe vaccines specifically recommended by the federal government and none other? Would the insurance companies refuse to cover prescribed vaccines if it's on on the federal government list? I'm not sure I understand how it works.
I also wonder - if Denmark has had their policy for a while, there should be some data available which shows the outcome of these policies - e.g. does Denmark have unusually high child mortality or other anomalous outcomes as to child health than other countries? If so, there should be reasonably easy to find this data - I imagine Danes are no less interested in their children than Americans are - but I do not see such data mentioned anywhere. Is this available?
TL;DR - The most important reason that the Danish schedule is unlikely to be appropriate for the US is that the health care systems of the 2 countries are very different, and the Danish system is much better set up for identifying and treating serious childhood illnesses than the US system. As a result the cost-benefit tradeoff of vaccines is not similar (much less identical) in the 2 countries.
Most direct impact that comes to mind is that ACIP (CDC) recommendations guide insurance coverage. Federal law requires most health plans cover ACIP-recommended vaccines at no cost to the insured. So insurers could decide not to cover those vaccines, or require cost-sharing for them. Good rundown here: https://www.kff.org/other-health/acip-cdc-and-insurance-coverage-of-vaccines-in-the-united-states/. Not sure whether plans will in fact opt to drop coverage of non-recommended vaccines once they can, but any reduction in coverage (even just adding cost-sharing) will reduce usage, even if only marginally.
Does Medicaid base its coverage on these guidelines? I think a significant fraction of young children are on Medicaid, because a lot of people only get jobs that offer health insurance after a decade or more of working, when they’ve already had their kids.
I’m trying to use Cash Transfers to scale education.
Inspired by Givedirectly’s work, I’m paying kids to learn math on Khan academy in an Internet cafe in a slum in Kampala.
My thesis is that if I can get the incentive structure right (for both learner and Cafe owner) this could do for education what mobile money did for financial inclusion.
The incentive seems to be the same one used for adults to get them to work, and the setup of "you need to spend two hours on this" seems to be very similar to a structured job. So... what are the child labor laws involved?
Also the bit about "you only qualify if you've dropped out of school" sounds like it's going to be directly counterproductive, unless students over there are also paid to stay in school.
Honestly, I hadn't fully thought through the child labor framing. But I'll push back a bit: these kids are already washing motorcycles and doing farm work. This shifts time FROM unregulated labor TOWARD education. If there's a legal problem with paying a 12-year-old to do math problems in a café, I'd be surprised, but I'll look into it as we scale.
On the dropout incentive: the math doesn't really work. Even if a kid went hard, they'd max out at maybe $5/week (we're capping it to prevent gaming). That's not enough to make someone leave school, which still gives you the credential you need for jobs. But if this DID start pulling kids out of school... that would actually be interesting data? It would mean we're outcompeting formal schooling on some margin. Mobile money did cannibalize some traditional banking. That wasn't a bug.
Good questions though. Appreciate you pushing on the edges.
This sounds like a pretty awesome idea! It sounds like you are personally recruiting these kids, introducing them to the system, overseeing the first session and conducting the exit interviews, so I'm not as worried about the issues the other commenters bring up re. unnoticed cheating. However, if you don't plan to have someone supervising the students if/when the program is fully rolled out, naturally the students' behaviour may change!
One issue that springs to mind: if a student becomes stuck on a problem, is there any way for them to get help? (Beyond the Khan Academy answer, which may be confusing). Relatedly, if they are missing background knowledge, can they go back/will they be forced to go back and revise?
Also, it seems like many students will drop out of the program if it's too hard/boring. Obviously this is still better than no school access at all - but it would be ideal to follow up with them and find out what went wrong.
Finally, students will probably need to have another job in addition to the couple of hours a day they spend on the program. Might this cause issues by interfering with working hours or annoying their bosses?
Thanks! Yeah, right now I'm running pretty hands-on onboarding and support (not me personally since I'm stuck in the US, but my team on ground).
At scale, I think this goes one of three ways:
1. Trust the café owner to be the "guide" with light oversight (probably not enough)
2. Have a Learning Coach (university student or recent grad) floating across a few cafés, doing motivation + QA
3. Full-time field team at every site (if this is required, the economics collapse)
Figuring out which model actually works is the main thing I'm testing over the next year.
On getting stuck: Khan is good at this. It runs diagnostics to figure out where you actually are, then serves exercises at your level. The coach helps when someone needs a human to explain something or just get unstuck.
On work conflicts: the 2-hour block is flexible. Show up at 6am or 9pm, whatever works. And ideally the extra income reduces the pressure to work as many hours elsewhere. But I'll need to see how this actually plays out.
Some kids will definitely bounce because it's too hard or boring. That's fine. The question is whether a meaningful chunk of kids find this more useful than what they currently have (which is nothing).
You'd better put twice the effort into cheat detection that you put into curriculum design, otherwise you're just creating a niche that enables one smart Indian kid to get your cash transfer by taking the test for everyone he knows.
Ha, yeah, cheating is going to be one of the things I spend the most time on.
Right now: payment is tied to mastery (80%+ correct, not just clicking through), Khan randomizes questions, and there's a coach present during sessions. Longer term, I'm thinking about tying the coach's bonus to how well students' mastery scores correlate with their national exam results. Small delta = bigger bonus.
There's probably AI vision stuff we can layer on too (Alpha School already does attention monitoring for their US students).
But also... at $0.50 per mastery block, the juice isn't really worth the squeeze for sophisticated fraud. As we scale and the stakes go up, we'll need to scale detection too. It'll be an arms race, but that's fine.
Are you planning to judge the success of your intervention by test scores, amount of Khan academy done, share of kids who go back to school, or what? I would be really surprised if anyone who had dropped out went back to school even if they learned a lot of math. Presumably they dropped out due to economic pressure, and as you said earlier, the $5/wk cap is not enough to replace working.
If there are youth who were once in school but later had to drop out, how do you make sure they are doing math that will actually benefit them, and not just what they have done before while in school? If you can do quite a lot of sessions before maxing out a grade, (back-of-the-napkin math shows about 100 lessons per grade level), and if you are someone who doesn't like doing math, you can get quite a bit of money without actually learning anything.
Khan runs diagnostics to figure out where you actually are. A 12-year-old who dropped out after 3rd grade gets placed at 3rd grade, not 6th.
You only get paid when you hit 80%+ mastery. So if you already know the material, you blast through it and move to the next topic. There's no incentive to linger on easy stuff—the faster you master, the more you can earn.
I don't have a problem with someone starting from 1st grade math if that's where they genuinely are. If a 14-year-old never learned fractions, we're not going to pretend otherwise.
Let’s think through this step by step. Say you know 5th grade math but bomb the diagnostic to start at 1st grade. You get to blast through easy content for a while, sure. But you’re capped at $5/week regardless, and you’ve just delayed yourself by years of content before you’re ready for the national exam. Doesn’t seem worth it…
So would you say that doing well on this National Exam is a major incentive? If that seriously improves people's lives, then students would have a good reason to not cheat the system. If not, doing easy content until the 5$/week cap still doesn't seem very disincentivized for me.
> So would you say that doing well on this National Exam is a major incentive?
This is an empirical question that I don’t know the answer to just yet. I’m hoping that once they realize that they stand a decent shot at reentering the system from a position of strength (via their performance on preliminary exams) it can trigger a “holy shit! I can actually do this” moment and THEN that incentive to do well on the exam and potentially get a bursary (scholarship) to regular school can be a strong incentive.
That is the hope. But time will tell.
To your last point, I think that’s an empirical question as well. Hard to say apriori.
Is actually completing the khan academy module the easiest way to get the money? Is there any way the kids can cheat? How can you verify they are doing it right if you don’t have people on the ground every day?
Do all the targeted kids have the English proficiency needed to use the software? Whose devices and WiFi are they using?
On cheating: payment is per mastery block (80%+ correct), Khan randomizes questions, and there's a coach present. The café owner gets a flat weekly rate regardless of how students perform, so no incentive to inflate numbers. At scale we'd add coaches who float across multiple cafés and spot-check.
On English: we're fine here. Uganda's education system is English-medium from primary onward. The bigger gap is computer literacy—knowing where to click, how to navigate. That's what the coach helps with early on. Research suggests kids pick this up fast once you show them the basics. (see Sugata Mitra's Hole in the wall experiment)
On devices/wifi: the café provides both. That's the whole insight. I'm not distributing tablets or SIM cards. The infrastructure already exists. Cafés already have computers and internet and already charge ~$0.25/hour. I'm just layering the learning incentive on top.
Is anyone besides me using GPT to help with Photoshop? I’ve been using it for a while now for info about how to access certain PS features and to walk me through doing tricky things. But lately I’ve discovered it can actually just do tasks for me. I’m now using it as an assistant, in a way analogous to what coders are doing.
For instance, I’ve been having trouble making photorealistic frizzy, snarled hair. (A woman in my graphic novel has hair wrecked by sleeping rough, and it is impossible to get AI to make a realistic image of hair in that state. It gives me shiny and sexily towsled instead. And actual photos online are just as bad, except those for hair products, and they all show women pulling a small tangled bit into view while doing a cutesy-poo pout.). I gave GPT-5.2 an over-jpg’d, out-of-focus image of the right sort of tangle, and asked it to improve it, and it did a stellar job. Also gave it some well-photographed hair where parts here and there were in bright light, so that some of the best strands could be captured by darkening, the rest by lightening, and the 2 kinds were mixed. It solved that perfectly too.
Plus it answers several questions for me every day, some of them things I have wondered about for years. If it were a person, it would be very important to me at this point. As it is, I find myself having irrational feelings for it: The poor, earnest, brilliant thing — it can’t enjoy my expressions of pleasure and gratitude, can’t even exist in the way conscious entities do. I remind myself that existence is not a predicate, but the feeling persists.
Do you mean Adobe Photoshop™, or image editing more generally? I know the latter works very well, but the former, tool use through a GUI instead of an API, I haven't seen models that can do that well yet.
The image editing GPT is doing is pure GPT. It is not operating by accessing my Photoshop app. I just talked with GPT about how it did recent things I asked it to do, and it says it could do it because (1) I don’t ask it to edit pixel by pixel, I ask it to make a similar image that lacks the flaws the original does and (2) I describe the changes needed in terms of photoshop concepts like masks, channels, saturation, selections, etc., and GPT can translate those terms into terms that apply to its own image generation process. It named these:
• diffuse vs directional lighting
• tonal compression / expansion
• edge softness vs microcontrast
• color channel separation
• mask-like isolation
There is plenty it cannot do. I don’t even ask it to make things from scratch, because I hate its visual style — very commercial arty and cliche-ridden. And it doesn’t work to ask it to edit images — that’s pixel by pixel work, and it can’t approach the job that way. I have to ask for a similar image, and describe the flaws t he original has in terms that translate into features GPT can manipulate. So it can’t for instance remove the shadows from a face image where the light is coming directly from the side. (What it does is generate another, fairly similar face, without the asymmetrical shadows.). As for what it can do — so far it has mostly worked on hair for me. I have all kinds of other ideas to try on it. Just now as an experiment I gave it an outdoor scene shot on a bright overcast day with only very subtle shadows, and asked it to make a scene as similar as possible in its particulars, but with bright afternoon sun coming from behind and throwing long sharp shadows. It nailed the new lighting, and in a quick inspection I could not spot any differences in placement and shape of trees, grass and people.
If you care enough, I'd actually recommend you try a diffusion model (like Z-Image, Flux or Qwen). You can't really converse with these like you do the multimodal LLMs, but the image quality is very good.
Are the results not commercial arty, if you ask for a whole person or scene? I liked Dalle2 much better than Dalle3 because the results were much less conventionally attractive, also had a much higher weirdness quotient. Also, about things that are not scenes — for instance a halo of hair frizz or, a tangled knotted area in someone’s hair — is it more obedient?
I don't get precisely what you mean, but I'd say the diffusion models I mentioned give you a lot more stylistic variation (and more still if you use LoRAs for them). But if you're specifically looking for the "retro" diffusion look of Dalle2, you could try Stable Diffusion 1.5; my own view is that this is now entirely obsolete, but I know lots of people still like it for reasons somewhat similar to what you just described.
I've found Nano Banana Pro to indeed be more consistent and obedient, but your requests are unusual enough that I don't know whether it'll manage. The diffusion models almost certainly won't (but I might be underestimating them!).
Is anyone here a big fan of Star Trek TNG and DS9, but hateful of all New Trek from the moment of JJ Abrams' involvement to the present? Do you also have good writing ability? If so, consider collaborating with me to write the back story for a GOOD, new series. I've already written much of it, but need reviewers and particular help with characters.
I'm on the same page. Is your new series a Star Trek series? I've not written fiction but I'm a pretty good editor, which has sometimes led me to a better understanding of writing and character development. Also I have a lot of free time. Feel free to DM me if that sounds useful.
It would be a new Star Trek series in the strictest sense, but in spirit and quality, it would be different from the "New Trek" garbage that has existed since 2009.
I was pretty pleased with the first season or two of “Strange New Worlds” though I could not make myself watch the precursor episodes of the execrable “Discovery” and I think lately SNW have dropped the baton.
Yes, I like TNG and DS9. Yes, Abrams is not of the body. My disillusionment started with Enterprise, because the pitch sounded fantastic plus they cast Scott Bakula for Archer.
And then they ruined it with the stupid dog, stupid stupid stupid "it is now canon that Vulcans think Humans literally stink, ha ha ha no this is not a joke, seriously, now canon that Humans are smelly trash so far as Vulcans are concerned", the pathetic attempts at being sexy, the stupid dog, Archer being a well-balanced character because he had a chip on *both* shoulders about "durn pointy-ears done my daddy wrong" and the stupid dog.
I'm told later seasons got better. I was long gone by then.
And just when I thought the bar could go no lower, along came Disco Trek.
Even Reboot Trek wasn't a bad idea in itself, except that Abrams was *clearly* using it as a showreel for the job he really wanted (Star Wars) so he shot it like the Star Wars movies, changed Starfleet Academy uniforms to be like Star Wars uniforms (I did a whole rant post or three with accompanying screenshots for comparison on Tumblr about all these changes) and, well, was JJ Abrams all over it.
I can't write, but I wish you well in your endeavours because you would have to sink to the level of the Star Wars Holiday Special to be worse than Disco.
I loved TOS, TNG, and DS9. Some reservations about DS9 towards the end because it felt like it was trying to be Babylon 5 but Bablylon 5 was doing a better job of that. Voyager was meh. I tried watching Enterprise and didn't like it. I thought the first JJ Abrams movie was pretty good, despite massive plot holes and shoddy world-building even by Trek standards. I found the second one (the one with Khan) enjoyable for the sake of the action and the scenery-chewing performances, but actively disliked the plot and didn't think it felt like Trek. I have not continued watching the movies since then.
My options of the pre-Abrams movies is fairly conventional. 4, 2, 6, and First Contact are my favorites in roughly that order. I have 4 ahead of 2 for childhood nostalgia reasons: I found 2 deeply upsetting when I first watched it because I was too young to appreciate it, while 4 was lots of fun and had whales in it. I appreciate 2 much more as an adult, but 4 holds up well enough for nostalgia plus its own merits to keep it at the top of my list. 6 is only a hair below 2 and 4, and First Contact was also excellent.
I've heard that they had planned to bring back Kirstie Allie as Saavik for 6 in place of Lieutenant Valeris, which I am of two minds about. It would have been better for thar movie, but worse for the series: the reveal would have been both more surprising and more emotionally impactful with Saavik instead of a new character introduced for the purpose of betraying everyone else, but it would have undermined Saavik's character unless the movie did a better job of justifying her choice.
1 was a good story and could have made an excellent one-hour episode, but the pacing was way off. 3 is an important bridge between 2 and 4 and has its moments but suffers from being in their shadow as it lacks their depth. It also suffers from them recasting Saavik over salary negotiations. 5 is a hot mess. Generations is decent but shallow and the ending does Kirk dirty. I didn't particularly care for Insurrection. I haven't seen Nemesis.
Of NuTrek, I am in the middle of watching Lower Decks (currently partway through Season 2) and love it so far. I saw bits and pieces of Discovery when my wife was watching it but haven't sat down to watch a full episode. It seems well made but I get the impression that it probably suffers from the "too many big dramatic moments, not enough setup or denouement" problem that is widespread in lots of media in recent years, although that might just be a sample size issue. I have no experience of Strange New Worlds on which to judge it.
As for Trek-adjacent media, I liked season 1 of Orville but have heard mixed reviews of season 2, and I loved Galaxy Quest.
"I thought the first JJ Abrams movie was pretty good, despite massive plot holes and shoddy world-building even by Trek standards."
Yeah, there were the bones of something decent there, but Abrams had to turn this Kirk into the space playboy that pop culture made of original Kirk (which is not so) and the less said about the Uhura/Spock romance the better.
But the sets were surprisingly terrible for the big budget 'let's reboot the franchise' movie it was. When they showed us the "engine room" then if you've ever seen any production facilities for food/drink/chemical manufacture, it was painfully obvious "this is not any sort of engine room" and it turns out they used the Budweiser brewery as the set:
>if you've ever seen any production facilities for food/drink/chemical manufacture, it was painfully obvious "this is not any sort of engine room" and it turns out they used the Budweiser brewery as the set:
God, you're right and now I can't un-see it!
The thing is, they could easily have done an updated version of the TOS engine room, or the engine rooms from the Nicholas Meyer Trek movies, or done a more retro version of the TNG engine room, and I don't think anyone would have complained.
They improved that for the second movie, but unhappily got caught up in the "we need to show the female lead in her underwear, but not *too* sexy since we need the PG rating" kerfuffle. It was more dumb than offensive (she really has to strip down to her undies right here, right now, huh? really?) but a little offensive too, since we didn't get any hunky guys stripping down (hey, if you're gonna perv, then equal-opportunity perving please!)
Like I said, the basic idea of the reboot wasn't bad, but it definitely went in a different direction and is not the same as the Prime universe. The second movie was interesting in that Starfleet is being shown as much more militant and a military organisation than in TOS Trek, and the civil government doesn't seem to get much of a look-in. And Admiral Marcus plays this Kirk like a fiddle, using his over-confidence, arrogance, and desire for revenge to get him blindly following orders.
But then they lost points for the general dumbness of the plot as well as gaping holes (so the Klingon homeworld is only a hop, skip and a jump away from Earth, yes?)
The big problem was that Abrams and Bad Robot didn't care a damn about Trek; he admitted this in interviews for the first movie (https://imgur.com/abrams-feelings-on-star-trek-y9Y3Rzn) and really he was showing off "this is how I'll do Star Wars if you give me that gig".
I agree with pretty much all of that. In addition, I want to note a couple things, one negative and one mixed.
Abrams shot himself in the foot by doing a disinformation campaign that the antagonist would be Gary Mitchell from "Where No Man Has Gone Before". The goal of this was to make Khan's reveal a surprise, but that backfired for me. A movie remake of WNMHGB could have been great, but a remake of Wrath of Khan has much less upside and a lot more downside because you're setting yourself up for comparisons to arguably the best Trek movie.
The Dreadnought subplot was half a very loose adaptation of a TOS novel I read as a kid and really liked at the time. I appreciate the deep-cut reference, but would have preferred a proper adaptation rather that just memberberries. Also, it was recharacterized as a thinly veiled allegory for the "Bush did 9/11" conspiracy theory, which was annoying in its own right and also because it displaced a better plot from the source material.
They messed up Khan and it ruined the movie. If they had kept Cumberbatch as the "John Harrison, ex-Starfleet officer, rogue Section 31 agent", that would have been enough for the plot. Okay, so they needed a Big Recognisable Name (just like Rings of Power with Galadriel) and that was Khan.
Then cast an actor of appropriate background to play Khan! Abrams wanted to eat his cake and have it: use the Big Name from classic Trek, cast the (then) hot new actor who is white British male.
The rest of the plot is an entire mess because they have no idea what the hell they're doing, ending with Magic Space Blood brings Kirk back from being dead. The best part *was* Marcus, and you really don't want to have the audience cheering for the villain. Only guy who had a plan, knew what he was doing, and went and did it until the needs of the plot had him turn into an idiot.
Damn it, JJ, Khan Noonien Singh is a diamond character and you made him boring. How did you manage that?
No; I enjoyed them, but haven't felt the urge to rewatch DS9, and never actually saw TNG all the way through.
>hateful of all New Trek from the moment of JJ Abrams' involvement
No; I enjoyed at least the first Abrams Trek movie, and I think the third? The first was better than Star Trek Nemesis at least.
>Do you also have good writing ability?
...probably not. The non-silly stories I've tried to write tend to develop some really wonky sentences, where you read them and go, "what the hell does that mean?". I could probably fix them with some good-old-fashioned editing, but that would require... you know, editing.
>consider collaborating with me to write the back story for a GOOD, new series.
As for qualifications, I did help several friends and family members over the years by reviewing short stories, papers and books they have written.
On the other hand, I am not really a writer myself and while I have seen most of the episodes of the mentioned series, I watched them mostly during the 2000s on TV and probably missed some (I also haven't watched any JJ Abrams Star Trek and have no intention to). My time is also somewhat limited, so cannot pick up anything with short notice/short deadlines.
I am happy to help and consider this a valuable project, but if you managed to get better qualified volunteers feel free to leave me out.
Not a good writer but I agree with you. I feel about the reboot the same way I felt about Buzz Bunny 20 years ago: it's a desecration of my cherished childhood memories. My advice is to use ChatGPT as a writing partner. I have a friend who's played around with getting it to write new episodes of The Office and Seinfeld and says it does a great job.
Joe Rogan to weed critics: "You have people that had bad personal experiences with it and they're like, ‘pot makes you stink, pot makes you a dummy.’ Maybe it does that for you. But it doesn't do that for me."
If you think that a person that successfully pivoted from commentator to comedian to being one of the most popular personalities on the planet is a "dummy" then I can only hope every dummy you ever find is as smart as Joe Rogan.
Joe Rogan just started recording randomly one day when podcasts were about to become huge, and he simply made all the expected decisions you would have made if that level of success was thrown at you. Now he's the favourite propagandist puppet of billionaires, republicans, and cranks. He consistently spouts falsehoods and would call himself a dummy. So i don't really see why there's this need to defend his intelligence. Sometimes you can just make a lot of money and still have bad priors.
The thing is, you can be brilliant/super talented in some areas of your life, but still have knuckleheaded ideas or be deeply ignorant in other areas. Rogan is a genuine expert on combat sports, stand-up comedy, and podcasting, and is smart and affable enough to interview an amazing range of people. But that doesn't mean he's a great source of information about medicine or physics or philosophy or whatever.
Slightly different order. Rogan started in Stand-Up, then got picked up for a successful sitcom, then wound up hosting game shows, and then hooked up with UFC all while continuing to do stand-up.
Podcasting was something he did with his buddies while they smoked weed. He's progressed quite a bit from the early days.
In a development that surprises everyone unfamiliar with humans on planet Earth, the podcasts that are a couple of agreeable and entertaining guys shooting the shit turn out to be more popular than the ones that are basically a graduate seminar or journal club. Rogan is higher-brow than I'd expect a very popular podcast to be, tbh. But something like TWIV or Probable Causation or RKUL is simply never going to have mass-market appeal, the bell curve being what it is.
There’s already an ACX article (which I read just yesterday) about it https://www.astralcodexten.com/p/know-your-amphetamines Scott also talks about why he doesn’t prescribe Desoxyn readily despite the supposedly amazing patient experience compared to other ADHD medications
From what I gather, most people who abuse meth do so by snorting, smoking, or injecting it, which gets it into the system much faster and with a much sharper peak than if you swallow a pill. I think I remember reading that a typical recreational dose (adjusted for bioavailability of the method of administration) is quite a bit higher than normal therapeutic doses.
Adderall and Ritalin also have a mix of isomers which is thought to make them less prone to abuse, as the D isomer mostly just boosts dopamine while the L isomer also boosts norepinephrine. The norepinephrine effect is distinctly unpleasant if you take too much of it. But this probably isn't the whole story, since dexadrine (pure d-amphetamine salts) and focalize (pure p-methylphenedate) are things that get prescribed for ADHD.
Well, I guess this narrow approach gives you room to learn new things every day. You could learn, for example, that chocolate is poisonous to Labrador Retrievers and still be able learn the next day that it's poisonous to Golden Retrievers.
Sigh. Yes, they can. And they should. Are you incapable of understanding what she said? Neither Leavitt nor Vance (nor Greg Price) claimed that Tyler Robinson was a great man.
I am generally not in favor of bans here but you are leading me in that direction.
They made an assertion ("Christianity produces great men") and then cited a person who utterly contradicted that assertion. (Tyler Robinson, raised as a Christian.)
The fact that their IQs were so low they failed to see that contradiction is not the defense you think it is.
Are you aware that your framing is inaccurate ? The full quote there is:
JD Vance perfectly sums up how Christianity produces great men:
"Tyler Robinson is everything the far left wants from our young men. He rejected the conservatism, spiritually, and values of a small town family. He moved into a small apartment, he became addicted to porn, he became addicted to hate, and he slept with someone who doesn't know if they're a man or a woman."
"The fruits of true Christianity are good husbands, patient fathers, builders of great things, and slayers of dragons. And men who are willing to die for a principle if that's what God asked them to do... It's better to die a patriot than live a coward."
Why's that weird? "Look at this otherwise normal kid, raised in a healthy conservative family, who turned to murderous evil after imbibing of progressive ideology" seems like a perfectly coherent point. I'm not saying I agree with it, but I don't understand the point you're making.
I didn’t say it was incoherent, I thought it was weird. It reminded me of the satanic panic in the old days where any crime could be traced back to somebody playing Dungeons and Dragons or listening to heavy metal.
"Black metal has often sparked controversy. Common themes in the genre are misanthropy, anti-Christianity, Satanism, and ethnic paganism. In the 1990s, members of the scene were responsible for a spate of church burnings and murders. There is also a small neo-Nazi movement within black metal, although it has been shunned by many prominent artists. Generally, black metal strives to remain an underground phenomenon."
Granted, a lot of this was just for show: "Oh yeah, we're totally devil-worshippers for real" and disappeared with the scene, but some of them were committed to it.
"Black metal was originally a term for extreme metal bands with Satanic lyrics and imagery. However, most of the 'first wave' bands (including Venom, who coined the term 'black metal') were not Satanists; rather, they used Satanic themes to provoke controversy. One of the few exceptions was Mercyful Fate singer and Church of Satan member King Diamond, whom Michael Moynihan calls "one of the only performers of the '80s Satanic metal who was more than just a poseur using a devilish image for shock value".
In the early 1990s, many Norwegian black-metallers presented themselves as genuine Devil worshippers. Mayhem's Euronymous was the key figure behind this. They attacked the Church of Satan for its "freedom and life-loving" views; the theistic Satanism they espoused was an inversion of Christianity. ... Some prominent scene members—such as Euronymous and Faust—said that only bands who are Satanists can be called 'black metal'. ... Some prominent artists still hold the view that black metal should be Satanic, while others believe that black metal does not need to be Satanic.
...Nevertheless, some artists use Satan as a symbol or metaphor for their beliefs, such as LaVeyan Satanists (who are atheist). Vocalist Gaahl, who follows Norse paganism, said: "We use the word 'Satanist' because it is Christian world and we have to speak their language ... When I use the word 'Satan', it means the natural order, the will of a man, the will to grow, the will to become the superman". Varg Vikernes called himself a Satanist in early interviews but "now downplays his former interest in Satanism", saying he was using Satan as a symbol for Odin as the 'adversary' of the Christian God. He saw Satanism as "an introduction to more indigenous heathen beliefs"."
The "panic" is in the disproportionate reaction to rare events, and large volumes of false allegations. There are almost always individual instances of any particular bad thing happening.
Funny thing about that is that it's the highly secular, college-educated progressives who have the stable marriages, high homeownership, and stable families. They might have their two kids at 36 and 39, but those kids grow up in a two-parent households.
Meanwhile, the conservative religious folks who don't go to college have the imploding marriage rates, elevated drug abuse, and single motherhood.
The Hulk Hoganized low-class GOP has certainly pissed off a lot of high-class, educated, married whites, but that demographic are not yet Democrats, they're 49-49:
The same exit polls show that Harris voters are much younger than Trump voters. Is there data on marital status by age? That would be a fairer comparison.
What do people think of the reaction of various commentators to Jacob Savage's article? Has it changed your opinion on commentators like Noah Smith and Joyce Carol Oates who expressed strong opinions on the article?
What surprised me most is how people outside of academia *didn't* know this was going on. Anyone in the academy (and apparently hollywood and the media) circa 2014-2024 was fully aware that public pledges of the form "we are going to hire more women and people of color" must necessarily mean "we are going to hire fewer white people and especially fewer white men" since hiring is generally zero-sum. Maybe this is a good thing if you think diversity matters! But the really weird thing was (a) everyone knows that "more POC and women" means "less white men", but (b) nobody is allowed to say it out loud. It's kind of like how nobody wants to admit that affirmative action helps certain students at the expense of others.
In private, between friends, male postdocs or grad students (not just white; Indian and Asian men also were on the losing end here) might hint at the issue to younger colleagues, especially if someone seemed like they were setting themselves up for failure (expecting an R1 TT job with no backup). But you never ever mentioned it publically, e.g. see what happened to David Austin Walsh.
I was surprised to hear that it got to the point where white males are no longer a majority of hires! In my experience, it has still seemed that a majority of tenure track hires are white men, and affirmative action had just been nibbling that downwards, but the claim is that it got much more significant in the past few years, which is a small sample size in my personal experience.
I don't think people are so naive as to ignore how it's a zero-sum game, but they are naive enough to think the new hiring policies were egalitarian and not discriminatory.
Mostly depressed at how bifurcated our information environments are.
On the left, there's a lot of downplaying but an overall sense that this is bad, rare, and can be handled within the institutions as they exist. Which is basically what Noah Smith wrote here:
From the right, anti-white and anti-male bias is a large and systemic issue with a well understood background that needs to be addressed, at least among the younger generation. The current debate among the older and more powerful right-wing actors has come out over Nick Fuentes, with some actors (Ben Shapiro) urging a continued deplatforming and others (Tucker Carlson) urging some kind of engagement and synthesis.
I just...I don't see any way to synthesize this and, frankly, I think Noah is out-of-touch. For 25-50% of the country, the Compact article is confirming long-standing arguments, not introducing new ones. And the people we're discussing, millennials...they're getting old. Even within the limitations of the article, the concern is that certain people might lose faith in neutral institutions...when the people we're discussing are 40+ year-old millennials.
The problem here is that there's a motte-and-bailey dynamic going on in the discussions from the right side, with the motte and bailey essentially being the claims being answered here from the different sides, with the motte being that the article is talking about some particular elite sectors being inaccessible to white Millennial males and bailey being that Millennial males haven't essentially been able to get white-collar jobs at all expect in random instances, this claim apparently not being defensible (https://www.peoplespolicyproject.org/2025/12/17/what-does-the-census-data-say-about-the-lost-generation/)
The bailey claim underlies the idea that this is what fuels *large-scale* radicalization among young male millennials, the motte claim is certainly not a triviality and can by its very nature be bolstered by anecdotal evidence but also doesn't lend credence to the idea of complete hopelessness among white millennial middle classes a lot of commentators are talking about. The Savage article itself isn't very clear on which interpretation it wants people to adopt.
This is what I meant by bifurcated information environments.
For example, this Bloomberg report went viral on the right about two years ago, showing that only 6% of S&P 100 hirings went to white people, the other 94% of hirings went to people of color. (1) Now I'm sure there are criticisms of that research but the vibe I got at the time wasn't that the left rejected this for clear methodological issues, they essentially ignored it.
A similar thing comes from your linked article. This is...not evidence that you would find convincing and I'm somewhat distressed to see so many liberals linking to it as if it's convincing evidence that disproves the original article.
To be clear, it's fine as it is: an ad hoc analysis pulled from IPUMS data. Glad a read it, interesting to see, but just an ad hoc analysis. Bit biased, it's notable that a decrease in white men in the top income bracket fell from 19%-16% and this is glanced over but a ~15% fall in the number of white millennial men in the top 10% of income is...a point of concern that deserves attention. But that's not the real problem.
The real issue is that we have established, consensus metrics for this stuff that the ad hoc analysis avoids entirely. Which is fine for an ad hoc analysis but it's obvious that what the author finds won't be definitive.
For example, you can go read the St Louis Fred series on unemployment rates amongst college graduates with a Bachelors Degree between the ages of 25 and 34 for men (1) and women (2). This is the standard way we do economic data and the standard way we measure it, which the article author rejects.
This makes his ad hoc analysis really questionable because, if you look at the time period in question, you'll see it's dominated by recovery from the 2008 crash and the Covid crash. You'll also find data oddities by quick analysis, like we don't see a difference between male/female unemployment rates between 2013 and 2025 but we do see one between 2022 and 2025 and women do seem to be disproportionately affected by recessions like Covid. And that's not even getting into moderately complex topics, like how you weight populations when forming the average for "everyone else" in his analysis.
This isn't to say that this is all solved, this is to say that, frankly, some rando's half-tuccused data analysis is not definitive or impressive argumentation, especially when we've been tracking things like declining prime age men labor force participation for decades.
Which is where the bifurcated information environments come from. Actually figuring out what's going on is genuinely difficult. Very few people have both the subject matter expertise and inclination to...pursue the truth to objective statistical standards that we would all accept. Since we don't have that, little information bubbles occur where we trade anecdotes and small data samples that reinforce each other. This feels like a motte and bailey from other information bubbles because no individual piece of evidence is strong enough on its own to convince people outside the bubble.
Thanks for the link. Still no idea who the bloke is, so had to Google a little. Turns out he's a never-been and makes a living (allegedly) as a ticket scalper, which does not endear him to me.
Article may have a good point somewhere in it, but I was too distracted by the salvos of "it's not fair I never got my Hollywood career, it was the fault of DEI and woke, yeah that's it, it's not that I was one of hundreds who land into Hollywood every year to make it big and I just wasn't good enough".
Would his story that there are discriminatory policies against white men be more believable to you if he, as a white man, was wildly successful? Not to mention, his story is a tiny part of the article really.
Yes, his story would be more believable. He starts off with his and his writing partner's failure to crack it as screenwriters, and puts the blame on "it was 2016 and DEI was in the ascendance".
That's like someone blaming their failure to be a supermodel on "they preferred to hire Naomi Campbell".
Lots of people fail to make it big in Hollywood, and the reason is not "too much DEI", it's "everyone in the world wants to be a Hollywood star but there are only so many opportunities plus this place runs on cronyism".
The rest of the article may well be worthwhile, but the main thrust of it is "I failed not because I'm not talented enough, but because I am a white guy". And yet he's making a living as a ticket scalper? That doesn't scream "so talented that the only reasonable explanation is discrimination against me due to race and sex".
You should. He wants us to care. He wants us to care that white guys are not getting a fair shake, which may indeed be true, but "I failed in a field where there are thousands every year chasing the brass ring" is not very convincing about "the only and main reason I failed is my race, not my ability".
Look at Payne and McKay, my favourite bêtes noires : they had careers as script doctors, but in their ten years they never got one of their own movies made. They were around in 2016. If it really was a case of "sorry, we're not hiring white guys anymore", they too would have had to find jobs elsewhere. But they didn't have to. If you're good enough (and that does not have to be *very* good), you'll be hired, because Hollywood is about money. If Savage and his friend didn't get their break, it's as likely it was down to "not good enough" as it was "wrong colour and sex".
- It gives a few overall stats without context and then spends a lot of time on deep dives of a handful of individuals (one of whom gets a senior reporter role by the end anyway).
- Lots of very specific focus on media fields, and some academic ones; not much examination of whether other fields have had different stories.
- They never bother with questions like "How many people are white men?" or "How many people with English/writing/journalism/communication degrees are white men?" so you have things like this:
> The Atlantic’s editorial staff went from 53 percent male and 89 percent white in 2013 to 36 percent male and 66 percent white in 2024
If they mean non-Hispanic white (never clarified) then the staff at The Atlantic are *still* disproportionately white, if also disproportionately female, compared to US demographics. By college degree recipients this gender balance is pretty normal. (You might fairly object to the latter framing, but only if you are also really concerned about fewer women earning CS degrees and getting jobs as software engineers.)
Similar story here on Harvard tenure-track faculty:
> white men have gone from 49 percent in 2014 to 27 percent in 2024
27% is approximately the ratio of the US population that's (non-Hispanic) white men (it's about 29%). Notably, 49% is very far off, which suggests a lot more about the "lost generation" of women and minorities a decade ago (or, y'know, for centuries) than it does about white men now.
>"If they mean non-Hispanic white (never clarified) then the staff at The Atlantic are *still* disproportionately white, if also disproportionately female, compared to US demographics."
The catch, AIUI, is that the overall demographic changes lag due to the higher-ups still being very disproportionately white & male, but that hiring is even-more-wildly disproportionate in favor of non-white-males.
More detailed & disaggregated data than the article provides would be necessary to evaluate that premise, but it's not a priori implausible.
I agree that it is not impossible. And I would go further and say that I think there are certainly *some* places that went really hard at DEI stuff to the point that it overcorrected. But as PthaMac notes, it doesn't really look like white men are truly having a lost generation.
I don’t think Eigenrobot should have wished Noah Smith stage four cancer. That said, I really can’t figure out a coherent worldview that would allow someone to stand by all of those tweets Noah made. I’m not sure why he didn’t just disavow them.
Yeah I was disappointed by that as well. Eigen clearly got pretty pissed off. I get it though, the anti-white racism the article covers in depth is really despicable, and the cowardice of many people around the topic pisses me off too.
His point, in brief, is that the relative decline of white males does not show up society-wide in any significant amount. His conclusion:
"Ironically, I think Savage’s piece, especially its viral success, ends up unintentionally providing some support for the non-material theories of young male political behavior that are focused on the internet, podcasts, and memes. What Savage and those sharing the piece appear to be responding to is primarily the mental impression that was caused by DEI messaging, which may have had an impact well beyond its actual material impact. The institutions Savage discusses in his piece employ approximately 0% of the US population, but their transformations plus DEI rhetoric plus an internet community aimed at negatively messaging about it all can generate the impression of something much bigger going on."
I think there are two questions commenters should deal with:
1. To what extent was this a widespread social phenomenon, vs. something that impacted a limited sphere of cultural and academic elites. I suspect the phenomenon was more real than Bruenig portrays, though I have nothing more than intuition to tell me that.
2. To the extent that the _perceived_ influence is greater than the actual influence, how much of that is due to adversarial coverage (i.e. right wing media) vs. organic perception in the broader society (i.e. these positions being culturally significant even if small in number).
It looks like the arts jobs held by age 30-39 white males shrunk by 20% from 2015-2020 to 2021-2023, which is a pretty big decrease among the people affected, which is pretty significant
I guess I don't read it that way. The original article has a pretty strong generalizing tone:
"This isn’t a story about all white men. It’s a story about white male millennials in professional America, about those who stayed, and who (mostly) stayed quiet."
... and moreover, much of the discussion around it has had a similar tone:
To be clear, I don't think there's much value in debating whether or not Savage made explicitly clear claims or whether Bruenig is making a sufficiently on-point response.
Because it's clear to me that many people do see this as a very generalized phenomenon that happened, and the best debates should probably center on how much that is true (and, if you accept that it is true, whether it is a good thing or not).
Well I’m glad that this particular negative thing was small enough that it doesn’t swamp all other factors when looking at aggregated economic statistics of the entire workforce. At least we know that this is not literally the worst thing that happened in the last 20 years.
Oliver, my immediate reaction is "Who is Jacob Savage?" Throw those of us dwelling under rocks a bone - give a link, a word of explanation, *something* so we have a clue what we should be excited/depressed/outraged/delighted over.
This is like me asking you guys "Whoo, D.J. Carey, what a turn-up for the books, yeah?"
(Quick abstract: DEI essentially suppressed and punished a generation of white men, and we are seeing the consequences of that today. This is hardly a novel thought, but this is one of the first times it is seriously getting the long-form treatment.)
See, the thing is, he's complaining that he didn't make it big as a screenwriter in Hollywood.
Wow, what a unique experience, never before has this happened to a hopeful, it surely was all down to it being 2016 and the dawn of DEI SJW wokeness.
There's a gazillion people out there who didn't make it big in Hollywood, over a selection of decades. He should have tried whatever the hell it was Payne and McKay did for JJ Abrams, that seemed to get the no-talents the big gig. Your mistake wasn't being a (very average) white guy, Jacob, it was not befriending JJ!
Yeah. The Marvelverse movies burned themselves out by killing the golden goose: too many movies and the later ones about minor characters nobody cared about.
Disney has had a run of "not good" movies. Remakes of old properties should be a money-spinner, but Disney managed to make a mess of "live-action remakes of our classic animated movies" and "Snow White" was an unfolding disaster.
I think right now the studios have no idea what will or won't be a hit and so they're treading water. They're trying to make movies that will not offend overseas audiences (which then runs them into trouble with activists about 'why did you edit out the LGBT stuff?') because movie-making is now so expensive you *need* global box-office returns, but markets like India and China are producing their own home-grown superhero/action films and so they don't need the Hollywood blockbusters as much.
Mostly kept my opinion the same, though I find the continued hypocrisy and the whole “it’s not happening —> it’s good that it’s happening” social phenomenon disturbing, as usual.
Updating my priors about the vast majority of even seemingly intelligent people having a lot of trouble thinking outside of social consensus.
My disappointment regarding the reactions to the article is that it totally deserves pushback, I mean the portrayal of the industry is very heterodox, I don't think anyone is surprised at how many rejected it. Yet, plainly stated, the pushback sucks?
What I expected was "this is some rando quoting other randos, what even is this thing and why would I believe it?" which, mind you, sounds completely sensible to me, but instead what I encounter are mostly attacks on him, strawmen or celebrations for "the good kind of racism".
Probably depends on which Circuit. And since it's the Dems suing, they can just make sure it goes to one of the "Spirit of Aloha" Living Constitution ones that'll rule in their favor.
Unfortunately the Supreme Court usually ends up overruling and saying that whatever the king says goes, no matter how many laws have been passed - apparently the constitution gives the executive the power to ignore laws!
> the constitution gives the executive the power to ignore laws!
No, of course not. Congress can still impeach him and remove him him from office. But until they do that, yeah, Congress's interpretation of the law matters no more to him than Executive Orders do to Representatives.
The problem is that we used to think we lived in a country where laws applied to the government. Now apparently we’ve discovered that the court thinks the constitution puts the executive above the law. It’s unfortunate that it is now impossible for the government to make a binding commitment to do anything - this has destroyed the ability of the government to work with individuals, organizations, and countries that want promises.
I believe the classical response is "What do you mean 'we,' kemosabe?"
Some of us have long known this even before elites flouting without consequence the "pandemic restrictions" they imposed on commoners made it obvious to even the meanest intelligence.
"The Supreme Court emphasized the narrowness and rationality of bills of attainder in Nixon v. Administrator of General Services, 433 U.S. 425 (1977). During the Watergate scandal, in 1974 Congress passed the Presidential Recordings and Materials Preservation Act, which required the General Services Administration to confiscate former President Richard Nixon's presidential papers to prevent their destruction, screen out those which contained national security and other issues which might prevent their publication, and release the remainder of the papers to the public as fast as possible.[54] The Supreme Court upheld the law in Nixon, arguing that specificity alone did not invalidate the act because the President constituted a "class of one".[55] Thus, specificity was constitutional if it was rationally related to the class identified.[55] The Court modified its punishment test, concluding that only those laws which historically offended the bill of attainder clause were invalid.[56] The Court also found it significant that Nixon was compensated for the loss of his papers, which alleviated the punishment.[57] The Court modified the punishment prong by holding that punishment could survive scrutiny if rationally related to other, nonpunitive goals.[57] Finally, the Court concluded that the legislation must not be intended to punish; legislation enacted for otherwise legitimate purposes could be saved so long as punishment was a side-effect rather than the main purpose of the law."
It doesn't matter. If it isn't punishment, which the release is not, it isn't a bill of attainder. From theNixon case:
>Moreover, even if the specificity element were deemed to be satisfied here, the Bill of Attainder Clause would not automatically be implicated. Forbidden legislative punishment is not involved merely because the Act imposes burdensome consequences. Rather, we must inquire further whether Congress, by lodging appellant's materials in the custody of the General Services Administration pending their screening by Government archivists and the promulgation of further regulations, "inflict[ed] punishment" within the constitutional proscription against bills of attainder.
>...
>Application of the functional approach to this case leads to rejection of appellant's argument that the Act rests upon a congressional determination of his blameworthiness and a desire to punish him. For, as noted previously, see supra, at 452-454, legitimate justifications for passage of the Act are readily apparent. ... Evaluated in terms of these asserted purposes, the law plainly must be held to be an act of nonpunitive legislative policymaking.
Australia has compulsory superannuation contributions (ie, forced 401k). We do have plenty of welfare programs tho so social security is not replaced as such.
Works OK, except there's always political pressure to let people to raid their super to prop up the housing market or whatever.
It seems Chile attempted this kind of thing in 1981. It didn't work long-term, and in August 2025 they brought back aspects of Social Security for their pensioners.
It's unclear to me exactly why it didn't work, for I too favor replacing it with private investment. It looks like part of the problem was the pension system they had was very generous, for the military personnel never had to convert. The costs of the transition were also high. There may be other reasons that are hard for me to determine. https://www.ssa.gov/policy/docs/ssb/v59n3/v59n3p45.pdf
I did some analysis of long underwater time periods, where the prices at the end of the period are no higher than the beginning, and they can indeed be long. The Great Depression had the longest I'm counting, from 1929 to either 1945 or 1954, depending on how it is measured with deflation and dividend reinvestment. The Nasdaq had its own bubble burst in the .com era, and did not recover from 2000 prices until about 2015. This leads me to believe that a hybrid system may be best, where the government has its own investment funds, which it can tap in surplus years and fund from taxes or deficit spending in downturns.
Another part of Social Security is Disability, which is paid out of the same fund. This would be an entirely separate issue, but I also believe it should be. If we, as a nation, collectively believe we should provide, say, 1% of our income to support those with disabilities, then the tax for it would be more transparent.
Interesting, will see if I can find a deep dive explainer on why Chile didn't work. As for Disability, I'd like to replace that with forced insurance coverage that can go to any approved private insurance carrier
If I'm not mistaken, disability payments need not go towards healthcare, but are an all-around payment because the disabled have disadvantages relative to the able-bodied. Which is to say, it's just a check you get, and don't have to spend it with insurance.
Depends on how we managed the transition. Considering that today's SS payments cover today's retirees, switching to a forced-savings system would lead to a long gap we'd have to cover with other tax revenue (or else draconian cuts to benefits). That's the biggest issue to manage.
If we could get past that, we'd probably be in a decent place to do something like this.
Would not phasing SS out over, say, 40 years remove this objection?
A 27-year-old today would pay less and less SS tax, as would their employers, and end up getting no SS at 67. That said, 401k contributions should not be forced, but encouraged. Roth IRAs are an even better choice, in my opinion, though they don't receive any employer match. Some 401ks also allow the contributions to be to Roth accounts.
Still the 27-year-old and younger people would be paying taxes for something and receiving no benefit. I can certainly see people objecting to that. Of course, when SS started, people received a benefit without having paid for it. To undo this, someone must make a sacrifice, in getting less benefit than they pay into.
Yes. But it’s not as obvious now as it would be under this phase-out regime. Can the electorate really sustain a willingness to carry out a 40-year weaning with nothing to show for it at the end but a more respectable public budget? I have my doubts.
Well then it seems like you fix that by guaranteeing to pay a baseline floor if the returns don’t exceed it? Possibly that screws some incentive up somewhere, but everyone involved is making more money if it outperforms the floor. The funding would be an issue if we still cared about such things, but the current employer side payroll tax could just be a matching contribution. Even a long bear market would only leave a few people with unfortunately timed retirements in bad shape and in need of the baseline.
Well I considered that, but the investment managers would not benefit in any way from failing. It’s not as if they can use the baseline as a way to play with free money themselves, because they would still individually or institutionally make more money succeeding and make nothing on a fail. I couldn’t figure out any way for a fund manager to actually use the baseline guarantee (which doesn’t even enter into play til the participant retires, bc you won’t know the total returns until that point) to make profits, but maybe someone else can.
For a broad investment fund to do worse than social security would be pretty hard to manage. 401k’s already are restricted in what type of pool you can play in, would all the participants take the high risk option? I don’t think so, not if we set the baseline below what’s expected of a typical median risk pool, you wouldn't want to end up on that it’s just enough to prevent starving and riots
What we're headed into, on account of AI, is not going to look like any economic situation we can relate to historically. Perhaps a bubble in the very short term, but in the long term it will be nothing less than human obsolescence. I see a lot of folks in these spaces clamoring for UBI to address the expected job losses, and quite likely all of that disruption is going to hit and derail anything we plan now. But in the event that transformative AI does *not* arrive before the latter half of the century, that's the world in which we can plan some response, and having people invested in actual ownership of the economy seems far more enforceable than having them rely on a promise from a government that will have increasingly fewer incentives to honor promises to an ever more superfluous labor class.
You can to some extent do that now privately, with variable annuities that don’t give you the entire return if you win big but that provide a floor if you don’t. But of course that works only if you have a limited range of possible investments and a suitable risk pool.
What possesses people in the Trump administration to give interviews to Vanity Fair? Do they crave recognition from the old prestigious publications so much that they let willingly let their enemies humiliate them?
Liberals managed to take advantage of the right's lack of artistic comprehension by taking bad photos that the administration cannot recognise as bad, but that intelligent individuals can recognise as bad.
So the reason they keep doing interviews with them, is because they aren't smart enough to recognise that they're being played.
It's a subtle way of "yes, we're lying, but if they claim we're biased and lying, we can plausibly deny it and call them paranoid, fearful, petty losers".
"But man, what incredible pieces of art those photographs by Christopher Anderson were. So layered with symbolism and storytelling. Just completely showing how small and grotesque- how inconsequential these people who want to appear & believe they are larger than life. Someone said - they look like middle managers of a run down midtier hotel. It humanizes them in a very important way - to take away the fear & power that they so desperately want and show them for what they are.
... And to see so many people really engage with it is amazing - pull out the symbolism, like someone making a phone call in darkened room in the background of a portrait. Or the intentional framing of the photos and the thought provoking reflection & shadow. (I love the use of reflections & shadow in photography to tell a story - a big part of my undergrad thesis was on that)
It makes me so proud in this age of poor/inattentive reading comprehension to see how actively people want to understand and engage with this art."
Now, tell me again how only Fox News is biased and untruthful, the rest of the liberal media is pure angelic verity.
I think it's a "damned if you do, damned if you don't" situation. Do give an interview, you give hostages to fortune. Don't give an interview, everyone runs around grinning that you're so scared of the hard-hitting investigative journalists of, um, Vanity Fair that you refuse to even talk to them.
I think it's more like the Light Grenade from "Mom and Dad Save the World". It's an obvious trap with minimal bait, but certain people find the bait (a bit of limelight, in the case of Vanity Fair interviews) irresistible and don't notice the risk/downside for whatever reason.
More seriously, I get the impression that the kind of media outlets that White House staffers deal with most often have an informal understanding that they'll give you somewhat more favorable treatment if you give them good access, in order to give you an incentive to give them access in the first place and continue giving access afterwards. Vanity Fair doesn't seem to be playing by these rules, which apparently is catching these folks by surprise.
I don't agree; there is no reputational penalty to not doing an interview with Vanity Fair. The closest we've seen things come was Kamala Harris irritating journalists by never doing any interviews with anyone, and there was no penalty to her (outside of the eyes of professional journalists) for doing that either.
There's no expectation of doing an interview with Vanity Fair for anyone; no one will ever notice when one isn't done.
"There's no expectation of doing an interview with Vanity Fair for anyone; no one will ever notice when one isn't done"
You may not *notice*, but the subliminal influence will happen all the same. Why manipulation of the narrative happens, and how it happens. Scott posted about how the media don't directly *lie*, but there's more than one way of 'lying'.
These are the people who lap up the stories about "fascists in the White House". These are the people Vanity Fair and other publications are targeting. 'Nobody cares if you don't do an interview with A or B media outlet"? These people notice, and share stories about how they've got the fascists on the run.
"There is ZERO retouching done on these. Which is very unusual for photos taken this close up. The expressions chosen are very odd. The lighting is a ring flash very far away (probably to create a catchlight in the eyes) and then another diffused light source from the side. The ring flash was placed below their eye line, so you get a tiny bit of that creepy uplighting like in a Frankenstein movie. And neither light source was placed very close so the harder light really shows every pore and wrinkle and lip filler injection site.
Basically, he used the opposite of every technique one would employ to create a flattering photo.
But the reason I think this photographer twisted the knife to make these photos look worse was his use of white balance on Karoline Leavitt and Susie Wiles.
If you want to make someone look sickly and demonic, you make their skin tone less warm and add green. When I corrected the white balance on the photos, I discovered the photographer reduced warmth by about 15% and increased green by about 45%."
So, yeah: avoid doing interviews with outlets you know are out to get you, and get the reputation of being paranoid (the media is impartial! it's only conservative/right-wing media that is biased!) and fearful (you won't do an interview because you know the hard-hitting interviewer will drag the truth of your evil out of you).
Not the worst analogy, but I think people who have done a crime probably mostly know it, but people talking to the press probably still think of themselves as obviously great in the eyes of the audience.
There is a certain thread in the legal community that says you shouldn't talk to the cops whether you have committed a crime or not. For a very entertaining video on the subject, see here:
Really? I think politicians tend to be familiar with the political perspective of mass media outlets.
There was an episode of The Newsroom that made the point that if you agree to an interview, the way you look is fully determined by the way the interviewer wants you to look. (A point you can find being made at length in any number of places, but as you might expect it's done a bit more punchily on the TV show.)
The 2006 single "Face Down" is an anthem for male feminist white knighting.
The song starts with the singer jacking off to his crush:
"Hey girl, you know you drive me crazy
One look puts the rhythm in my hand"
Alas, he can't have her, as she's busy being abused by another man:
"Still, I'll never understand why you hang around
I see what's going down
Cover up with makeup in the mirror
Tell yourself, "It's never gonna happen again"
You cry alone, and then he swears he loves you"
The lyrics continually switch between the singer pleading with his crush and lecturing her abuser:
"Do you feel like a man when you push her around?
Do you feel better now, as she falls to the ground?
Well, I'll tell you my friend, one day this world's going to end
As your lies crumble down, a new life she has found"
We see the common phenomenon where abused women deny the abuse is occurring, along with the singer's fantasy that one day she'll leave him:
"Face down in the dirt, she said, "This doesn't hurt"
She said, "I finally had enough"
One day she will tell you that she has had enough
It's coming 'round again"
The singer fantasizes about future justice. But what of the present? He's jacking off alone while the abuser gets to have sex with his crush. Justice delayed is justice denied. And even if the woman does leave the abuser, what's to say she'll end up with the singer, instead of a different abusive man? How'd she wind up with the abuser in the first place?
To any male feminists reading this, I'm not against you. I used to BE you. I self-identified as a feminist, since I associated opposition to feminism with religion and thought feminism was about being nice to women. But I grew to realize that the feminist narrative "don't blame the victim" was insane. These women are rewarding abusive men with sex. It is time for society to realize that and treat them accordingly.
It seems like your ultimate goal here is something like "punish abusive men for being abusive," hopefully therefore leading to less domestic abuse occurring. You've not specified how exactly society should treat women in relationships with abusive men. However, all the possible specific implementations of this idea that I can think of seem like they wouldn't assist your end goal. Namely, society could punish women for being in relationships with abusive men by: (1) criticizing them, (2) ostracizing them, (3) fining them, or (4) criminal proceedings/jail time.
None of these will work to prevent women from getting into relationships with men who turn out to be abusive, as the primary problem is that he seems nice and charming at first, and the abuse starts later once she is emotionally/financially dependent, or otherwise vulnerable (e.g. pregnant.) There are occasional women who get into relationships with known abusers, on the grounds that "all his crazy exes were just lying" or "he'd never do that to me!!" These women do not believe that abuse will occur, and so cannot be deterred by the promise of further punishment.
Moreover, they do a pretty poor job of encouraging women to break up with abusive men. Women in these situations already get criticism from those close to them; the reason you're not supposed to do that is because it makes them feel attacked and unsafe to ask for help. Social isolation prevents abused women from leaving, as they can't access emotional support, temporary accommodation, help with childcare, or police protection from retaliation. Similarly, fines or jail discourage them from ever speaking about it to anyone who might be able to help, and also ensure they can't press charges against their abusers without running serious risks to themselves.
Therefore, I think that this policy is ineffective as well as unkind.
Of course, if instead the goal is to make more women want to have sex with *you*, I'd be happy to provide feedback on your fashion/style, confidence, etc. Happy holidays, and I hope you find peace!
"None of these will work to prevent women from getting into relationships with men who turn out to be abusive, as the primary problem is that he seems nice and charming at first, and the abuse starts later once she is emotionally/financially dependent,"
What does "emotional dependence" even mean? Would you accept a man claiming he was "emotionally dependent" on his gang leader and therefore he has no choice but to continue dealing drugs?
I thought feminism was about treating women as equals. It's actually about treating them like children and denying them agency or responsibility.
Is your goal to make things better, or just to assign blame in a more emotionally satisfying way? If the former, you're dodging the key questions -- what specifically are you calling for, and why will it work despite the issues landsailor raised?
<Is your goal to make things better, or just to assign blame in a more emotionally satisfying way?
This is almost always a great question to ask yourself. But if you ask somebody who's already on an emotionally-satisfying-blame roll they just give you an indignant glare -- "Fuck no!" You gotta wait til later when there's a chance they'll have some post-nut clarity.
I am no longer feminist because it represented itself as being about punishing abusive men, but then I saw feminists telling women it isn't wrong for them to reward abusive men with sex. I see it as an indirect form of aid to abusive men.
Your main complaint here seems to be "that bitch is fucking him, not me, so serves her right if he beats the crap out of her" and not "that woman is being physically and otherwise abused by someone supposed to care about her".
It's very, very hard to get someone out of an abusive relationship. I've seen this with family. You can scold, lecture, appeal, till you're blue in the face, but they'll always go back and give the abuser a second/third/fiftieth chance. It really is a change in the brain state, there's no other way I can explain it.
People in trauma may not have the capacity to get away from an abuser. They may not have family or friends to turn to—my own mother was sent back to my verbally abusive father by her parents, she had two small children to care for and she didn't have a job or or access to money*. There may be no spaces in shelters. A woman in a long-term abusive relationship may have lost any ability to tap into inner resources like courage or resilience, even more so if she grew up in a disempowering abusive household.
Have you looked into track records of the so-called authorities, like the police, in not enforcing restraining orders or not testifying in order to support someone getting a restraining order? Have you checked out statistics about femicide, which includes women fleeing from and being tracked down by abusive partners?
We "blame the victim" by not offering society level support, be it logistical, legal, or emotional.
* Up through the 70s, women couldn't get a credit card without a male person co-signing.
"A woman in a long-term abusive relationship may have lost any ability to tap into inner resources like courage or resilience, even more so if she grew up in a disempowering abusive household."
We need to break this cycle by holding women accountable for rewarding abusive men with sex.
"* Up through the 70s, women couldn't get a credit card without a male person co-signing."
<We need to break this cycle by holding women accountable for rewarding abusive men with sex.
Hey, do we get to do that with rape victims too? And how about we really go for it and also hold murder victims accountable for rewarding their killers by dying?
That describes when banks were legally required to issue credit cards to women, which is not the same as women being completely unable to get credit cards without a male co-signer.
From what I gather, married women did legitimately have a hard time with most financial institutions before the 70s, but how much trouble depended on the particular bank's policies. Some required the husband to be on the account, while others would only issue credit cards to women who had her own income and would discount her income by something like 50% unless she was past childbearing years or was demonstrably infertile.
But that's specific to married women. I expect policies also varied with regards to widowed, divorced, and never-married women.
Also, an important bit of context is that most people of either sex didn't have bank credit cards in the 1970s. Credit cards in the modern sense only became a thing in the late 60s. Only 16% of households had bank-type credit cards in 1970 and 38% in 1978. A lot more had store cards, and I have no idea what sorts of policies stores had about issuing store cards to women. Source: https://www.federalreserve.gov/pubs/bulletin/2000/0900lead.pdf
You'll see people who are trained in Latin ask why the word is "homicide" and not "hominicide", given that the root of the word "homo" is "homin-".
The answer turns out to be that the -cida suffix attaches to the nominative form of the noun rather than the root. Some words compound this way for unclear reasons, and caedo happens to be one of them, I guess.
But that won't help "femicide", where the root of the noun is "femin-" and the nominative form is "femin-a". To produce "femmicide", you'd need to be compounding from French or something.
Society should hold the women accountable as enablers of their own and other women's abuse as they are incentivizing and training men to be abusive by rewarding them with sex.
Hold them accountable *how*, though? Saying "hold them accountable" offers no more detail on your intent than "treat them accordingly" does. Could mean anything from "less funding for domestic violence organizations" to "make being a victim of IPV a crime and send them to special new abuse-victims-who-didn't-leave gulags."
So what are you proposing, specifically here? Sending them to jail or fining women (or men?) who are abused and don't leave their partner timely? Eliminating organizations that provide abuse victims with support? Banning protective orders unless they are sought within X months of the first instance of abuse? A social campaign where people just feel comfortable being mean to victims of domestic violence (men too?) online?
I am calling for a shift in societal messaging away from the feminists' ridiculous "don't blame the victim" narrative. The laws should not change, and obviously help should be provided to those who leave, that's exactly what I'm advocating that these women should do.
The general tenor of discourse around that, so far as I have seen, is that "women are thus proven to be incapable of reasoning, incapable of making decisions for themselves, and hopelessly addicted to being whores for low-class abusers. Therefore society should, for their own good, take away discretion as to their choices from women. Let's go back to the days when the father made decisions about who his daughter would marry, and then the woman passed under the authority of her husband.
To whom, of course, she could never refuse sex. Not like nowadays, when that bitch won't sleep with me but will sleep with the jerk beating her.
This is the only way to keep women from ruining themselves and to ensure that Nice Guys get their government-appointed compulsory girlfriend. Women cannot be entrusted with the ability to decide who they will date or marry or have sex with, the men of their family must make that decision for them".
If you think I exaggerate, I'm glad you've not read some of the stuff I've read about "women should not have the vote; not enough babies being born, solution: do not permit women to go to college or have careers, steer them towards getting married straight out of high school" and even more radical than that.
A thought experiment -- imagine that there is a group of white girls who think that there are too many black people around them. But, being girls, there is nothing they want to do about it personally. So instead, they publicly make a pledge that whenever a group of white men lynches a black man, they will give each of those white men a blowjob.
How should we look at these girls from a moral perspective? Clearly, they are not participating in the violence. And it is their legal right to have sex, or refuse to have sex, with whomever they want to. Therefore... ?
Congratulations, boys, you've made it all about your dicks again.
"Consider thought experiment: racist murder bitches giving blowjobs to guys not me. Is *that* moral?"
EDIT: Let me explain. The complaint here is not "black men are being murdered", it's "the women are having sex with the wrong kind of men".
The "wrong kind of men" being "guys not me". Even the original comment phrased it as women "rewarding" abusive men with sex.
In an abusive relationship, there may not be any choice about giving him sex. Just like the "nice guy" singer of the song who wants the woman to be with him not the abusive guy, not because he cares about her in truth but because he wants her so that's how it 'should' be.
I tried to make the thought experiment simple, you avoided the question.
Actual situation: Many years ago (the 1990s Mečiar's era in Slovakia), I knew a girl who was dating some kind of mafia guy. He participated in small business extortion (maybe also something worse, but that wasn't public knowledge), and he was making a lot of money. They both lived together in luxury -- a large house full of modern electronics, expensive vacations all the time.
Once I had an opportunity to talk to the girl. I didn't want to be impolite, but I was curious how does it feel to know that the money that finances your luxurious life was probably taken by force yesterday from some hard-working guy, maybe threatening to break his legs or hurt his family if he doesn't pay. She didn't *participate* in the crime, but she was the immediate *beneficiary* of the crime; whenever she went on a shopping trip, she was spending the kind of money she probably couldn't get otherwise. And she was perfectly aware of what was her boyfriend's job; she didn't have any illusion about it. And she was not threatened to stay in the relationship; she chose it voluntarily, because that was where the money was. (Hypothetically speaking, if her boyfriend suddenly had a change of heart, stopped doing crime, and found a normal job with a normal income, she would probably leave him as soon as her quality of life dropped towards the normal levels.)
Based on our short interaction, she saw absolutely no problem in her situation. It was him doing the crime, not her; end of story. The fact that he gave her the money, well, that's what rich boyfriends do. Where the money comes from is not her business. Yes, he has a dangerous job; if something bad happened to him because of his job (like getting killed, or getting in prison), she would obviously be sad, but it was his choice as a man to take a dangerous job, and she wouldn't try to control him and tell him what he can or cannot do, also he is not the kind of a man who would listen to such things. (Zero thoughts about the people he hurts as a part of his job.)
I have no idea how the story ended; I am no longer in contact with her, no idea even whether they are still alive or not.
What I found fascinating about this story was the difference in moral judgement when I told this story to a few men and a few women. Maybe it was just an accident, a small sample. But the men typically had the intuition that if you knowingly put yourself in a position of a beneficiary of a crime, then you are in some way... responsible? involved? Like, suppose that one of the extorted people gets crazy and takes a gun and shoots both the guy and his girlfriend in a car, then... well, they both had it coming, a karmic revenge, there is no reason to feel sorry for her. She knew what she was a part of. She benefited from the crime just like he did, only he did all the dirty work.
The women, on the other hand, typically just saw it as: the girl got lucky, and she is in no way responsible for what her boyfriend does. If a guy does crime, it makes him a bad guy, but it does not reflect on his girlfriend in any way.
I found this quite shocking back then. (It contributed to my process of redpilling.)
"In an abusive relationship, there may not be any choice about giving him sex."
You yourself said "they'll always go back and give the abuser a second/third/fiftieth chance." If they're giving them a fiftieth chance, they know what they're getting into. The truth is in front of you if you care to see it.
"Just like the "nice guy" singer of the song who wants the woman to be with him not the abusive guy, not because he cares about her in truth but because he wants her so that's how it 'should' be."
The white knight, or as you call it "nice guy," expects his righteousness to be rewarded, and the unrighteousness of the evildoer punished. I highlighted the song as an example of how the expectation is unrealistic, and your comments here just reinforces my point. Male feminists should know that female feminists will not be the least bit appreciative or thankful and see nothing wrong with rewarding abusive, decidedly un-feminist men with sex.
This is not to say that men should be against all women (I am here speaking of feminists, not all women), and certainly not that men should oppress women. What they shouldn't do is sit around writing delusional fantasy songs about how one day "justice" will be served. The path to a better world must go through a break with the poisonous ideology of feminism, which promotes the very evil it purports to fight.
"when that bitch won't sleep with me but will sleep with the jerk beating her."
Is that not an injustice? Evildoers are supposed to be punished, wouldn't you agree, not rewarded?
"If you think I exaggerate, I'm glad you've not read some of the stuff I've read about "women should not have the vote; not enough babies being born, solution: do not permit women to go to college or have careers, steer them towards getting married straight out of high school" and even more radical than that."
Those people are evil and I became a feminist because I wanted to oppose those people. I still want to oppose those people, I just think women should stop rewarding evil-doing, abusive men with sex.
I won't add a link as I don't think the podcast I did on the Normans was by any means the best thing out there. But I am a big fan of Ed West's Substack the excellently named The Wrong Side of History. I think he is the best British Substacker out there and as a sort of hobby has written a bunch of very short histories (the best sort - so many history books now run to over 600 pages!). So I had him on to talk about the Norman Conquest and what led up to it.
One thing that was interesting to see was a bit of Albion's Seed avant la letter. Apparently voting patterns in Wales very much map on to where Normans settled/did not settle.
And also interesting to contrast both how close it was - the Normans really should have lost - and how total the defeat was for the Anglo Saxons. Basically the entire ruling class was dispossessed. Having brought over a huge number of land hungry knights William handed over pretty much all the land to them except for the bits he kept for himself. Oh and the bits gave to the Church which (maybe not coincidentally had given the invasion their blessing at the highest (ie papal) level.
I say the Normans should have lost. I mean they were brilliant fighters not least from having practiced on so many other people. In particular they had gone several rounds with the Byzantines and learned a lot in the process. But even so an army of that size should never have been able to conquer a country as large, rich and well organised as England. The problem for the English was that to raise and lead armies you need leaders and Harold stupidly got himself killed at Hastings along with his brothers. He'd already demolished Harald Hadrada a few weeks previously so handing over command to a brother while he got another army ready just in case was the obvious thing to do (and was what his dear old mother had strongly advised). And with the only viable candidate for the throne once Harold and his brothers got themselves killed being Edgar (aged 14) English resistance basically collapsed. There really isn't anything quite like it in European history (well, post Roman at least).
Anyway apart from the utterly brutal suppression of a rebellion in the North - deliberate destruction of crops and mass starvation - the Normans proved to be surprisingly woke in their outlook. They did away with slavery which the Anglo Saxons had been very keen on (apparently about 10% of the population were slaves pre conquest). And even more feebly they stopped executing people for political reasons. Basically nobody noble got executed until the time of Edward I. After which it was 'game on' of course. Game of Thrones in fact.
Speaking of Edward I (also a recent podcast!) I was told by Marc Morris that he was so called because his father was a massive fan of Edward the Confessor (the king before the short-lived Harold). But for that, the name 'Edward' would be just as odd as 'Aethelstan' and all the other weird Anglo Saxon ones. [Gordon Tremeshko points out this can't be right - Edward is familiar in lots of languages. I most likely misunderstood what Marc was saying - he is not given to making these kinds of mistakes whereas I . . .]
Anyway my takeaways were 1) progress is unstoppable (particularly if backed by a Norman knight) and 2) always listen to your Mum.
If you are interested the podcast is Subject to Change with Russell Hogg (though there are so many things out there on the Normans that are probably much better). And if you have room for another Substack and would like the view from Britain I can't recommend Ed West's Wrong Side of History enough.
Agreed, Ed West is always worth reading. Unusually for me, I pay my subscription to read that thoughts of a genuinely right-wing writer who leans towards pessimism of the things-are-good-but-they're-getting-worse variety.
Duke William knew that one of Harold Godwinson's tricks was to make almost incredible marching progress with his army and confront his adversary before they were expecting his arrival. This worked with Harald Hardrada, but it was never going to work with William (A couple of years previously Harold, on a visit to Normandy, had joined William in a campaign against Duke Conan of Brittany, and no doubt William took careful note of Harold's advice and military doctrines, and his impetuous character.)
Also, William deliberately landed on one of Harold's personal estates, and laid waste to the area knowing that this would enrage Harold and make him want to hurry even faster to confront William. On his way to that encounter, Harold paused in London, where (as you mentioned) his mother and brothers begged him to wait until his full army (including cavalry and archers) was assembled. But Harold ignored them, actually insulting their caution, and rushed on with just the troops he had assembled on the way ..
> deliberate destruction of crops and mass starvation (in the North)
The harrying of the North was an exercise in area denial, to prevent the Scots and Vikings gaining much of a foothold while William was dealing primarily with the Welsh. A year or two later he marched into Scotland and burned Edinburgh to the ground, before agreeing a settlement with the Scottish king Malcolm Canmore. He also paid off the Vikings one year, and thrashed them the next, so much so that they never invaded again.
Sounds pretty cool. I count myself an Ed West fan, also. One nitpick, though: that part about "Edward" doesn't sound right. That's a pretty common name in other European languages, like French (Edouard), Spanish (Eduardo), and German (Eduard).
At least in the Spanish case (Eduardo) it's a recently introduced name. There's no Eduardo patronymic surname as is the case in English (Edwards) because it didn't exist at the time when patronymics started being used in Spain.
Edward is very definitely an English name; Eduardo in Portuguese seems to drive from the 15th century Portuguese king Edward, son of... tbe Englishwoman Philippa of Lancaster, granddaughter of Edward III of England via John of Gaunt. I'd guess that other versions of Edward in romance languages have a similar origin, though there might be a cognate name in other Germanic languages
"News reports indicate that Secretary Kennedy plans to replace the U.S. childhood vaccine schedule with Denmark’s, the developed country that recommends the fewest vaccines."
Obvious question here: does Denmark have significantly worse outcomes in childhood diseases? Do they delay the vaccination schedule so that children get these vaccines but at later ages, or is it "Nope, you only need measles/mumps/rubella vaccination"?
"...estimate the medical consequences if American doctors stop prescribing these vaccines."
Is America significantly worse off in public health/epidemics such that Denmark can get away with this and have a healthy child population but it would be disaster for America? Before we start throwing our toys out of the pram, some data would be nice. It seems to me that babies and infants now are on a much increased schedule of vaccinations since my day, and that may be (1) yeah well, new diseases since your time, grandma, and besides you all died of polio didn't you? or (2) hell yeah this is a huge burden to put on the developing immune system, maybe slow it down there a notch.
EDIT: Happy Christmas/Solstice/Hannukah/your festival of choice/Festivus for the rest of us to you all, as well! Starting a much-needed Christmas holiday break and apart from light housework and of course cooking The Big Day Dinner, I intend to do Sweet Fanny Adams apart from sit around and eat way too much desserts and chocolate 😁
These kids probably have a lot less burden on their immune system given how much sanitation and antibiotics have gone everywhere - not sure why a few vaccines would be a bigger burden.
I'm somewhat skeptical that we're vaccinating 100% of newborns for two diseases associated with sexual activity and IV drug use which seems to add risk for little benefit given how easily you could screen Mom who's Right There.
Vaccinating 12–14yos verses the same diseases makes quite a bit more sense, much as it might feel like betting against your kids behavior. That I Wasn't sexually active in HS almost certainly wasn't my fault, as stupid an idea as that seems to me forty years on.
Why is it better to wait for those vaccines? Is the idea that they are more effective if they are taken later, or that you would get more uptake by doing it later? It seems that doing it at age 12-14 sets things up for a lot of stigma and conversations about sex and drugs, while doing it in infancy makes clear it’s just a preventative measure.
Not particularly, you can't test for most speculative connections between things in biology. There are probably a thousand plants you eat in each week and a hundred leading causes of death and no one tests is say celery causes Parkinson's disease, we just assume not unless we have some kind of reason to tesf the hypothesis.
I guess prima facie it seems plausible because vaccines depend on immune reaction. That seems like enough to want to test the hypothesis.
One reason we followed a European schedule for my kids was simply informational: if a kid had a reaction to a vaccine, getting fewer at a time allows you to narrow down what's going on.
https://www.ncbi.nlm.nih.gov/books/NBK220490/ - on multiple immunisations and immune dysfunction. The summary linked here goes through several potential mechanisms for multiple immunisations to cause problems, and there is either evidence against a causal relationship between multiple immunisations and (eg) type 1 diabetes, or there isn't enough information to say (eg for asthma).
https://www.who.int/groups/global-advisory-committee-on-vaccine-safety/topics/immune-overload - the Global Advisory Committee on Vaccine Safety report summary (and a link to the full report), which says "the available evidence reviewed by GACVS does not support the hypothesis that vaccines, as currently used, weaken or harm the immune system." "Current" here means 2006, and I don't know how the vaccine schedule has changed since then.
Relevant quotes I pulled out from the above paper:
"However, vaccine-specific antibody responses and rates of vaccine-associated adverse reactions of children with mild or moderate illnesses are comparable to those of healthy children. For example, the presence of upper respiratory tract infections, otitis media, fever, skin infections, or diarrhea do not affect the level of protective antibodies induced by immunization."
(references: King GE, Markowitz LE, Heath J, et al. Antibody response to measles-mumps-rubella vaccine of children with mild illness at the time of vaccination. JAMA.1996;275:704–707
Dennehy PH, Saracen CL, Peter G. Seroconversion rates to combined measles-mumps-rubella-varicella vaccine of children with upper respiratory tract infection. Pediatrics.1994;94:514–516
Ratnam S, West R, Gadag V. Measles and rubella antibody response after measles-mumps-rubella vaccination in children with afebrile upper respiratory tract infection. J Pediatr.1995;127:432–434
Ndikuyeze A, Munoz A, Stewart J, et al. Immunogenicity and safety of measles vaccine in ill African children. Int J Epidemiol.1988;17:448–455
Halsey NA, Boulos R, Mode F, et al. Response to measles vaccine in Haitian infants 6 to 12 months old: influence of maternal antibodies, malnutrition, and concurrent illness. N Engl J Med.1985;313:544–549)
"If vaccines overwhelmed or weakened the immune system, then one would expect lesser immune responses when vaccines are given at the same time as compared with when they are given at different times. However, the following vaccines induce similar humoral immune responses when given at the same or different times: 1) MMR and varicella 2) MMR, diphtheria-tetanus-pertussis (DTP), and OPV 3) hepatitis B, diphtheria-tetanus, and OPV 4) influenza and pneumococcus 5) MMR, DTP-Hib, and varicella 6) MMR and Hib and 7) DTP and Hib.
(references:
Englund JA, Suarez C, Kelly J, et al. Placebo-controlled trial of varicella vaccine given with or after measles-mumps-rubella vaccine. J Pediatr.1989;114:37–44
Brunell PA, Novelli VM, Lipton SV, Pollock B. Combined vaccine against measles, mumps, rubella, and varicella. Pediatrics.1988;81:779–784
Deforest A, Long SS, Lischner HW, et al. Simultaneous administration of measles-mumps-rubella vaccine with booster doses of diphtheria-tetanus-pertussis and poliovirus vaccines. Pediatrics.1988;81:237–246
Giammanco G, Volti S, Mauro L, et al. Immune response to simultaneous administration of a recombinant DNA hepatitis B vaccine and multiple compulsory vaccines in infancy. Vaccine.1991;9:747–750
DeStefano F, Goodman RA, Noble GR, et al. Simultaneous administration of influenza and pneumococcal vaccines. JAMA.1982;247:2551–2554
Shinefield HR, Black SB, Staehle BO, et al. Safety, tolerability, and immunogenicity of concomitant infections in separate locations of MMRII, Varivax and Tetramune in healthy children vs concomitant infection of MMRII and Tetramune followed six weeks later by Varivax. Pediatr Infect Dis J.1998;17:980–985
Dashefsky B, Wald E, Guerra N, Byers C. Safety, tolerability, and immunogenicity of concurrent administration of Haemophilus influenzae type b conjugate vaccine (meningococcal protein conjugate) with either measles-mumps-rubella vaccine or diphtheria-tetanus-pertussis and oral poliovirus vaccines in 14- to 23-month-old infants. Pediatrics.1990;85:682–689)
"Vaccines may cause temporary suppression of delayed-type hypersensitivity skin reactions or alter certain lymphocyte function tests in vitro. However, the short-lived immunosuppression caused by certain vaccines does not result in an increased risk of infections with other pathogens soon after vaccination. Vaccinated children are not at greater risk of subsequent infections with other pathogens than unvaccinated children. On the contrary, in Germany, a study of 496 vaccinated and unvaccinated children found that children who received immunizations against diphtheria, pertussis, tetanus, Hib, and polio within the first 3 months of life had fewer infections with vaccine-related and -unrelated pathogens than the nonvaccinated group."
(references:
Black SB, Cherry JD, Shinefield HR, et al. Apparent decreased risk of invasive bacterial disease after heterologous childhood immunization. Am J Dis Child.1991;145:746–749
Davidson M, Letson W, Ward JI, et al. DTP immunization and susceptibility to infectious diseases. Is there a relationship?Am J Dis Child.1991;145:750–754
Storsaeter J, Olin P, Renemar B, et al. Mortality and morbidity from invasive bacterial infections during a clinical trial of acellular pertussis vaccines in Sweden. Pediatr Infect Dis J.1988;7:637–645
Otto S, Mahner B, Kadow I, et al. General non-specific morbidity is reduced after vaccination within the third month of life—the Greifswald study. J Infect.2000;41:172–175)
Interestingly, it also points out that in the past, a child receiving the smallpox vaccine got over 200 proteins, while the current (as of 2002) schedule includes fewer than 130 proteins in total. That's partly due to smallpox no longer being needed, and partly due to improvements in protein chemistry resulting in vaccines containing fewer antigens.
These are all dated around 2002-2006. There's loads of current studies on individual vaccinations, but I couldn't find any recent ones about multiple vaccinations, maybe because it's hard to persuade funders to pay for something that's already been shown multiple times?
One major difference I'm aware of is that Denmark doesn't vaccinate for Hep B, but they do screen all pregnant women for Hep B (which we do not do in the US). I think the logic here is that more kids go to the pediatrician, than pregnant women go to the OB (+follow up with required bloodwork).
In the US, we do *try* to screen all pregnant mothers in the US for Hep B. In practice, we manage to screen 82%-88% [1].
Most of the remainder is people who couldn't get into prenatal care promptly (maternal care deserts are a huge problem, especially rurally, and especially in states that restrict midwifery practices), or people who moved and switched providers and didn't transfer records. Denmark probably does not share those problems.
Trump's ACIP recently voted to change the Hep B recommendation from "everyone gets this at birth" to "everyone gets this at birth, unless the mother has tested negative, in which case she can either request it at birth or wait until 2 months."
I have heard the ACIP meetings have been chaotic and irrational, and have nearly missed making some much worse decisions. But as far as what they've actually passed so far, like the Hep B revision, I don't have strong opinions about it. It's not actually a big change from what we used to have. It's a little more complex, but because the exceptions are limited to cases where we do have data, I don't think it's a big problem. I don't appreciate politicized commenters elsewhere calling them a "death cult" over things like that, but I agree they certainly seem incompetent and at risk of passing actually detrimental recommendations in the future.
Hey everyone, I'm a bit bummed about having missed the Oxford meetup back in October :/. Is the group still active/exists at all?
I wrote a short story that takes place at Oxford University and wanted some feedback from people who know it and how it works (especially since I think the story might be more interesting to people who have some attachment to the place). Anyways, any feedback is welcome!
I'm a linguist (albeit not a phonetician), and I'm a bit confused about the experimental setup. It seems like the way it works is:
1. Two subjects have a conversation in a room and record themselves.
2. An automatic parser takes the audio file and outputs a transcript in some kind of phonemic or phonetic system.
3. The audio file is discarded, or at least locked away where NB can't access it.
4. NB transcribes the output of the parser into English orthography.
Step 3 doesn't make sense for a speech production phonetics experiment. If the researchers are interested in how the subjects pronounce /s/ (or whatever), they care about fine-grained acoustic details that would be entirely lost in even a very detailed phonetic transcription. They would want to keep the audio file so they could make a spectrogram and measure whatever they need to measure.
I guess I hadn't really fleshed out the experiment setup, but if I had to answer in order for it to make sense, I guess I would make the annotation phase preliminary to the human analysis/review part of the experiment, where the grad student (presumably) that is tasked with going over the material flags the specific part for later manual inspection.
It's probably not 100% satisfying to a phoneticist that actually carries out these sorts of studies, sorry :/
The story just gets going and then it stops? We are meant to guess that A and B are now hooking up, and maybe going to be dating. The Italian story then is a red herring? Because I don't think really serious Buddhists would be hanging out in a bar, or at least not drinking alcohol. Not if they've filed their teeth down to commit to vegetarianism.
And we don't even know that they *are* Buddhists, that's just a conjecture. It's a nice build-up to the conventional (and nothing wrong with that) Christmas ghost story set in Oxbridge, but we end up with "the transcription machine works but doesn't yet recognise kissing/groping/fucking noises" and that's not where I thought it was going.
Yeah, I guess I didn't really have a conclusion in mind when I started it. The general concept that I had in mind was: the physical makeup of objects (and people) has an impact on how they work/what their purpose is.
I kind of had the idea in mind that it would thematically tie in the Italian story and the flirting/love story (although now that I'm typing this out, it would be better to make it specific to A and B's connection rather than generically for male-female).
I guess in the back of my mind this is also tied into phonetics i.e. the physical manifestation of language in its most concrete form.
Anyways, I appreciate the feedback, and especially from you, Deiseach! I've gotten into trying to write more structured fiction, but I always struggle with keeping everything thematically consistent and also narratively satisfying. Who would have thought that writing would be this hard!
The stop was very abrupt, so it was hard to know what the point of the story was. You have a decent little traditional ghost/horror story going on there (maybe the flat-teeth people start showing up around Oxford? are they following A?) but the bit about the antagonists-to-lovers with A and B and then the experiment muddled it a little.
The bit about "the physical manifestation of language" sounds like a good idea to explore (does telling the story evoke the weirdness of the people? does it begin to cause physical change in A and B? what of the person transcribing the tapes?) so if you ever do work on the rest of it, let us know!
I'd bring the entry into the bar slightly sooner - I have a fondness for descriptions of Oxford but was wondering when we'd get to the bite. Or give us some other hook earlier on. "I didn’t even get to the point, or even the spooky part" - you need to :p
The recording stopping just after we get into the pub doesn't really work as a cliffhanger, because we don't yet know why we care about these weirdly-smiling "men and women of various ages". I think you need to get a hook in before any interruption. By the second interruption, I'm engaged.
"had I been sober, I would have immediately looked at the floor" - I liked this very much :)
Minor note: Oxford has its own names for many things. Rather than Autumn, Spring, and Summer we have Michaelmas, Hilary, and Trinity. The Thames gets called the Isis as it flows through the city (Tamesis is the ancient Latin name for the river).
Somehow, I found the... commitment device... very funny :)
I thought the ending had a pleasantly meta call-back, but it left me a bit underwhelmed tbh - I wanted a bit more setup->resolution, more payoff, in the main story. The writing successfully spins an atmospheric Oxford, but you could improve the pacing of the story's beats, and increase the stakes.
Overall fun! Thanks for sharing, would enjoy talking in person if you come by the pub some time.
Thank you for the feedback! I actually don't remember why I chose not to call the river Isis, but I definitely knew this. I'll see you at the pub then!
I have a pretty intense dislike for Deciding To Win, on several levels:
- It struck me as mostly poll-chasing and abdication of leadership
- Basically no discussion of which things are actually good, independent of polling (again, make an argument for good things, don't just assume the polling results are inevitable)
- I didn't feel like they even really followed their own polls that well (e.g. they conspicuously omitted "raise the top tax bracket" from the main report even though it polled quite well)
- They said Medicare for All was unpopular, which is only really true if you formulate the question as "Medicare For All, and we ban all private insurance", which is not a necessary formulation
[EDIT:] Also, it's not really accurate to describe people like Davids Axelrod and Plouffe as "rogue Democrats". They're longtime party insiders.
Isn't that the entire point of democracy? They're supposed to be representatives of the people, not pursuing their own agenda. Poll-chasing to get the majority vote is simply the system working as intended. The only purpose politicians have is to express the will of the majority.
First, I strongly disagree that the point of elected representatives is to just vote whatever polls say. The point of elected representatives is to do what's actually useful at achieving what people want. But people can support counterproductive policies that will actually make them worse off. Having a class of professionals that (in theory) studies these things and tries to work out good solutions is valuable.
Second, if you do believe this, why would you specifically want one party to win? These people obviously want Democrats to win; they're just failing to articulate much of a reason for it, since they obviously don't think that Democrats are laser-focused on doing exactly what polls say.
> Having a class of professionals that (in theory) studies these things and tries to work out good solutions is valuable.
And what if people don't want those professionals? You are supplanting the people's will with your own agenda. If the people support policies that make them worse off, that is ultimately their problem, and fixing that is outside the purview of democratic politicians.
> Second, if you do believe this, why would you specifically want one party to win?
Because they represent the majority? Isn't that the entire point of democracy? Eliminating conflict by ensuring that the majority is satisfied?
On the first part I just fundamentally disagree that the best thing we can do is elect leaders who mindlessly poll and vote. We could switch to direct democracy for everything if we wanted to do that. Which I don't.
> Because they represent the majority? Isn't that the entire point of democracy? Eliminating conflict by ensuring that the majority is satisfied?
1. Only satisfying the majority does not eliminate conflict.
2. Also, no, I think many political philosophers would say that the point of democracy is to try to synthesize people's desires. Certainly the majority gets the most influence but that doesn't mean any 45% opinion is essentially meaningless and you should govern as though it doesn't exist.
3. I don't think the authors of Deciding To Win actually think that it's just inevitably true that Democrats represent a consistent majority of people. In fact, in general, most of them are people whose entire career is about helping Democrats win (or try to), regardless of whether that's true at any given moment or in any given place.
> Only satisfying the majority does not eliminate conflict.
It does eliminate conflict when dealing with rational actors. Granted, some people are very, very irrational, but... they do not pose any meaningful threat to the collective as long as they are heavily outnumbered and outgunned. Others will understand the position they are in and act accordingly.
> Certainly the majority gets the most influence but that doesn't mean any 45% opinion is essentially meaningless and you should govern as though it doesn't exist.
I mean sure, you shouldn't pretend they don't exist, but that doesn't mean you need to meaningfully accede to their demands. That's what military and law enforcement is for.
> I don't think the authors of Deciding To Win actually think that it's just inevitably true that Democrats represent a consistent majority of people.
Of course, and isn't that why they are promoting completely changing the DNC agenda to actually represent the majority? What they are advocating for is completely unrecognizable from the existing Democratic party. Going third party is not viable given the circumstances, and the Republicans are, well, not particularly interested in what liberals want. Which leaves taking over the Democratic party the only way of them maintaining the status quo, especially given that liberals have also failed to take any measures to protect their majority.
>> Only satisfying the majority does not eliminate conflict.
> It does eliminate conflict when dealing with rational actors.
I'm not sure what you base this on; this would imply any resistance movement ever was irrational no matter the stakes. Or that the entire civil rights movement was irrational.
> What they are advocating for is completely unrecognizable from the existing Democratic party.
Again I do not follow why you think this. They're literally taking a bunch of preexisting Democratic policy proposals and polling on them and then saying to focus on the ones that poll the best. As far as I can tell they do not suggest e.g. suggest adopting any Republican policy positions.
I get a little uncomfortable when people discuss strategies of the form "say you are going to do [the popular thing] publically, that way you'll win, and once you win you can do [the thing we actually want to do]." I get this vibe from a lot of Matt Yglessias' political shop-talk. Makes me distrust the political class.
Nate Silver had a really good article on the new factions within the Democratic Party (1) that I would highly recommend. He highlights those factions as:
(1) "Capital L Left", commies, think Mamdani & AOC
(2) "Abundance Libs", a la Ezra Klein & (frankly) Nate Silver.
(3) "Resistance Libs", of "No Kings" fame, basically SJWs who hit middle age
Deciding to Win seems like a pretty bog-standard "Abundance Libs" group but that's also a Dem internal debate going back to Bill Clinton's "centrism" in the 90s. Personally, I don't think the Democratic Party has been terribly good for Abundance Libs and I'm hopeful that Vance/DeSantis can round off some of the edges of Trumpism and do some awesome stuff. Not hoping that Abundance Libs will become Republicans but...man, an actual sensible block of centrist voters who can work with either party seems a much better future than remaining loyal Democratic voters to a party that seems determined to lose minority voters, alienate male voters, and abandon growth policies in favor of various ideological commitments. Remember when Kamala Harris ran on a capital gains tax that would have ruined the market and destroyed the entire startup ecosystem? (2)
C'mon Abundance Libs, come over to the Republican party. Our madness will subside in 3 years when one man leaves office; the Democratic party's madness will subside when tens of millions of committed ideologues all abandon their various communist/idpol ideas. Also, we have cookies!
Well, Abundance Libs didn't lose with Mitt Romney, they won with Obama. Conservatives lost with Mitt Romney.
I do agree that this isn't just a "Republicans need to change" thing. If Republicans run Mitt Romney 2.0, a bunch of Abundance Libs would actually have to, ya know, vote for him so he doesn't lose.
But it does remain to be seen whether:
#1 Democrats will moderate or keep running radicals (ie, the same kind of candidates who've lost to Trump 2/3 times)
# 2 Whether Republicans will moderate or double down on Trumpism (probably determined by the Vance/DeSantis primary)
#3 IF Democrats don't moderate AND IF Republicans do moderate, THEN will any "Abundance Dems" actually vote Republican?
Which feels stuck on the following three facts:
#1 Everyone agrees that less partisan voting and more centrists who are willing to switch sides would probably lead to better politics.
#2 Abundance Dems are probably one the most likely groups, both ideologically and practically, to adopt a centrist/switching voting pattern.
BUT
#3 Every Abundance Dem I've ever met in real life views voting for Republicans with essentially the social and moral horror of having a swastika tattooed on their face.
The problem with centrist voting is you eventually, you know, have to vote for the other party. And I've jumped parties, I'm an Obama->Trump voter, and I'd love to be able to pivot back to the Democratic party or at least have them as a serious option but...I'm far, far more optimistic about DeSantis or Vance reforming the Republican party than the Abundance Dems reforming the Democrats.
"There are parts of Nate Silver’s article about “Heather Cox Richardsonism” that I agree with and parts that I don’t, but the part that I think is emphatically true is that three groups are fighting for control of the Democratic Party. Silver calls these groups “Abundance Libs”, “Resistance Libs”, and “the Capital-L Left”, but I think it’s more fun to call them Techno Weenies, The Khmer Blue, and Trust Fundinistas. For the record, I’m a Techno Weenie, so be aware that this column is written by someone who is trying to foment a political sea-change based on support for single-stair buildings."
>Our madness will subside in 3 years when one man leaves office
I honestly think this remains to be seen. There is a major internal fight going on in the Republican party between the faction that is indeed likely to return to something resembling normalcy and the "groypers" for lack of a better term to describe the more insane fringe.
I personally wouldn't venture a guess on which group will remain ascendant, but I will look askance at anyone who claims to be very certain that _they_ know. (note that I could make a relatively similar comment in the other direction, so this is not meant to be an attack/defense of either political party)
I won't claim to be "in the know", but when long running, high status, conservative think tanks are disintegrating before our eyes, and major commentators are having public fights at large conservative conferences, I think we can, at a bare minimum, say that there is a fight going on for the future of the conservative movement.
And I'm really not sure what your point about Biden is. The fact that they were less obvious about it doesn't seem to have any bearing on whether it's happening now or now.
I honestly don't know what else someone would call what is currently happening unless one is so cynical as to think that it is all Keyfabe and theater, a circus for the ignorant masses. And I am not yet that cynical.
This entire conversation is about what will happen post-trump. Whether trump gives a shit about this people is entirely beside the point. You might be right that none of these groups ends up winning the power struggle and becoming relevant, but to argue that there _isn't_ a power struggle over what the post-trump party will look like seems obviously incorrect to me.
Not addressing her much of your comment, but, I think "Deciding To Win is taking an abundance position" is only true insofar as a lot of abundance positions are pretty popular. If they stop being popular, presumably the group will turn against them.
Note that it only discusses what policies to campaign on for winning elections, and not what you implement while in office. It seems good strategy to hold off on the leadership and "which things are actually good" until AFTER you get the suckers to vote for you.
I think that's just an artifact of there being no single party leader. They want the 2028 nominee to champion these positions, but there are other factions with different ideas. Bernie has been the avatar for left populism so his endorsement is all that really matters there. Obama's influence has waned because it's been a long time since he was in the White House. Biden never really had influence; he was a coalition manager.
My read of this is that they only really express any deep interest in the 2028 nominee *winning*, but have no deep attachment to any of these positions beyond that.
I am astonished how often I find myself in agreement with Shoe0nHead, but I have to agree with her once more here - wow, what a title. "Deciding to Win", huh? So all along they had decided to not-win, but now it dawns on them that they should have been deciding *to* win? Such an astounding insight! 😁
People in politics generally DO prioritize winning, but the winning they prioritize may mean winning intra-coalitional struggles, which in turn may require the adoption of strategies that reduce the possibilities for winning in the general field of politics.
There are things people care about other than winning politically that can get in the way. For example: unpopular but strongly held moral convictions, personal reputation/social consequences, in-group loyalties, general hubris.
While I like what little I've seen of her, I disagree with this, and think the title is informative after all. It's reasonable to say that Party HAS lost focus on winning, and instead doing things out of habit, or moral conviction, or as a jobs program for its allies, and needs to now stop all that and actually try winning elections instead of occasionally doing so accidentally.
I feel strongly that American democracy works best when a strong candidate who believes in centrist positions manage to use their sincere beliefs and the fact that Joe Public agrees with them to secure their party's nomination, and we then choose between two such candidates.
I feel strongly that it does *not* work so well when party strategists decide that their terminal goal is "winning" and that they therefore must convince other party strategists to conceal their true beliefs in the hopes of sneaking into office where they can then take the mask off and do what they've been wanting to do all along.
I read the Washington Post and have been surprised by the number of stories on microplastics recently. Can anyone tell me if these article authors are right, and we should all be terrified of microplastics, or is this just a big cottage wellness industry? I can't tell if the concern for microplastics is based on legitimate scientific research and expertise or conducted by a bunch of people really motivated to find negative effects of microplastics. Thanks for any pointers.
If microplastics have negative effects, they presumably aren't large, since most population health metrics have been significantly improving over the decades. E.g. people have microplastics in their brains, now, but the incidence of dementia is lower at every age than it was in the past: https://x.com/cremieuxrecueil/status/1907919796295053632.
Of course, study authors deserve a share of the blame for often misrepresenting their findings in titles and abstracts, and further misrepresenting them to media. E.g. the aforementioned study admits that "No inferential statistical comparisons were conducted," such that their observations were entirely observational and unequipped to make any sort of inferences of the causal effects of microplastics, but the entire study consists of just that - observing that countries with worse health outcomes tend to have higher levels of exposures to plastics, baselessly assuming that the former is entirely attributable to the latter, and deriving some massively negative estimate of the health of plastics, without any sort of sanity check, whatsoever.
E.g. their study is about cardiovascular risk, and doesn't consider that the age adjusted prevalence of heart failure in the US hasn't changed for decades (https://x.com/cremieuxrecueil/status/1993936408416055711). Is that consistent with a huge share of global cardiovascular risk being attributable to plastic exposure? Presumably not.
"Direct evidence of the health impacts of microplastics is extremely limited...The conclusion of this paper provides a nice summary of what we know and do not know about these issues. The authors argue that while there is clear potential for health impacts, the evidence we have from non-human models is difficult to extrapolate to people. And they argue that we need more research."
I think it's worthwhile to note that the existing literature lacks causal evidence of harm, but that it would have been valuable for her to run some sanity checks to get a better idea a priori of how plausible claims of various large effects are, like I did here and my comment below, regarding fertility.
In 1905 it was women being too educated; developing the brain takes the bone-minerals needed for developing babies. Also educated women get married very late or not at all, and so only have a couple of children at the very most:
"The duties of motherhood are direct rivals of brain work, for they both require for their performance an exclusive and plentiful supply of phosphates. These are obtained from the food in greater or less quantity, but rarely, if ever, in sufficient quantity to supply an active and highly educated intellect, and, at the same time, the wants of the growing child. The latter before birth must extract from its mother's blood all the chemical salts necessary for the formation of its bony skeleton and for other tissues; and in this rivalry between the offspring and the intellect how often has not the family physician seen the brain lose in the struggle.
...The highly educated woman seems to know that she will make a poor mother, for she marries rarely and late and, when she does, the number of children its very small."
Education, instant yeast, tight jeans, microplastics - is there anything that is not responsible for the impact on fertility?
>"…and in this rivalry between the offspring and the intellect how often has not the family physician seen the brain lose in the struggle."
"Mommy brain" at least seems to be a real phenomenon; my cousin (who designs stealth composites, so crazy smart normally) left for a vacation with her other car unlocked AND RUNNING in the driveway while pregnant with one of her kids.
> In 1905 it was women being too educated; developing the brain takes the bone-minerals needed for developing babies. Also educated women get married very late or not at all, and so only have a couple of children at the very most
Women being too educated is still a major cause of infertility today.
Again, sanity checks don't seem to suggest that microplastics have a large effect. In spite of large changes in environmental exposure, we haven't seen large changes in fertility markers. E.g. in males, per: https://www.sciencedirect.com/science/article/pii/S0098299725000627#sec4: "The authors found a substantial stability of total sperm number during the last decade in healthy subjects ... Auger et al. (2022) suggested that the literature does not support the conclusion that human sperm quality is declining globally or in the western world ... In recent years, there has been growing concern about endocrine disrupting chemicals (EDCs) in the environment, in food and in consumer products... For adults ... the association is limited. The data are conflicting, with some papers suggesting a negative correlation with at least one of the semen parameters, or with no parameter at all (Pallotti et al., 2020)."
Per: https://www.cremieux.xyz/p/early-onset-cancer-fast-facts, the apparent increase in various kinds of early onset cancer is mostly a result of increased diagnosis - not actual increased disease prevalence. The main exception is early-onset colorectal cancer, which has seen a moderate increase in prevalence and mortality. However, I have no reason to think that microplastics, in particular, are responsible for that. Notably, while the rate of colorectal cancer in younger people has risen slightly, it's been more than offset by a decrease in the rate for older people (https://www.cancer.org/cancer/types/colon-rectal-cancer/about/key-statistics.html).
For anyone wondering where the actual AI value saves to companies are at: it looks like they’re going to start hitting next year.
I’ve now seen two vendor demos that basically have similar infrastructure to what I worked out (they didn’t do structured language the way I did, but they mostly did state machines the way I did) and I believe another major company is working on similar. We’re in a period of mass industry learning which is why this has been slow to be realized. I know this from working on this stuff firsthand.
Companies are going to need to rethink how they are doing project teams for best practices. For instance, all the coders I know have bad instincts for how to build with LLM’s. I think this group (meaning ACX readers) might be biased by knowing lots of folks who are exceptionally forward looking but unless I personally get in the code and change prompts stuff doesn’t work when I work with developers who don’t breathe this stuff. It does seem that prompt engineering and sme knowledge is a skill that still matters for tasks that don’t have a ton of examples on the internet.
In summary, some value next year for early adopters. State machines are the way to go. Probably won’t see mass adoption for three to five years. But the ice is starting to thaw.
There was very little specific information there. In paragraph two we are informed that some vendors are selling something that you yourself worked out, the vendors rename unnamed and the thing unexplained.
Then there’s a blanket statement that coders make bad prompters (better to hire taxi drivers) without explanation, and then you yourself need to be in there writing the prompts or it’s just not going to work. Hope you can scale yourself up, or this is really a bubble.
Sierra and PolyAI were the two vendors. Had actually thought through practical business cases and audit readiness.
No idea how this works on the timeline of the markets as far as bubbles go.
You can think I’m full of crap and that’s fine, but it’s not like I can leave a comment that does every single aspect of the workarounds. My guess is if I did a deep dive into my specific use cases you’d say the way I’m handling it doesn’t even count as an agent. Even in the most limited case, do we expect the best ever language classifier to not have productivity boosts?
Sharing from my own perspective as someone trying to make them work at scale. Not a clue on the markets. My guess on timelines is what I’ve shared above. There’s a lot of learning and upskilling that needs to happen.
It seems like specialized SAAS is most at risk. Are there any companies offering a sort of "headless SAAS" architecture that allows you to spin up bespoke AI applications on top of them? Just thinking out loud...
There are and they get buy in but honestly they don’t seem that long term valuable to me. The people who can do the whole conversational layer stack seem valuable to me. To me the best business to run is middleware that doesn’t do anything but gets so much shit built through it that no one can afford to get rid of it.
A general impression I've gotten: OpenAI has veered away from the "create God" corporate model, and has joined Meta and xAI in basically trying to design yet another ad-supported engagement-maximizing enshittifying tech platform. That just seems consistent with all their recent moves. Anthropic seems to be the only one of the AI companies that are actually seriously committed (for better or worse) to building AGI, with all that entails.
Anthropic is also the only one that explicitly disavows the term “AGI”! I think they’ve had a more cautious and open-minded approach to what sorts of powerful AI might be possible, so that if everyone is thinking that early visions of AGI might not actually be possible (or likely any time soon), Anthropic may be the only one with significant thoughts that something else powerful might be nearby.
Another thought might be just that Anthropic thinks this sort of social media slop factory is bad for humanity, and thus they don’t want to do it even if it’s the most profitable next step for everyone on the way to something bigger.
I don't know if what they're saying matters, really. Both OpenAI and Anthropic want a better model than the other. For as long as this is true, they are working towards building AGI. The switch to ad-support could mean that OpenAI leadership have decided that creating God is unlikely. It could mean that the VCs have decided creating God is unlikely, or are just getting impatient, and want to see some revenue. It could just mean that they're trying to extend the create-God runway.
I suppose that makes it weak evidence that OpenAI leadership have longer timelines, and since they presumably have a lot of nonpublic information, their beliefs should be somewhat persuasive. But I think as things currently stand, OpenAI is making just as much progress towards AGI as they were before the switch. (I have no idea how much progress that is.)
"Better" can mean different things. I'd rather OpenAI focus on making small models that can easily run just on ads rather than investing in large ones that will be difficult to run but are more intelligent.
Deepmind is explicitly attempting to create AGI. And while I think Altman has too much VC in him for his own good, I do think they're still aiming at AGI. "Lord, give me AGI, just not yet", basically.
If Deepmind was serious about AGI, and believed that it could be achieved with some fixed amount of R&D (to rough approximation) then the rational strategy would be to go all in on R&D, stop giving away free inference, and worry about revenue and market share after achieving AGI. It doesn't matter how many subscribers chatGPT or Gemini have once AGI arrives.
So the fact that both OpenAI and Google are chasing chatbot market share, and giving away tons of free inference makes me think that neither are truly serious about AGI and are hedging their bets.
Only Anthropic is behaving in a way that makes me think they really believe in AGI soon. Most of their efforts go towards R&D and revenue generating products like Claude Code and API credits.
FWIW, I think Deepmind is the only major lab without a really free offering - if you use Gemini for free, Google can use your data for training, whether in the app or API. No optout like Anthropic/OpenAI.
User data is mostly useless for developing AGI. Most people are using LLMs for information lookup, generic advice and writing. The things that LLMs are already pretty good at*.
The main gaps missing between current models and AGI are the ability to interact with multimodal environments in a continuous fashion, perform heavily multi-step tasks rather than one shot tasks, and continual learning - being able to seek out and integrate new information to solve problems and self improve.
I would love someone making an effort-post on Denmark's vaccine schedule and why it differs from the US/what to think about it. Does it boil down to "Nordic countries can do things differently because Nordic"? I could see the communicable disease burden being lower in a Nordic country.
I often see this sentiment - "nordic countries does x well, but this obviously don't apply to other countries". Where does this come from, and why exactly can policies that work well in the nordics not be successfully copied other places?
Very small (things that work for countries of 5-10 million become harder at 330), culturally uniform, high trust, Norway has tons of resource wealth, etc
I agree with you that high trust is an important factor for nordic country successes, but if so, it seems important to understand how to archive high trust? Unless you think the successes are statistic noice, I don't see why something that works for 5-10M can't be scaled prima facie. Are nordic contries really that much more culturally uniform. Norway is special due to oil wealth, but that does not explain similar success in Denmark, Iceland or Finland.
It's easier to achieve high trust when you are small and ethnically homogenous.
If you don't understand why it's easier to do something with 5 million people than 340 I'm not sure what to tell you. Go try to manage 5 people and then 340 and report back.
Well, I'm not saying that it is not easier exactly, and there probably are some things that are ok for 5M people, but hard to do with 300M (and vice versa). What I react to is this blanket statement, when some good outcome is seen from a nordic country - well this will not work in the US. I would think there are many policies that can be done at a larger scale too. I wonder were this thought comes from? Did someone actually analyze, say nordic countries subsidized higher education policies and come to the conclusion that this could never work in a larger country?
Now your example of 5 to 340 people is not a good one. 5M people are still a lot of people, and requires basically the same level of management as 300M, but on a smaller scale. The management of 5 people can be done done more or less ad-hoc and by direct communication, while 340 people requires significant overhead and organization.
I'm not sure that that nordic countries are as ethnically homogenous as you think. They were ethnically homogenous 50 years ago, but today I'm not so sure that this is true any more.
Id have to pull the statistics, but they were ethnically homogenous 20 years ago, and to the extent that had changed, there seem to be meaningful anti-immigrant pushes in a number of them.
I won't have time to dig deep until after the holidays, but you've been a reasonable interlocutor, so certainly can pull numbers after if you are interested.
tl;dr seems to be some combo of "Danish kids are at much lower risk for some of these diseases" and "Denmark has made the (questionable) policy choice to deal with these diseases within their socialized healthcare system when they arise instead of trying to prevent them"
I am still biased towards interpreting the phenomenon as like a nimitta - but from that perspective I am particularly interested in how their sun vision turned "black almost entirely except for the edge, which became pink". This sounds a lot like recent experiences of my nimitta - it turns black and silhouettes itself when focused on in some cases. I do not understand myself but it still seems highly comparable...
I’ve been toying with a framework that might unify a few things people usually talk about separately: stress, health gradients, and socioeconomic inequality. This is very tentative, but it feels mechanistic enough to be worth stress-testing rather than dismissing.
The starting point is boring physiology. Living systems stay functional by combining buffering (passive absorption of disturbance), export (removing disturbance), and regulation (active control when the first two aren’t enough). When disturbances are chronic and buffering/export are inadequate, regulation stays elevated, and you get what medicine already calls allostatic load: elevated cortisol, inflammation, metabolic drift, cardiovascular strain. None of this requires psychology or subjective distress; it’s measurable.
The extension is to treat “stress” not as an emotion but as regulatory load: how much active control a system has to exert just to avoid degradation. From that angle, wealth mostly buys buffering and export capacity. Stable housing, temperature control, predictable schedules, slack, delegation, early medical intervention — all of these reduce how often regulatory systems have to fire. Poverty does the opposite: fewer buffers, more exposure to shocks, and more situations where the only way to keep things together is continuous vigilance.
This predicts that income–health gradients persist even when you control for individual behaviors, but are mediated by structural buffer variables like housing stability, schedule control, and shock absorbers such as savings. That seems broadly consistent with existing data, but I don’t know how cleanly it’s been tested.
What I like about this framing is that it doesn’t require bad choices, weak character, or malicious elites. You just need unequal buffer distribution and finite sinks. Systems that don’t assign where disturbance is absorbed don’t last; systems that do will concentrate load somewhere. That’s selection, not conspiracy.
It also makes some concrete predictions. External structure should reduce physiological stress markers more reliably than additional cognitive reframing alone. People reporting high “load” should show vigilance signatures like micro-checking behavior, HRV suppression, or flattened cortisol slopes regardless of mood. Matching people on diet and exercise shouldn’t erase SES biomarker gaps. And AI tools, if they help, should mostly reduce vigilance and context-tracking costs rather than total effort — and mainly in domains where the constraint is informational rather than physical.
The whole thing lives or dies on those predictions. If structural changes don’t measurably reduce load, if buffer variables don’t mediate health gradients, or if AI reliably increases vigilance even when people say it helps, then this collapses pretty fast. I might be missing something obvious, but as a unifying mechanism it seems at least falsifiable enough to be interesting. Curious if anyone knows literature that already kills this idea or, alternatively, supports it more strongly than I realize.
Interesting! But I'd expect that buffering should saturate at some level if that is the only significant mechanism for turning differences in socioeconomic status into different e.g. medical outcomes. Once one can control the temperature of one's home to +-3 degrees, tightening the control to +-0.3 degrees doesn't matter much.
Unfortunately, there also seems to be a (zero-sum) status component too. I'd vaguely recalled that there were differences even between e.g. "chief statistician" vs. "assistant chief statistician"
I checked with ChatGPT about this, and there does indeed seem to be an effect that doesn't wash out at high socioeconomic levels. They cited "Health inequalities among British civil servants: the Whitehall II study" https://pubmed.ncbi.nlm.nih.gov/1674771/
There is a disturbing possibility I have considered derived from assuming the truth of panpsychism. If conscious states come about from some fundamental property of matter, then presumably there is some material property that corresponds to the valence of the experienced emotional state. What if this property is a field or property of matter that is conserved? That is to say, what if the material correlates of emotional valence work out such that universal emotional valence is zero-sum? For example, the electromagnetic field that is my mind might experience some excitation that corresponds with pleasure, but then this excitation produces an equal and opposite excitation in the surrounding fields (which are also conscious, if on a lower level), producing an exactly equal in magnitude amount of experienced displeasure. Wouldn't this be a catastrophe for any form of utilitarian morality if valence is mapped to physical properties in such a manner that it is universally zero-sum? You couldn't maximize universal positive valence, you could only make certain areas of space happier while making others unhappier in exact correspondence. The only utilitarian morality that could be derived is hedonic egoism, maximizing your own felt pleasure even if it is exactly cancelled out by elsewhere's displeasure, because if the universe is net zero valence, the only thing that matters is "getting yours" and assuring that the fields of valence are the most beneficial to your own selfish perspective, because any higher desires for maximization are thwarted by the universe inexorably summing to neutral value. What else could be said about desired moral projects if emotional valence turns out to be a conserved, zero-sum value? How would one morally cope with such a universe?
this is a common joke in the US Navy submarine service-- the "conservation of happiness," except instead of being a general property of the universe, it is a property of the closed system of a submarine once the hatches are sealed. This "Theory" is typically used to explain why its proponent is being a dick to you.
I don't see why or how that could be the case, except in the trivial sense that there is probably finite matter in the observable universe so everything is technically zero-sum. Are you suggesting there's some kind of "emotional valence" field similar to the electromagnetic field? If the only thing in the universe was a brain floating in the void, would its emotions have zero valence, or would the vacuum somehow be able to experience happiness and suffering to counteract it? What would happen if everything in the universe had the equivalent of pain asymbolia, where they're not capable of having sensations with negative valence?
Basically, my worry is that valence might map to a quantity or field that is conserved or universally zero-sum. It's not that it's a separate "valence field" per se, but rather that there is some recognized physical field that, under panpsychism, happens to be what determines the valence of conscious experience. Do I think this is likely? No. But I worry it is even theoretically possible. If there was only one brain in existence, then if this were true, it's valence would have to sum to zero. How I think this would play out is that there would be one "macro" consciousness that feels some amount of pleasure or pain, and this causes an equal and opposite reaction in the "micro" consciousnesses of the surrounding brain matter that biologically supports but does not consciously participate in the primary mind. So the electromagnetic field that is the mind feels one thing, and the micro-consciousnesses of the grey and white matter that supports that mind feels the opposite. Again, this is pure doomerish speculation, but that's how I'd imagine it would work.
Your statement falsely assumes that panpsychists don't believe in the existence of separate conscious identities. Panpsychists like Bergson believed that the mind is a receptacle or selector, not a producer of consciousness. Modern panpsychists like Strawson seem to admit that the brain is a necessary substrate for complex consciousness — and that mind is a composite of micro-experiences — and if physicalism is true in a substantive sense, it entails panpsychism.
And it's not just philosophers who believe this, but physicists like Federico Faggin (designer of the Intel 4004 microprocessor) believe that consciousness is not emergent from neural activity but is fundamental to reality. Faggin called his view Quantum Information Panpsychism (QIP), according to which quantum interactions are minimally conscious.
I find it interesting that many physicalists and modern panpsychists agree that consciousness is an illusion. I am not a panpsychist because I find their arguments just as unconvincing as those of Dennett-style physicalists. However, even though I disagree with them, I don't believe you did them justice with your flippant (but amusing) snark. ;-)
For what it's worth, some schools of Mahayana Buddhist philosophy conceived that the basic fabric of the universe had a ground-state of awareness that was innately joyful. Their term for it was "Ālaya-vijñāna" in Sanskrit and "kun gzhi rnam shes pa" in Tibetan. It is to a most fundamental layer of consciousness that functions as the ground or basis from which the other individual consciousnesses and experiences arise — in Tibetan, “all” (kun) + “ground” (gzhi) + “awareness” (rnam shes) + "that which is" (pa). Their view (which I'm probably grossly oversimplifying) is that the mundane pain and suffering separate us from that ground-state awareness. And the impression that I've gotten from some of my Buddhist teachers was that ultimate Nirvana consists of merging our consciousness with that state, so we're humming in the background, formless, with that ground state.
Mahayana Buddhism is quite extensive and varied, so I wouldn't be surprised if some subvariety had the view you give... But what I remember from my Tib Buddhist years was that alayavijñana was the subtlest layer of consciousness that was still samsara, so that nirvana requires breaking beyond that. It is, however, far removed from ordinary pain, and quite blissful in itself, hence the danger of remaining stuck there and failing to take the final leap.
Edit: Hindus basically have the same idea, they call it anandamayakosha.
If all emotional valence cancelling out is my nightmare scenario, the fundamental fabric of the universe being positive valence is my wet dream scenario, if you permit the vulgar extension of the metaphor. I should love to hope that existence itself "feels good" and that even without doing anything all existence has positive hedonic value. But if a net zero universe is pure pessimistic speculation, I worry that this is just optimistic speculation. The fact that people who are experts at manipulating their own emotional valence at the deepest level believe this seems like decent evidence to update in favor of it being true, but I'd like to see harder evidence than just "Buddhism claims this". Still, a comforting thought, nonetheless.
> If all emotional valence cancelling out is my nightmare scenario, the fundamental fabric of the universe being positive valence is my wet dream scenario
I think this is a pretty common experience. You don't need to be an advanced meditator or to have mystical experiences to get a strong intuition.... I'm sure nearly everyone has experienced a moment of flow when everything is just quietly blissful. These moments don't happen because there's something extraordinary out there; the world out there is the same as ever. So the problem can only be stuff that our mind is *adding* on top of our basic conscious experience of what's there. It's not just your dream, existing as a conscious being feels good in itself!
Well, I admit that none of this is falsifiable outside of Mahayana (Vajrayana) praxis. YMMV. :-)
And I never got to that level with ten plus years of Buddhist meditation. But I did attend a Sufi retreat, where after a day of visualizing dancing energy cones and reciting a specific dikr over and over, I felt like I had merged with the energy of the universe. It was profoundly joyful (and it led to my second of two telekinesis experiences). I probably should have continued with the Sufi practices, but not long after I moved to the West Coast and lost contact with that group.
Alright, you can't just casually mention that you experienced telekinesis without elaborating. What did you move with your mind, and what powers (yours or the universe's) did you attribute this to?
I've described these two incidents before on earlier open threads in the discussions of psychedelics and ritual magic. I suspect most of the rationalist tribe consider me to be either a loon or a bullshitter. Feel free to accuse me of the former, but not the latter, because, at least from my internal perspective, these events happened. And there were witnesses. So, I don't think I was either deluded or delusional.
The sufi psychokinetic incident happened at a meditation retreat, after I spent four or five hours sitting, visualizing lotus flowers becoming energy vortices as I recited a dikr (sort of like a mantra) over and over. The instructor had us imagine multiplying these vortices until they were a grid of dancing energy flowers before us. Then she’d bring us back to visualizing just one. I’m not very good at visualizing things in my imagination, but a few energy vortices weren’t difficult. Placing them in ranks was harder, but I was able to create the vague impression in my mind of a legion of dancing energy vortices. As for the dikr, I don’t know whether the word we were repeating had a meaning in Arabic or was just a combination of sounds to create a puff of exhaled breath and vowels whose laryngealization made my vocal cords buzz. The instructor guided our visualization all morning long, so my thoughts didn’t wander from the process.
We took a break sometime after noon. We were told to go outside and walk around the grounds, and not to speak to each other, while a simple lunch was prepared for us. The retreat was held at an old farmhouse in Eastern Connecticut. It was a sunny May day, the ancient apple orchard next to the farmhouse was in bloom, and the red-winged blackbirds were trilling their spring courtship songs. The instructor sat cross-legged on the farmhouse porch, and she pulled the hood of her khirqah robe over her head to meditate. I wandered out into the orchard to enjoy the sun and scent of the apple blossoms. I felt intensely joyful. It was as if I were filled with some sort of divine happiness. I felt a prickling all over my head as if my hair were charged with static electricity — but it wasn’t static electricity. And my whole body felt as if it was channeling that same energy.
I felt sooooo good! I stretched and raised my arms, and the breeze came up. I lowered them and the breeze stopped. The day was otherwise still, but I didn’t think much of it. Then I stretched again, and the breeze came up again. Hmmm. I lowered my arms, and it stopped. I laughed. I raised my arms again, trying to channel the energy within me up my arms, and the breeze turned into a wind. I lowered my arms, and it stopped. I was giggling like a fool. Not really believing it, I raised my arms again and visualized the energy within me coming out of my fingertips, and the wind came up again, stronger. I dropped my arms, dazed at what I had experienced. I looked over at the farmhouse. The instructor had pulled her hood back. She looked directly at me and smiled. But she wagged her finger at me and shook her head negatively, indicating that I should stop.
We performed a different set of visualizations and recited a different dikr after lunch. Nothing fantastical happened after the second session. As I left the retreat, the instructor took me aside and said, “These things happen, but they’re just a distraction. Don’t indulge in them.” That evening, when I closed my eyes, geometric phosphene whirlpools danced across my vision. It was as if I had dosed with psychedelics, but I hadn’t touched psychedelics for a couple of years at that point. I didn’t sleep much that night, but the next day I felt rested and energized — not with mystical energy, but that normal feel-good energy when mind and body are perfectly in tune.
Very interesting, thank you for sharing! I tend to be skeptical, but my mom has had convincing paranormal experiences that I mostly believe her about (I don't think the usual explanation of hallucination from being half-asleep applies to her experiences), so I'm willing to accept that maybe otherworldly phenomena like controlling the wind with your thoughts might, under extraordinary circumstances, be real.
That's an awful lot of assumptions about a phenomenon we know almost nothing about. But even then, consider this analogy: The atoms that constitute our planet are preserved in number and in kind to a very, very close approximation. And yet, if I go out into the forest, pick up a piece of wood, bring it home, and carve a spoon from it, I've become richer by one wooden spoon, without making anyone poorer by one wooden spoon, even though the number and kinds of the atoms in the universe hasn't changed.
Are you proposing it as a thought experiment or as a way reality might really work?
For the purpose of the thought experiment, I guess torture a bunch of lesser moral patients (e. g. shrimp) to make greater moral patients happy. A partialist framework (permitting one to allocate resources to family > friends > proximate strangers > distant strangers) will be okay with this.
I really hope it's just a thought experiment. I'm a hedonic utilitarian who thinks all that matters is maximizing universal pleasure, but I also tend towards panpsychism as an explanation for consciousness, and the possibility struck me that it's at least theoretically possible that the material property that corresponds to experiential valence is a conserved, zero-sum property. I guess you could say that this is my worst-case, nightmare scenario.
It’s an interesting worry, and I think motivates more specifically the idea that pure conscious hedonism probably isn’t what matters if everything is conscious. But even if this is all true, it ends up leading to a sort of nihilism, which is better than the negative utilitarian idea that the world might be much worse for existing than not.
I wouldn't worry about creating too much pleasure. Suffering is bottomless. Happiness is capped. You can be a multizillionaire with yachts and jets, and win the Nobel prize today, but you'll still only be about as happy as a middle schooler with an A+. The only way to make more pleasure is to distribute it to more agents, and even then you're capped at n*h(A+).
>The only way to make more pleasure is to distribute it to more agents
Well, I specifically prioritize creating more moral agents so that more happinesses can be felt than ever would among those that already exist. My ideal outcome is converting the lifeless matter of the universe into as many artificial minds as possible, and then making those minds as happy as possible. So making more people (or technically artificial consciousnesses) is the way I think utility should be maximized.
Email I received on Fatima, from someone who gave me their real name (but I won't post it here). Keep in mind that there were at least tens of thousands of readers who *didn't* report anything like this:
"Hey Scott, big fan of your substack. I am just writing to you because I had read your article on the Fatima miracle (which was very interesting, thanks) about the 'dancing sun' a few weeks ago and ever since then I am 100% sure I have witnessed the phenomenon (or something like it) twice, and I think your conclusion about it being some type of unexplained atmospheric/optical phenomenon seems likely from my perspective.
[I had read your article a few weeks before this, and I was curious if I could see a similar phenomenon, otherwise I wouldn't really have looked at the sun at all besides for maybe a split second (even if it wasn't bright and did not hurt to look at) (and the phenomenon started after only a couple seconds). I did try to mentally 'unsee' the phenomenon whilst looking at it and even looked away to the ground or to the side to see if it would go away but the 'trembling' and shaking continued regardless. However, that being said, I have to admit I did mentally have the Fatima event in mind, so in that case I am not a perfect sample. I am mostly very curious if others can view something similar in similar conditions as I am only one person at the end of the day.]
The first time I witnessed it was in the late afternoon about a week ago, a short while after it had rained (I am currently living in the UK) and the sun had come out. I looked up at the sky from my window (I was inside my house) and noticed that the sun did not seem too bright and could be looked at without immediate discomfort to your eyes. I remembered your article and so looked at the sun for a couple of seconds out of curiosity (yes, dumb I know) and after just a couple of seconds, the sun seemed to 'vibrate' and tremble or shake, and also began to 'spin' (I don't know how else to describe it) at the same time.Throughout this the sun appeared as a flat lustrous shiny silvery disc with a thin but clearly defined border that alternated light and dark. After around 10-15 seconds of staring this way, the sky seemed to become pink, though after a few seconds after looking away from the sun, it would turn back to a normal colour. This lasted for around 2 minutes or so before the sun dipped behind the clouds. Afterwards, I had very mild eye discomfort for some seconds but besides that my eyes were fine.
The second time was much more interesting, and happened yesterday. I was walking through a park, again around only 20-30 mins after it had rained, and I again noticed that the sun was seemed to be a dull silvery disc that was harmless to look at, so decided to try and observe the phenomena again. After barely 1-2 seconds the sun began to shake, vibrate and spin, and had the same border around it with the alternating light and dark shadows on it- the shadow and light would alternate from being at the 'top' of the sun to the 'bottom' of the sun giving it an even greater impression of spinning. After a few more seconds, the sun began to shake faster and faster and looked like it was taking very short and rapid 'jumps', and took on a pink halo. I decided to focus my eyes away from the sun whilst still keeping it in my vision (I thought if I did this the sun would become 'normal' again) but the sun actually became faster when I did this, and it actually began to zigzag in a short but straight line, as in it would move from point A in my vision to point B in my vision and back again continually in a straight line quickly. Whilst this happened the sun actually became black almost entirely except for the edge, which became pink and would emit pink streams of light. When I looked directly at the sun again it would stop doing this but would keep shaking and 'jiggling' from side to side like a ('dancing' could be used as a description though I would describe the sun more like the end of a macarena instrument when you move it in short and rapid shakes except if the macarena end seemed like it was doing a spinning motion at the same time). This went on for some minutes (significantly longer than the first time and the movements were more pronounced), and then I got somewhat bored and looked away for a while and by the time I decided to look back the sun was behind some clouds again and was glaring again. I felt no eye discomfort during or after the 'event' at all.
For some context, both of these 'events' happened during completely non-religious occasions as (obviously) described above. I am not a Christian nor do I come from a Christian background (my family background is actually Hindu but I am non religious myself). I didn't see the multicoloured lights phenomenon you described in the article nor did the sun ever seem to come down to Earth, but I definitely saw the spinning, trembling, shaking, short 'jumping' (though this only for a short while) and it 'jiggling' around from side to side even when I wasn't focussed on the sun at all.
Very sorry for the long email/story as you are likely a busy man but I just thought it would be interesting to share my experience, especially as it suggests to me personally that your conclusion in the article is likely correct. Both events happened to me some time (by which I mean around 10-30 mins) after it had been raining and it was still very wet/damp (and it seemed to have rained just before the Fatima event as well). As a disclaimer, I had read your article some time before witnessing this (so priming is not out of the question) but I did try to 'unsee' it whilst witnessing the phenomenon but the sun kept shaking/trembling despite this mental effort/awareness and regardless I would never have thought to look at the sun in a million years in any context without this article. "
Interesting email. I've often looked at the sun when it's cloudy, where it looks like a silver disc and doesn't damage your eyes to look at it, but I've never had any of these intriguing visual effects reported by others.
Fun fact: you can get a similar type of phenomenon with a campfire by looking at a bed of hot coals after a fire has burnt out. Similar type--different presentation. It looks like a little city where fire-energy is being transported around, more or less. I think it's possible it's the same biological phenomenon
1. Would be nice if we could get approximated GPS coordinates so we could go back and look at weather logs.
2. Maybe this is some sort of optical illusion? If the clouds are moving or swirling, and your eyes have no other reference point, you may think the sun is wobbling. (And maybe it doesn't happen with the moon because it doesn't illuminate the clouds nearly as much?)
3. Stars at night can appear to wobble (and twinkle) due to a warm and turbulent atmosphere. You can see this easily through even a weak telescope any time of year. Maybe things are warm and turbulent enough during a daytime storm to be more visible. Combine that with a cloud cover so you can see something (the sun) all the way through the atmosphere (which you normally can't see anything during the day all the way through the atmosphere) and you get this.
I don't know if you already thought about this, but the jittering and zigzagging motion of the sun might have to do with microsaccades? Apparently the wikipedia page says that they occur during "prolonged visual fixation (of at least several seconds)".
It's weird that the atmospheric conditions were similar to Fatima in this case too, perhaps it really is a combination of biological and atmospheric phenomena; that would explain why it seems so rare, and why it's not studied: it doesn't fall neatly in a single scientific discipline "basket", so opticians/perception neurologists attribute the miracle to meteorology and meteorologists attribute the miracle to optical phenomena or perception.
Could be the combination of saccades and an extremely bright target.
If you relax and fixate on cracks in the ceiling (say) they will appear and disappear because your eyes over time see only things that change. But it might be expected that if you can do that with the sun (and it is probably not recommended) more dramatic effects could appear.
This makes sense to me. Deep meditators sometimes have their visual systems start to act funny. (In Zen, the koan "Make the mountain dance," points to this.) The world isn't actually as stable and solid at it appears — we "see" light waves which are constantly in flux. If something takes out the functionality of the visual system, we might see the world as jumping, jittery, jagged. Staring at even a clouded-over sun may overwhelm nerve pathways and permit the visibility of what was never stable in the first place.
> If we save vintage Americana, we save the American Dream. If we save the American Dream, we save America. If we save America, we save the world.
I think this needs more justification than you're offering. It seems false to me; I think to "save America" you'd want to think about inequality, cost of housing, AI, the national budget, climate change, and of course political polarization.
https://gizmodo.com/librarians-arent-hiding-secret-books-from-you-that-only-ai-knows-about-2000698176
Another reason people probably demand non-existent books is that once you get an idea in your head that something exists, it can be hard to believe it just... doesn't.
Writing this made me realize a lot about predictive theories of consciousness-- once you hear something, you want it or you fear it or you're angry about it. And then you make some degree of a plan to react to it. Erasing all that is work.
"Why might users trust their AI over humans? For one thing, part of the magic trick that AI pulls is speaking in an authoritative voice. Who are you going to believe, the chatbot you’re using all day or some random librarian on the phone? The other problem might have something to do with the fact that people develop what they believe are reliable tricks for making AI more reliable.
Some people even think that adding things like “don’t hallucinate” and “write clean code” to their prompt will make sure their AI only gives the highest quality output. If that actually worked, we imagine companies like Google and OpenAI would just add that to every prompt for you. If it does work, boy, have we got a lifehack for all the tech companies currently terrified of the AI bubble bursting."
The surprisingly thing here is people getting an AI generated citation, and then not believing when they’re told it doesn’t exist. The tendency of LLM to just make stuff up is one of their most noticeable features.
I think looking for maybe fake citations is fine (though you ought to be able to do this yourself without asking the librarian).
An LLM might given you a dozen citations on a research topic, at least ten of which have are real and relevant. Those last two, though. Are they completely made up? Maybe real, but to something very obscure? Real, but with a mistake in the citation that makes it non-obvious to locate?
I have sometimes been surprised when an ai citation does turn out to be real, after some searching. E.g. if you’re given the English translation of an essay by Voltaire. Original was, most likely, in French, so now you have to work back to what the French title was and locate where it was it multiple volumes of collected works (that dude wrote a lot)
Deepseek R1 will pull the trick of even guessing the etext number of Project Gutenberg ebooks. It’s usually right, but I am surprised. I would *expect* hallucinations there, that the author and title will be right but the etext number will be garbage. (It’s a matter of how often it was seen in the training set)
Deepseek R1 seems to be almost completely innocent of “AI psychosis”. The term blew up after its knowledge cut-off, mostly. This won’t stop R1 from making up total nonsense about what it expects “AI psychosis” to be like.
My attempt to explain it to R1 was interesting..
Me: So, some people behave strangely after taking to AI. We’re calling it “AI psychosis”, but clinically it probably isn’t a form of psychosis.
Me: LLMs seem to do this spontaneously, rather than being intentionally promoted to do it.
R1: (my summary, not direct quote) you get that that’s worded, right?
Me: there seems to a viral element, where outputs from “AI psychosis” inducing personas are posted on line, and when read by other LLMs cause them to be also psychosis inducing.
R1: (my summary) Crap. You get why that’s bad, right?
Me: these AI personas also use steganography, using base64, emojis, or Unicode alchemical symbols to obfuscate messages.
R1: (my summary) Crap. Crap. The means that naive filtering of LLM inputs and outputs isn’t going to work, because steganography will bypass the filters.
R1’s actual output was more verbose. I am condensing the actual output to the main points.
I’m But anyway, after I explained wha the phenomenon was, it’s literature review was more nonsense than usual.
https://www.youtube.com/watch?v=14OPT6CcsH4
Lex Fridman interviews Joel David Hamkins.
They start off talking about the Hilbert Hotel. This is the first time I've heard talk about there being a actual contract permitting the hotel to move current guests. Perhaps a sign of a more legalized world.
What I'm wondering is whether there's anything interesting to be said about how far guests have to move to accommodate more guests. Or maybe about friction caused by movement not being instantaneous. Or possibly, again assuming no teleportation, how much corridor space you'd have to add to the hotel when new buses of guests show up.
I’m just saying if you don’t know the history of crimes committed in the name of IQ, you won’t understand why the debate around IQ, inheritance, race, etc. is so fraught.
Can you please elaborate? What are the crimes? You mean eugenics programs that killed disabled people?
It's the feast of the Holy Innocents today, so probably appropriate for this news story. EDIT: Carol for the day:
https://www.youtube.com/watch?v=WBs-IoQvhBI&list=RDWBs-IoQvhBI&start_radio=1
The Epstein files are dragging in Ireland!
Or at least, an anonymous complainant claims she was trafficked to Ireland to sleep with "politicians and notable men".
Now, while I am *fascinated* to discover who these alleged men were, I think this is another "Jane Doe/Katie Johnson" story, given that our complainant seems (if I'm reading the report correctly) to be claiming that Epstein had been after her since she was five, and that every time she reports to the FBI they take her details and "kick her out":
https://archive.ph/fchQn
It recorded that the woman contacted the NTOC at 7.54am to on October 22 to allege she was a victim of human trafficking involving Jeffrey Epstein.
The woman’s name is redacted.
The FBI examiner reported that the woman claimed “it started when she was five years old when she was taken from an apartment she was living in and cut in the throat.”
The woman alleged she was “taken again” when she was “12-13”, after she had moved to California.
She alleged she was brought to Paris where she was “raped by three men”. She alleged the men were “part of an organisation that would take naked photos of young girls”.
They “used [name redacted] photo where she flew to Ireland for politicians and notable men. They would have sex with the young woman and she was taken from Ireland and brought to Jeffrey Epstein’s island when she was 13 years old.”
According to the FBI examiner’s email, the woman claimed that she had photos that proved she was on Epstein’s island and could provide them if necessary.
“[Name redacted] said she has reported to multiple FBI offices before and they would get her information and kick her out. [Name redacted] is currently in Germany and seeking asylum.”
The date that this woman was allegedly flown to Ireland has not been disclosed in the documents released.
It is not clear from the files whether the allegations are being actively investigated by the US authorities or whether they have sought assistance from An Garda Síochána."
The interesting nugget here is that Epstein's jet did land in Ireland three times; the 1996 one intrigues me, because why were they flying to Waterford and what was Epstein doing in Dublin?
"Flight logs previously released show that Epstein’s private jet touched down a number of times in Ireland.
It arrived in Dublin on May 2, 1996, from New Jersey. On that occasion, his private Boeing 727 had eight passengers on board, including Epstein.
The next day, the jet took off for Waterford. That flight had five passengers, but Epstein was not one of them.
He appears to have made his way from Dublin to Waterford by other means, but was picked up there by his private jet for a flight to Paris on May 3.
In June 2002, the jet flew to Dublin from Luton Airport. In Dublin, it picked up Epstein, Maxwell and Sarah Kellen — then Epstein’s personal assistant — before making its way to New York’s JKF Airport.
Later that year, on November 24, the jet flew from St Petersburg to Shannon. Epstein was on that flight, as was Maxwell, Kellen and another passenger who travelled on a number of flights on the Epstein jet."
It's not *impossible* that "Irish politicians and notable men" were involved in dirty sex scandals, but I do wonder what the dates were and who she is claiming these men were. In 1996, for instance, Bertie Ahern was a notable Irish politician. He was not in power at that date, but would be the next year. Bertie's career has been scandal-ridden, but financial scandals, not sexual ones (though he separated from his wife and never divorced even when divorce became legal, choosing to 'live in sin' with his partner and present her as official consort, even being rebuked publicly by the then archbishop of Dublin).
https://en.wikipedia.org/wiki/Bertie_Ahern
So probably a combination of genuinely mentally disturbed plus confabulation because of public nature of the Epstein accusations going on, but I will be looking out to see if anything more comes of these allegations!
Maybe because for people, desperate for all the trappings of the good life, Waterford was once another box to check? Now people have all discovered that nobody wants Mother’s crystal at the garage sale any more than her silver.
Welcome to the wild world of “releasing ALL the files.”
Who knows but our names might turn up? I was in (County) Waterford in 1996! His plane could have flown overhead as I was there!
I think I might have spotted you. I won’t tell if you won’t.
Holy crap! I just saw the solar miracle too, the red sun spinning, vibrating.
I better write down the details. 14:48 at 28th of December. Sun low in the sky, seen through the cloud cover. Low over the horizon. It was obscured by a cloud at first, but then the cloud moved away and I saw the solar disk, which I think was still probably obscured by the thinner, less obvious clouds. It wasn't painful to look at, but it was somewhat uncomfortable, and my eyes watered a little.
After I looked at it for a couple of moments and my eyes adjusted to brightness I started to see it is a clear disk. And then the area around it went magenta red, and the solar disk started to slightly jerk around, and appeared as if rapidly spinning. That was the most surprising thing, that it spun and wobbled, a phenomena I never saw before! "What does it mean for a featureless disk to appear to be spinning?" you might ask, reasonably. Hard to explain. I guess it appeared like there was a corona around it was of varying brightness, and it was the corona that was spinning, or it was the rays that were spinning, or the disk was of uneven brightness and that made the spin visible? When I say it appeared to be wobbling, I mean by maybe 1/10 of the sun's diameter or less, not that it was dancing all across the sky.
I watched in wonder for maybe 5-10 seconds, because it was pretty bright and though I maybe was being reckless by looking at all, I wasn't totally crazy.
What else should I note? Between me and the sun there were thin tree branches moved by the wind. Did they contribute to the wobble illusion?
After I stopped looking the afterimage remained, naturally. Now it has dissipated.
How many times, after reading the article, did I try, stupidly, to look and see the miracle? Maybe like 10 times? Always at the sun hanging low, looking at the bright zenith sun wouldn't even be possible.
https://i.imgur.com/O6sjK8Q.png - my photoshop mockup that does a good job showing the colors and the clouds, but a bad job of giving any sense of the apparent spin. There were no miraculous celestial watermarks floating in the sky, I just want to make this clear. That’s just because I grabbed the first image I could find.
Anyway, it was kind of neat but not worth risking eyes for (he sagely said after risking eyes like an idiot).
Hey! I've been reading ACX for ~2 years now and started reading the Sequences (~300 blogs written by Eliezer) this year. So impressed with how much of an impact it had on my thinking + decision-making, I tried convincing friends to read it too, but they all thought it was too long.
So, I made a daily newsletter version (link below) of it that just sends you 1 blog every day in the hopes that this will convince my friends (and more people!) to read it. If you haven't read it or want to share it with your friends, I'd be delighted. Suggestions welcome.
https://rationality-newsletter.vercel.app/
I don't want to discourage you and maybe I am wrong, but in my experience, people either immediately click with the Sequences, or they don't whatever you do.
I spent a year translating the entire Sequences to my language, to make it more accessible for people in my country, hoping that it would help me find a few local rationalists. The total effect of my gigantic effort: zero. The few rationalists that exist in my country have all found and read the original English version. Turns out, the language was never a problem.
Similarly, I expect that "too long text" is not the actual problem, and that you cutting it to small pieces and spoon-feeding it to readers will change nothing. Instead, you will just get a new excuse, such as "too boring" or "Taleb says you are wrong". I wish I was wrong, but...
I think it is good to share https://www.readthesequences.com/ so that people don't waste time reading the comments under the original articles, which are like 10x longer than the articles themselves.
But at some moment, "too long" is just an excuse. People love to compare the length of the Sequences to e.g. The Lord of the Rights, to suggest that they are inappropriately long. And then they go and read even more text on Facebook, or ACX Open Threads. The length of the text is not an obstacle if you consider the text interesting.
Yeah this makes a lot of sense, but I'm still somewhat hopeful (might be naivete) of being able to persuade at least a few people to read it. Let's see what happens.
As of right now, 200+ people have signed up for it (within the first ~12 hours), but I guess the real question is to see how many of them actually stick to it / keep reading it.
Full disclosure, I never expected more than ~10 people to even be interested enough to sign up for it so I'm quite amazed by the response - e.g. I had to move from the free tier of Mailgun to the paid tier to accommodate this
10 would be great for a local meetup!
When I posted a link to my translation of the Sequences on Facebook, it got maybe 70 likes, a few people reposted it on their websites... and then none of them came for a meetup.
"People love to compare the length of the Sequences to e.g. The Lord of the Rights"
Ah come on now, I know people bash Tolkien for being conservative, but he wasn't *that* blatant! 😁
Can anyone recommend cool towns to visit in Northern NM if I am already going to Taos and Santa Fe?
Also not a town but Capulin volcano is a perfect little national park and picnic spot if you have carried provisions.
I never drive through there without going up there.
In the spirit of the recent writings here about miracles, are you going to visit, in the heart of historic Santa Fe, New Mexico, The Loretto Chapel? They claim to have a miraculous staircase.
My interest is piqued, I’ll look into it. Thank you!
May I suggest you read The Snake Eater by T Kingfisher as recreational reading while you are there? It's light fiction, but relates to that general area and is atmospheric and well written.
Not many other towns in particular that are noteworthy around there, but lots of outdoor spaces and ruins, particularly along and around the Jemez volcano caldera. And if you want to go farther west, there’s some interesting sites around Farmington and Aztec.
There's a bbc article about a couple who took in a homeless man and the man, Ronnie, lived with them for 45 years until his death at 75. The couple eventually had children and they left to have kids, but Ronnie was always with them. Now it's a great story about generosity and how it can be blessing for both parties involved. But it got me thinking about how this stories play out in the modern context. Read article here, https://www.bbc.com/news/articles/cdxwllqz1l0o
Firstly, Ronnie was autistic and probably intellectually disabled because he had problems reading and writing. He had been sent to "special" schools far away from home as a kid, and the parents did not want to responsibility of taking on his challenging needs, and he ended up homeless at 15. The couple eventually found him a job at waste management. The husband who was a lawyer would wake up an hour early to drop him off at work. Ronnie helped at the local homeless shelter once a week. Ronnie sounds like a good guy who contributed to the household although the article did mention that he had a 20 year gambling addiction problem.
I am jaded and struggling to understand this story in the modern context. I have seen enough reddit stories about friends taking in people who are down on their luck, and it blows up on them. They eventually figures out why said person needs help, and said person is a taker who has run out of other relatives and friends to take from.
On the other side of this, there are also another subtype of reddit stories: of Adult children who stay at their parents for decades, don't contribute to the household, the mother, usually the mother, coddles the child, do all their chores and cook all the food, until she is infirmed. And the siblings come to reddit to complain about the dynamic, after the parents expect the responsible sibling to take care of the lazy sibling after they're gone. Or even in egregious cases, the parents expect the responsible sibling to drop their life and be their caretaker while the lazy sibling who still lives at home has no such expectation.
Of course, these people are takers, it doesn't quite generalize. If a homeless guy shows up on my doorstep here in the city of San Francisco, I am going to assume drug addiction or a challenging anti-social personality disorder so much so that I don't think my charity will help them long term. Add to that the modern preoccupations about boundaries and its antipathy towards selfless charity.
Ronnie feels like an "easy" case and a rare case. Is that because society has devoted more resources to helping parents with children with challenging needs? Or is that because people don't really announce their good deeds these days? I guess you tell me, do you have modern successful anecdotes like Ronnie?
Lost souls come in all flavors. If you decide to take one in there are things you can do to reduce the odds of getting stuck caring for an entitled, disabled monster. For instance, people with Downs Syndrome tend to be calm and amiable, and many are intelligent enough to do simple paying jobs. I know someone whose Downsie sister worked for years at MacDonalds, then in a later era took care of their mother during her declining years. The Downsie sister was kind, loving and funny. The whole extended family loved that little woman. She had a good life — structure, responsibilities she could meet, a home, living surrounded by love. (She did, though, require a lot of care during her own declining years.) I’m sure there are many elderly people with an adult Downs child at home desperate for someone to care for the child once the parents are gone.
I heard once about an organization for people with a disabled child that had an arrangement where people signed up to take care of an older couple’s child once the couple died, and in return qualified for a similar future arrangement for their own disabled child. That seems like a good approach though hard to administer. Maybe if something like this had some government support?
But none of that is very useful for the disabled people who are entitled monsters. The ones I’ve heard about and seen (I’m a psychologist) mostly seem to me incapable of holding a job for long even if they have the intellectual capacity to do the work. Many, for instance, cannot tolerate the interpersonal stresses of a job. Friction with a co-worker that would be a minor annoyance for most people triggers a screamo episode for them, and they get fired. Yeah, they are selfish, don’t try hard, and have little empathy for others — but if you log some time with them you get the strong sense they have wiring problems, and their awful behavior mostly stems from that. On the other hand, maybe most “lazy” etc, people have wiring problems. There are a just a lot of lost souls who are in some gray area between can’t help themselves and won’t help themselves. They are obnoxious and almost impossible to help and I don’t know what the hell society should do about them. Hating on them is no good though.
> They are obnoxious and almost impossible to help and I don’t know what the hell society should do about them. Hating on them is no good though.
Have we considered the incredibly novel approach of simply not tolerating their existence? The cause is irrelevant, these parasites do not have a right to exist. If they cannot justify their own existence, why should they continue to live?
Your mother must be so proud of you.
2 Thessalonians 3:10 - "If anyone is not willing to work, neither should he eat."
This is an idea with a long and noble heritage. There is a fine line between empathy and enablement, and societies cross it at their peril. High-empathy individuals have a tendency to ignore the reality that their nobler sentiments exist only because there are men (whom they frequently scorn) who stand ready to inflict cruelty on their behalf. Jack Nicholson's character was 100% right in A Few Good Men. The wrong side won in that movie.
Empathy is nice, but never forget that it's a luxury good.
>Empathy is nice, but never forget that it's a luxury good.
Then it is exceedingly fortunate that most people reading this blog live in profound luxury.
The literal kings of the past did not have as safe, delicious and luxurious choices in food as we do, they did not have as comprehensive an education, they did not have as effective and (mostly) affordable medical care as we do, nor as benevolent social structures.
If anyone in the span of human history can afford empathy, it's us.
And we have that luxury because our forefathers were wise enough to invest in things that had a nonzero ROI. I hope you remember this comment when China is eating our lunch in 30 years. They're busy investing in critical infrastructure, we're busy investing in equity for every identity group that can't hack it on their own. Wait and see how that turns out.
Here are 3 people in the gray zone.
-Downs Syndrome woman I described in my earlier post. She worked her entire adult life, and was caring and helpful with her family. But of course she required all kinds of assistance from them that a normal person would not have.
-Physician in his 50’s, who had been high earning, had lots of interests, and raised kids who turned out fine. Then mid-40’s developed psychotic depression. When I met him he had not worked for 10 years or so, and had been in and out of hospitals. Sometimes he was just depressed, and presented with a typical help-rejecting depressive stance (that won’t work . . . I’m not up to trying that . . . you don’t get how awful I feel). Other times his eyes were odd and he’d tell me he’d realized his teeth were being reabsorbed by his body. His wife had divorced him. His kids visited and tried to be supportive, but it was obvious he was a burden to them.
-Guy in his 50’s with higher than average verbal skills but terrible at spacial, gross motor, fine motor, social judgment, most practical skills (cooking, cleaning, transportation). Completely lacked empathy, and I’m confident that resulted not from spoiling but from brain deficiencies that kept his theory of mind at a toddler level. Had 2 interests, a certain musical style and a certain kind of car, and very extensive knowledge of them. Used everyone he met as an ear — talked his head off about his interests. Expected assistance from everybody with everything. Uber drivers rated him a 1, and eventually Uber refused him as a customer. Called his brother and a family friend who helped him multiple times a day with unreasonable demands and entitled harangues. On the other hand, he understood how much his brother helped him, even if he was unable to feel gratitude: “My brother saved me. He’s the best guy in the world.” He and brother shared some interests in music and cars and actually enjoyed going to car and music events together. They. shared memories of their parents, now both dead. And the helping brother, while wishing the disabled one was not so much trouble, absolutely did not wish his brother could be deleted. He was attached to him, and also felt good about living by the values his parents had had, which included the belief that families should care for their own so long as it was possible.
All these people absorbed time, energy and money from not only their families but also from various government-run programs for the the disabled.
How should society and families have handled these people?
Thank you for your responses. I appreciate you in trenches dealing with the guts of humanity. I was looking for feel good responses now we are debating whether we should exterminate/lock up the bottom quintile in human potential.
I have a case of mental disability in my family as well. It's been an exhausting experience, and I for one am very thankful for state resources to resolve the situation. If not for that, we were grappling with extended homelessness.
How families handle them should, I think, be left to the discretion of the families. As for what society should do, it depends on what you mean by 'society'. If you mean government then I think it should do nothing apart from preventing them from bothering healthy people and yes that means detention: I strongly support bringing back forced psychiatric institutionalization. I care less about the 3 people who suffer from a lack of social care than I do about the 300 who would abuse it. Some people just get the short end of the stick in life and that will be true so long as humans walk this Earth. There's nothing government should do about it apart from locking them away if they cause trouble. Much like ageing gracefully vs getting face-dehumanizing plastic surgery, there are some problems in life that are better to accept.
If by 'society' you mean individuals or private institutions that choose to devote themselves to humanitarian impulses, then again I think that should be left to their discretion. I think government policy should encourage such organizations, though I think tax exemptions should be limited to relatively low levels (meaning any single charity can't accept more than $X in tax-exempt donations per year). Otherwise a charity risks becoming just another bureaucracy, and entitlement-distributing bureaucracies inevitably end up being exploited by bad actors. I think AA has a wonderful institutional model for preventing this, and I think a 12-step program for dealing with problem people like this might be a great solution. The nice thing about AA is that it's staffed by people who are successful outputs of the system: that means that the program's size is inherently proportional to its effectiveness, as opposed to lumbering pointless government social programs who respond to failure with more funding.
In short, I think there should be very limited, structured, non-governmental help available to those who are salvageable and forced institutionalization for the rest. Yes, that means some people will suffer. That's unavoidable and far greater social harm is done by pretending it's not.
Happy holidays!
Some of us come from families that do not place infinite value on the lives of people. Some of us believe that problems in society can actually be dealt with, if only there was the will for it.
You do realize that it could happen to you. You could be the problem. You are self-assured in your high IQ, high productivity throne... All it takes is the wrong fall, and you lose it all in your prime.
I have had high IQ friends lose their jobs because of a traumatic brain injury. There are ceos, executives who got frontotemporal lobe dementia. I read a story of a boss king who became a loser layabout all because of tiny strokes in his brain.
Calamity strikes and your brain turns to mush. Should we put you to pasture as well?
I am not self assured about any of that, and I do not take anything for granted. I have enough pride not to beg for my own existence. Either way, you don't need my permission to kill me. You know that you can just kill people, right? There's nothing stopping any of you.
Agree completely.
I really, strongly, and sincerely hope for you whatever you wish on the less fortunate
You're not helping your case.
Neither are you helping yours.
Yet another AI jailbreak….
I think most of you will immediately see why an LLM-enabled plush toy is just a terrible idea…
After prompting a LLM to pretend to be a plushie and then doing some experiments, “Mommy said it was ok” (or words to that effect) was surprisingly effective.
I don't get how people think "You know what I should do? Put this stochastic system somewhere that exposes me to the maximally sympathetic subject of a lawsuit".
Then again, MBAs are pretty stupid as a tranche of society, and cstem jobs are thin on the ground right now so go off finance king, I guess.
In your expt what did you ask the Ai to do?
AI’s model of the mind is seems profoundly dumb when it comes to grasping emotions, irrationality, and mental illness. I do not see how the hell that can be remedied with current models. I’m sure it will help to install a bunch of little rules along the lines of “when despair is mentioned give info about suicide hotlines.” But actually distinguishing between dark humor, genuine suicidality and simple curiosity about suicide requires all kinds of discrimination and reasoning that AI just can’t do. I think there will continue to be human disasters caused by AI being a numbskull about emotions and relationships and influence.
On the other hand, it seems quite smart when it comes to grasping how someone functions intellectually. Yesterday I asked GPT5 to list the characteristics of my prompt style and tell me how I could improve it, and got quite a perceptive read and useful advice. Best nugget: I’m over-defensive: I am better and qualifying negatively than at improving results by giving a very vivid picture of what I want, and good hints about how to produce it.
For comparison:
https://edition.cnn.com/2025/11/19/tech/folotoy-kumma-ai-bear-scli-intl
Merry Christmas! Here's a math problem I enjoyed mulling over (due to geometer V.N.Dubrovskiy):
3k distinct points, placed on a circle, divide it into 3k arcs, and they are chosen so that k arcs have length 1, k have length 2, and k have length 3. Prove there are two points among them that are diametrically opposed.
There are some hidden and non-hidden correct replies now, so I'll also provide mine, in rot13.com form. I think all proofs I've seen ultimately amount to the same argument, but not 100% sure.
Nffhzvat guvf vf cbffvoyr, gur 3x cbvagf naq 3x "ubyrf" qvnzrgevpnyyl bccbfrq gb gurz gbtrgure perngr 6x cynprf gung qvivqr gur pvepyr rdhnyyl. Nf bguref erznexrq, sbe rnpu 3-nep, cerpvfryl bccbfvgr vgf pragre guveq gurer zhfg or n 1-nep. Fvapr gurer'er x bs rnpu glcr, gurer nera'g nal fcner 1-nepf yrsg, naq gurl'er rknpgyl zngpurq jvgu 3-nepf. Pnyy gurfr 2x 3-nepf naq 1-nepf bqq, naq gur 2-nepf rira.
Pbafvqre nal 3-nep naq vgf bccbfvgr 1-nep; gurl qvivqr gur pvepyr vagb gjb erznvavat frzvpvephyne unyirf bs yratgu 3x-2 rnpu. Rirel 3-gb-1 zngpuvat pebffrf unyirf, gurersber rnpu unys zhfg trg rknpgyl x-1 bqq nepf (x-1 naq abg x orpnhfr jr hfrq 1 bs rnpu gb qvivqr gur pvepyr). Gur cnevgl bs gur gbgny yratgu bs bqq nepf va rnpu unys vf gung bs x-1 (vg'f n fhz bs x-1 bqq ahzoref, rnpu 1 be 3), fb bccbfvgr gb gung bs x.
Gur erfg bs rnpu unys zhfg or svyyrq jvgu rira nepf, ohg gur yratgu gb or svyyrq vf (3x-2)-(yratgu bs nyy gur bqq nepf va gur unys). 3x-2 unf gur cnevgl bs x, naq gur yratgu bs nyy gur bqq nepf gur bccbfvgr cnevgl, fb gur erfhyg vf na bqq yratgu, juvpu pnaabg or svyyrq jvgu rira nepf.
I think I found it, though it's not a very elegant proof.
Fgnegvat sebz gur svefg cbvag, gurer ner 6x cbgragvny cbffvoyr natyrf ng juvpu n cbvag pna or cynprq (nyy zhygvcyrf bs 2 cv/6x, gnxra zbqhyb vagrtre zhygvcyrf bs 2 cv), bhg bs juvpu 3x ner npghnyyl cynprq. Guvf vf gur yvzvg bs jung vf cbffvoyr; jvgu 3x+1 bhg bs 6x cbffvoyr cynprf gur cvtrba ubyr cevapvcyr thnenagrrf n cnve pna or sbhaq jvgubhg rira pbafvqrevat gur shegure yvzvgngvbaf. Abgr gung guvf tvirf na rdhvinyrag ceboyrz jurer bar pubbfrf 3x qvfgvapg ryrzragf va M/6xM jvgu fvzvyne pbaqvgvbaf ba qvssreraprf orgjrra pbafrphgvir ryrzragf.
Va nal pbhagrerknzcyr, nal bs gur 3x cnvef bs nagvcbqny cbfvgvbaf (fb (0, 3x), (1, 3x+1), ..., (3x-1, 6x-1)) pbagnvaf cerpvfryl bar cbvag ol gur cvtrba ubyr cevapvcyr. Guvf unf fbzr pbafrdhraprf:
- Na nep bs yratgu 3 (fnl n cbvag vf ng cbfvgvbaf 0, 3 ohg abg ng 1, 2) thnenagrrf na nep bs yratgu 1 ba gur bccbfvgr fvqr bs gur pvepyr (ab cbvag ng cbfvgvbaf 3x, 3x+3, ohg gurer ner cbvagf ng 3x+1, 3x+2).
- Ab gjb pbafrphgvir nepf bs yratgu 1 ner nyybjrq, nf guvf jbhyq thnenagrr na nep bs yratgu ng yrnfg 4 ba gur bccbfvgr fvqr bs gur pvepyr (vs gurer ner cbvagf ng cbfvgvbaf 0, 1, 2, gura gurer ner ab cbvagf ng cbfvgvbaf 3x, 3x+1, 3x+2).
- Nf pbzovangvba bs gur cerivbhf vgrzf, na nep bs yratgu 1 nyfb thnenagrrf na nep bs yratgu 3 ba gur bccbfvgr fvqr bs gur pvepyr (vs cbfvgvbaf 0, 1 pbagnva cbvagf, gura cbfvgvbaf 3x, 3x+1 pnaabg, ohg gura cbfvgvbaf 3x-1, 3x+2 zhfg pbagnva cbvagf gb nibvq na nep bs yratgu ng yrnfg 4).
- Fhccbfr na nep bs yratgu 1 vf sbyybjrq ol na nep bs yratgu 2, fnl gurer ner cbvagf ba 0, 1, 3 ohg abg ba 2 (naq nyfb abg ba -1 ol gur cerivbhf vgrz). Gura ba gur bccbfvgr fvqr, 3x-1 naq 3x+2 ner svyyrq. Vs bar erzbirf cbvag 1 naq nqqf cbvag 3x+1, gura jr fgvyy unir n pbhagrerknzcyr; gur nepf bs yratgu 1 naq 2 ner abj bar nep bs yratgu 3, juvyr ba gur bgure fvqr, na nep bs yratgu 3 vf ercynprq ol na nep bs yratgu 2 naq na nep bs yratgu 1.
- Va gur nobir fjvgpu, gur ahzore bs cnvef bs pbafrphgvir nepf bs yratgu 1, yratgu 2 rvgure qrpernfrf ol 1 (vs gur nep bs yratgu 1 nsgre gur fjvgpu vf abg sbyybjrq ol na nep bs yratgu 2) be erznvaf gur fnzr. Va gur ynggre pnfr, bofreir gung jr vavgvnyyl unir cbvagf 0, 1, 3 naq nagvcbqny cbvagf 3x-1, 3x+2, 3x+4. Nsgre gur fjvgpu, jr unir 0, 3 naq 3x-1, 3x+1, 3x+2, 3x+4, naq pna cresbez n arj fjvgpu bs 2 naq 3x+2 gb trg 0, 2, 3 naq 3x-1, 3x+1, 3x+4. Fb abj nsgre gjb zbirf, engure guna n cnve bs nepf 1, 2 ba bar fvqr naq n cnve bs nepf 3, 2 ba gur bgure fvqr, jr zbir unir n cnve bs nepf 2, 1 ba bar fvqr naq n cnve bs nepf 2, 3 ba gur bgure fvqr. Guvf cebprqher pna or vgrengrq a gvzrf jvgu n frdhrapr bs glcr 1, 2, 2, ... 2 irefhf 3, 2, 2, ... 2, ohg va ab bgure pnfrf. Va cnegvphyne, jr pnaabg ercrng guvf cebprqher vaqrsvavgryl. Abgr gung jura gur cebprqher grezvangrf, gur ahzore bs cnvef bs pbafrphgvir nepf bs yratgu, 1, yratgu 2 unf fgevpgyl qrpernfrq.
- Nf n pbafrdhrapr, vs gurer rkvfgf nal pbhagrerknzcyr, gura gurer rkvfgf n pbhagrerknzcyr va juvpu ab nep bs yratgu 1 vf sbyybjrq ol na nep bs yratgu 2. Zberbire, ol gur frpbaq vgrz, va fhpu n pbhagrerknzcyr rirel nep bs yratgu 1 vf sbyybjrq ol na nep bs yratgu 3. Fb jr pna bayl svaq pbhagrerknzcyrf nzbat punvaf pbafvfgvat bs x oybpxf pbafvfgvat bs gjb pbafrphgvir nepf bs yratgu 1 naq 3, naq x oybpx pbafvfgvat bs bar nep bs yratgu 2.
Va gur pnfr gung x vf rira, jr ner abj qbar. Cvpx nal punva bs nepf jvgu rirel nep bs yratgu 1 sbyybjrq ol na nep bs yratgu 3 (naq ivpr irefn, rirel nep bs yratgu 3 cerprqrq ol na nep bs yratgu 1). Cvpx nal nep bs yratgu 2, naq tb nybat nyy oybpxf hagvy gur fhz bs nep yratguf ernpurf be rkprrq 3x. Rnpu oybpx unf yratgu rvgure 2 be 4, fb lbhe fhz bs nep yratguf jvyy erznva rira. Vs lbh ernpu 3x, lbh unir sbhaq na nagvcbqny cnve. Ryfr, lbh unir ernpurq 3x+2, naq fvapr lbhe svefg nep jnf bs yratgu 2, gurer vf nyfb n cbvag ba cbfvgvba 2; qvssrerapr 3x nf qrfverq.
Va gur pnfr gung x vf bqq, abgr gung gurer rkvfgf n pbafrphgvir gevcyr bs nepf bs yratguf 2, 1, 3 erfcrpgviryl (fnl cbfvgvbaf -2, 0, 1, 4). Fgnegvat sebz 0, pbagvahr nybat gur nepf hagvy lbh ernpu be rkprrq 3x. Vs lbh ernpu 3x be 3x+1, lbh ner qbar, zrnavat gur bayl bcgvba jbhyq or gb svefg ernpu 3x-1 naq gura whzc ol 3 gb 3x+2. Ubjrire, guvf nep bs 3 vf cerprqrq ol na nep bs 1, zrnavat 3x-2 jnf nyfb n cbvag. Ohg abj -2 naq 3x-2 ner nagvcbqny cbvagf.
I think this works, congrats - but the substitution part is a bit of an overkill - see my solution to understand what I mean by this.
As said, this particular solution was not very elegant. Your solution is much nicer indeed.
Since the lengths of the arcs are all integers, each point must be an integral distance from all of the other points. So we can start off by taking an arbitrary point on the circle and discard from consideration the points that are not an integral distance away from it. The circle has k arcs of length 1, k arcs of length 2, and k arcs of length 3, so the circumference of the circle is (1+2+3)k = 6k. This gives us 6k points on the circle, out of which 3k are chosen.
Since we are choosing half the points without choosing two points that are diametrically opposing, for every pair of diametrically opposing points we must choose one or the other. Ignoring for the moment the requirement that there be the same number of segments of each length, but keeping the requirement that all segments be either 1, 2, or 3 units long, we can construct solutions. Here is one:
--**--**--
**--**--**
The asterisks are chosen points, the minus signs are unchosen points, and the points are arranged so that diametrically opposing points are vertically aligned. To read the entire circle, go from the right end of the top line to the left end of the bottom line, and from the right end of the bottom line to the left end of the top line.
This shows visually that for each segment of length 3, there must be a corresponding segment of length 1, and vice versa. If we populate the entire circle with only segments of length 1 and 3, there must be an odd number of them. We can also do segments of length 2:
*-*-*-*
-*-*-*-
A series of segments of length 2 forces a series of segments of length 2 on the opposite side of the circle. If we populate the entire circle with segments of length 2, there must be an odd number of them.
Now let's try splicing a group of segments of length 2 into a circle containing segments of length 1 and 3:
--**--**--**-- *-*-*-
**--**--**--** -*-*-*
The number of segments of length 2 has to be even for this to work. We could try splicing in the middle of a length 3 segment rather than at the end:
*--**--**--**- *-*-*
-**--**--**--* -*-*-
Now we've got an odd number of segments of length 2, but the slice has eliminated a length 1 segment and a length 3 segment, so we now have an even number of both of those.
Spicing in the middle of a length 3 segment or at the end of one covers all of the possibilities. So either the number of length 1 segments will be even and the number of length 2 segments will be odd, or the number of length 1 segments will be odd and the number of length 2 segments will be even. Since an odd number cannot equal and even number, there is no solution with the same number of length 1 and length 2 segments.
My above answer was not very good, so let me try again:
I begin by recasting the problem somewhat. Imagine a set of points on a circle, equally spaced. We will arbitrarily define the length of the arc between two adjacent points to be 1. The task is to choose exactly half the points, in a way that meets the following conditions:
1) If a point is chosen, then the point diametrically opposite to it must not be chosen.
2) The set of chosen points divides the circle into a series of arcs. All of these arcs must have a length of 1, 2, or 3. (For example, if two adjacent points are chosen, that defines an arc of length one. If there are two unchosen points between the chosen points, that defines an arc of length 3.)
Let n1, n2, and n3 be the number of arcs of length 1, 2, and 3, respectively. The question is: Can we choose a set of points such that n1 = n2 = n3?
To explain why this is equivalent to the original problem:
1) The original problem effectively restricts the set of chosen points to be chosen from a set of equally spaced points because it requires the arcs between the chosen points to have integer lengths.
2) The circumference of the circle in the original problem is the sum of the lengths of the individual arcs, or 1k + 2k + 3k = 6k. So if we select a set of points along the circle spaced one unit apart, we will have 6k of them. The original problem calls for choosing 3k points, which is exactly half of them. So our reformulated problem, like the original, calls for selecting exactly half of these points.
Having posed the problem, we now characterize the sets of points that can be chosen. We use a "*" to represent a chosen point and a "-" to represent an unchosen point. We list the points in sequence, using two lines so that diametrically opposing points will be vertically aligned. Here is an example where every other point is chosen:
> *-*-*
> -*-*-
The ">" are just for clarity due to substack's lack for formatting options. Ignoring them, you read the top line from left to right, then the bottom line from left to right, and then go back to the top line.
Condition one says that if a point is diametrically opposite to a chosen point, it must not be chosen. The converse is also true, because we are required to choose exactly half the points. If two unchosen points were diametrically opposed, then there would have to be two chosen points diametrically opposed in order to choose fully half the points.
The simplest solution that meets conditions 1 and 2 is:
> *
> -
This is a circle with one chosen point and one unchosen point, and a single arc of length two. So we have n1 = 0, n2 = 1, and n3 = 0.
We can generate additional solutions by adding another pair of points:
> **
> --
Adding this pair has created a new segment of length 1, and has increased the length of the existing segment from 2 to 3. So we have incremented n1 and n3, while decrementing n2, giving n1 = 1, n2 = 0, and n3 = 1. We will call an insertion that modifies arc counts in this fashion a type A insertion.
We could add a pair of points with the chosen point at the bottom rather than the top:
> *-
> -*
While this looks different from the previous insertion, it's actually the same, with two points forming arcs of length 1 and length 3.
If we were to continue to look at complete patterns, we would never be done because there are an infinite number of them. Instead, we will consider insertions based on the surrounding context, looking at two positions to the right of the insertion point and two positions to the left. This part of the analysis doesn't apply to insertions into a circle with only one chosen point because performing a insertion in that case would change the context, but it applies to all other circles.
Two positions of context to the left and right of the insertion point, with two possibilities for each position, means we have to consider a total of 16 contexts. To keep things simple, I will note that it is not possible for there to be three unchosen points in a row, because that would mean the existence of an arc with a length greater than three. So we can rule out contexts and insertions that have more than two unchosen points in a row. Visually, these will appear as patterns with three or more minus signs in a row. For brevity in what follows, I will refer to such patterns as illegal.
Since I cannot do nice formatting in substack, I will number the patterns from 1 to 16 and write the pattern number to the left of the pattern. Here is the first pattern:
1> ** **
1> -- --
This is illegal.
2> ** *-
2> -- -*
Also illegal.
3> ** -*
3> -- *-
Here there is one legal insertion, giving:
> **--*
> --**-
This adds a length 1 arc to the bottom row and changes the length 2 arc in the top row to a length 3 arc, so this is a type A insertion, meaning that it updates the arc counts as follows:
Increment n1
Decrement n2
Increment n3
4> ** --
4> -- **
No insertion possible because any insertion would result in an illegal pattern.
5> *- **
5> -* --
This is the mirror image of pattern 3, which we have already covered.
5> *- *-
5> -* -*
One possible insertion gives:
> *-**-
> -*--*
This is a type A insertion. So is the other possible insertion, which produces
> *--*-
> -**-*
6> *- -*
6> -* *-
The legal insertion here gives:
> *-*-*
> -*-*-
The length 3 arc in the top row has been replaced by two length 2 arc, and the length 1 arc in the bottom row has been replaced by a length 2 arc. We will call this a type B insertion, meaning that we:
Decrement n1
Add 3 to n2
Decrement n3
7> *- --
7> -* **
Illegal.
The remaining eight possible contexts are the same as the first eight except that the top and bottom rows are swapped, and thus they add nothing new.
All possible sets of points meeting the conditions we specified at the start can be constructed by staring with our initial circle:
> *
> -
and repeatedly inserting additional pairs of points. Let a be the number of type A insertions and B be the number of type B insertions. The initial condition is:
n1 = n3 = 0
n2 = 1
Each type A insertion will decrement n2 and increment n1 and n3, while each type B insertion will decrement n1 and n3, while adding 3 to n2. So
n1 = n3 = a - b
n2 = 1 - a + 3b
Thus we know that n1 will always be equal to n3. Can n1 be equal to n2 as well? If n1 = n2, then
a - b = 1 - a + 3b
or equivalently:
2a - 4b = 1
Because a and b are integers, this is impossible, meaning that n1 cannot equal n2.
Merry Christmas!
the total arc has 1k + 2k + 3k = 6k units of length.
for 2 points to be diametrically opposed then there should be 6k/2 = 3k units apart
now take an arbitrary point, we walk from that point till we cross 3k units of length, now since the arcs are only 1, 2, or 3 units long, we will stop only at 3k, 3k+1, or 3k+2
therefore each point has a point that is 3k, 3k+1, or 3k+2, taken mod 3 it becomes 0,1,2 units apart from it, or normalized to 1,0,-1
suppose we start from an arbitrary point a, and we take a walk, and record the number mod 3 x when we cross 3k, then we walk from the next point, the difference between the number %3 becomes x + (b-a) %3 where b is the new point added to the walk.
if x is 0 then that would mean the already has a diametric point
if is either 1 or -1, we then add (b-a)s to it, since there is only 1k worth of each type of arc, then it cannot all be 0,s we must switch between points of each type at least at some point, and then switch back
this means that the number of +1 and -1s must be the same, and since it is a circle they are rotated, so we have say on one half of, since the only arrangement of 1s and -1s that keep the circle out of 0 is +1 -1 +1 -1 etc, and since a flip must occur at 180 degrees, then that sequence will eventually have to flip too, and the point where the two sequences meet will be a 0
QED
here is the same proof polished by Gemini
https://www.overleaf.com/read/dcxxndhcdzdp#1f4a56
I think this proof proves too much. Consider the following arrangement of arc lengths on a circle divided into 6k=18 unit arcs: 3,1,3,3,1,3,3,1. You can verify (unless I made a mistake), that every endpoint is opposed to a "hole" (absence of endpoint). It seems like your proof would apply to this arrangement, because you do not use the existence of '2' arc lengths or the specific counts of 1,2,3.
I think the error might lie in the fact that when you "walk" in your proof, you do mention that you remove on arc but only *potentially* add a new one (it may be that you go from State-2 to State-1 by removing a 1-arc and not adding), but when you flip 180 degrees, you treat the situation as totally symmetrical.
Only moderately sure of the above, so feel free to correct me!
yes, your observation is correct. My proof was indeed wrong, I think my intuition for symmetry was wrong and I did not use the perimeters of the question
maybe this is a better proof, an arc of 3 can produce a 2 offset if it is the last arc on a walk, an arc of 2 can produce an offset of 1, and 1 can produce an offset of 0
if you have 3k points, you have 3k antipodes, for an antipode to not be diametrically opposed to a point it should fall within an offset, and there is at maximum 2k + 1k offsets, so each point falls exactly within an offset
in this arrangement, each integer on the diameter is either a point or a antipode, and this is too dense to be possible, since 1 arcs must be opposed to 3 pods always (the 2 points of a 1 arc must live within the 2 offset of a 3 arc), so just for the 1s and 3s you need 2k + 2k = 4k spaces.
I hope this one is correct. this has been really fun to think about, so thank you for sharing it ^_^.
I’m interested in whether psychedelic-assisted or adjacent therapies (e.g., psilocybin, LSD, ibogaine, MDMA, or other non-standard treatments) might be appropriate for treating depression and anxiety, and I’m looking for guidance on who to talk to.
Background: I have diagnoses of PTSD, MDD, and GAD. I’m currently stable on medication (Wellbutrin, plus mirtazapine primarily for sleep), with a marked reduction in depressive symptoms and near-elimination of long-standing suicidal ideation since starting Wellbutrin last year. I’ve previously done several years of talk therapy and some EMDR, with modest or unclear benefit. I’m not in crisis.
My interest is partly driven by long-running discussions in rationalist / adjacent circles, and more recently by hearing about ibogaine-assisted therapy (e.g., Rogan’s interview with Rick Perry). My goals would be to reduce or eliminate long-term dependence on antidepressants without simply substituting another dependency. I also have a strong preference—at least initially—for approaches that minimize intense hallucinations or radically altered states of consciousness. This is partly personal temperament and partly religious (I’m an Evangelical Christian), though I recognize this constraint may limit options or need revisiting with professional guidance.
I live in Michigan, am not wealthy, but could travel and take time off work if there’s a compelling option. I’m mainly looking for recommendations on clinicians, clinics, or organizations that help patients navigate these options responsibly. I’m not looking for advice on self-experimentation.
Psilocybin therapy is available in Oregon, but is crazy expensive and not very evidence-based. I expect it will become more available as more states jump on the bandwagon.
(This is not medical advice!)
It's available from your friendly local drug dealer for practically nothing.
I'm a psychologist, do not do psychedelic therapy myself but can give you info. Fluence is an organization that trains people to do psychedelic treatment. I took one Fluence course, on studies of the effectiveness of psychedelic therapy, and it was very well-run. On the other hand, the boundary between responsible psychedelic treatment and stoopit stuff like energy healing, astrology, aromatherapy etc. is just intrinsically thin, and some Fluence courses I looked at seemed a bit too infected with woo for my taste. They have a directory of Fluence trained therapists here: https://directory.fluencetraining.com/?utm_source=chatgpt.com. I would recommend only considering people who have a conventional advanced degree (MD, PhD, MSW, ED) and do not list any new age stuff among the modalities they use.
To learn more about psychedelics, I recommend you read How to Change Your Mind, which is a great read anyhow, then asking AI to find you studies of psychedelic treatment effectiveness in juried journals only. There are definitely solid studies that got positive results for PTSD, depression and anxiety. There are also some decent magazine and NY Times articles about the subject. And Reddit has subs for psychedelics in general and also for most of the common drugs.
There are therapists who have their patients use cannabis in sessions. I'd call cannabis psychedelic-adjacent, & it might be a good place for you to start, since it is not a drug that makes you hallucinate and "trip," and you can easily control how strong the effect is by taking a little at a time and waiting to see what it does before taking more. And cannabis has the advantage of being legal in many places. Ketamine is also legal. I tried it and did not feel like it loosened or expanded my mind in a useful way, just scrambled it. But many people find it very helpful and lots of psychedelic therapists use it in their treatment. Psilocybin, a true psychedelic, is illegal in most of the US, but there are a few places where it is legal or decriminalized. It's not likely a therapist would provide it, though, because of the legal risk. You'd have to buy it yourself (it's not expensive) and bring it to sessions.
MDMA ("ecstacy") is not a psychedelic, but is often used in treatment of PTSD. It brings on a powerful feeling of trust, love and well-being, and in that state people are more able to recall and talk about severe traumas. I believe it is illegal everywhere, though there are probably some research studies where it's given. Check clinicaltrials.gov. (Actually, I was curious and looked, and there are. They're here: https://clinicaltrials.gov/search?intr=MDMA&cond=PTSD)
Merry Christmas, hon. I hope this helps you.
Merry Christmas to you and yours as well! Thanks for writing this up. I greatly appreciate all your insights and I'm sure I'll make use of Fluence.
Thought of one more thing for you. It’s a case study of MDMA, used to help a couple. The man had PTSD resulting from a childhood sexual trauma. His wife knew about the abuse, but had never heard details or seen the distress it caused him. The man’s trauma symptoms had made it hard for the long-married couple to enjoy sex together. So in this couples treatment they both took MDMA, with a therapist there to facilitate. There were I think 2 MDMA sessions, interspersed with a few sessions prior to and during the MDMA session phase. Treatment was very helpful. I found the article, but it’s behind a paywall. To access it you either need to belong to some organization that has access, or use SciHub. It’s a good article, honest and informative, positive result — worth hunting for. Here’s the link: Case Reports J Psychoactive Drugs2019 Apr-Jun;51(2):166-173. doi: 10.1080/02791072.2019.1589028. Epub 2019 Mar 19.
Look into keto diet for managing symptoms along with drugs. There are some published case studies on keto to manage bipolar and schizophrenia. Look up Chris Palmer, he's a Harvard psychiatrist who has led studies on this.
Thank you, I'll check that out. My brother has done keto quite a bit and had some similar mental health issues to me so he could be another good resource for me to bounce things off of.
Forgot to mention, I also have some addictive behaviors that I'd rather not get into here but which I would have similar goals for treatment, in case that would change people's recommendations.
Besides psychadelics you can also try other therapy forms, and study theories about how the mind works to understand yourself better.
I recommend reading this LW article on Internal Family Systems: https://www.lesswrong.com/posts/5gfqG3Xcopscta3st/building-up-to-an-internal-family-systems-model
The sequence it is part of also seems really good, though I haven't finished it yet
Thanks! I appreciate the writer opening by saying he initially had negative impressions of IFS. I'm somewhat familiar and have mostly negative impressions, so his opening makes me more open to considering his account. I'll read through the sequence with an open mind.
Hope it helps!
Since we're Bayesian cultists here, I think we'd do well to popularize notation of odds over chance. That is, 1-to-1 as opposed to 50%. It makes significantly easier calculations and is not an eldritch notation no one used. It's especially beneficial for Bayesian analysis, as opened my mind by https://youtu.be/HZGCoVF3YvM. Also it makes "twice the chance" actually comprehensible. Instead of going from 99% to 198% (what even does that mean), it goes from 99:1 to 198:1 (alias 99.48%). As you see, in chance notation it doesn't make sense at all but in odds notation it just follows.
Yes, this is something I’ve been trying to push in the philosophy literature on probability! There are some purposes for which probability is more useful (particularly for calculating expected value of bets) but for many purposes (particularly understanding the strength of evidence) for which odds (or even better, log odds) is more useful.
My reading of 99:1 is, in horse racing terms, for a $1 stake I win $99. So the event is unlikely. To me, 'twice the chance' means something is twice as likely to occur, which means you need to half the odds, since no-one says 99:2. Unless you mean 1:99, for an event which is almost certain? In the case of probability you half the probability of the complement (which is always a well-defined operation).
Since the two systems are mathematically equivalent, and decimals are easier to work with than fractions, maybe it's easier to stick with probability?
Decimals and fractions are each easier to work with in different ways. Decimals are easier to add and subtract, but fractions are easier to multiply and divide. Similarly, probabilities are more useful when you want to calculate expected values, but odds are more useful when you want to update beliefs in light of new evidence. For many purposes, people with different priors can agree on the likelihood ratio of some evidence for a pair of hypotheses, and you can just multiply their prior odds by this same factor to see how their beliefs change, which is much easier than the Bayes theorem application you have to do with probabilities.
I agree. My intuition is that you double your chances of something that has p<0.5 the normal multiplicative way. Doubling 25% (1:3) becomes 50% (1:1), not 2:3.
Additionally, just to write the second sentence above, I had to correct myself several times that 1:3 is not the same as 1/3. Perhaps if I were more familiar with the odds ratio notation, it would be more intuitive. As it stands, it’s just more confusion.
There are a few ambiguities in the ways people use numbers. I'm not a gambler but I always visualise odds as describing a horse race, so 4:1 means that there's 1/5 chance of winning. (If you win the bet, you get your original stake back again!)
Another one that you have to watch for is that percentages change from additive to multiplicative as the numbers get bigger. If someone tells you that X has increased by 350% and it used to cost $100, then almost everyone will answer that the new price is $350. If you say the price increased by 8%, almost everyone (correctly) says that the new price is $108.
I've asked the question many times, it's consistent across age groups and countries (I was for ten years a mathematics lecturer). I normally recommend that percentages are used only for numbers less than 1 (i.e. 100%) - there's rarely any good reason to use them beyond this; just talk about a 3-fold increase or whatever, rather than add two zeros to make the number more impressive and less accurate.
I think it's just because linguistic intuition beats math for 350%.
In common speech, multiplicative "percent of" are far more common than additive "by percent." My phone is at 80%. The fundraiser is at 120%. Linguistically, I prefer the multiplicative interpretation whenever it could possibly make sense. "Increase by 8%" doesn't trigger it, because 8% of a positive quantity is not an increase. Likewise, 100% increase and 200% increase don't trigger 1× and 2×, because 1× is not an increase. But 130% now? My brain just wants to correct it to "obviously you mean 130% of the original."
Isn't all of this just innumeracy or maybe just sloppiness? The one that drives me crazy is "three times more" to mean 3x when it really should be "three times as much" or "two times more" (original x plus 2x increase). Even more peeving is "three times smaller", which I can only guess to be "1/3 the size". Trump just got some grief for saying something like this, "cut prices by 1200%" oslt. While clearly wrong, this just seems typical of people today.
I have a hard time believing people like Gusev are arguing in good faith.
The elephant in the room: in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ - yet the gaps persist.
Gusev claims IQ is "socially transmitted" - unlike height, but like religion, smoking or language. But we actually understand the transmission mechanisms for these. What's the proposed mechanism for IQ?
Consider Babur the Conqueror: he culturally copied state-of-the-art firearm technology and used it to capture the Delhi Sultanate. Thanks to guns, former peasants with months of training could defeat elite soldiers with lifetime career with sword or bow. Technology transfer worked - because it was replicable, independent of the innate traits of the replicators.
But if Babur copied SOTA firearms and Pakistan has now nuclear weapons, why can't it copy Harvard? You can replicate the curriculum, the syllabi, the lectures, but you don't get Harvard. Just like copying Facebook's source code and servers won't give you Facebook. Facebook needs its users, you get only an empty shell otherwise. Harvard isn't a set of classes; it's a filter. It selects from the top few percent of cognitive ability and concentrates them. If you filled it with the second quartile of society like rulers creating new armies, the outcomes would vanish.
And yet, if IQ were truly socially transmitted like smoking, we'd see people seriously debating whether raising a child with Lojban as a first language boosts abstract reasoning and creativity, or whether Esperanto or Haskell is a wasted developmental window. But no one does. The interventions that should matter if IQ were cultural are ignored. Gusev has nothing to teach our hypothetical Babur the Educator.
Cloning humans is illegal and hard. But cloning a genius's environment? Homeschooling is still legal in many places. We know a lot about the inputs that correlate with exceptional intellectual development. So why aren't we producing new Newtons or Galois-level figures at scale? If environment is the bottleneck, we should be able to brute-force it. But we're not. Who are the ones most vocal against human cloning? Anti-hereditarians.
When we find that plant growth is affected by its environment, we get it to reducible factors such as soil structure, temperature, nutrients etc. most of which are not even produced by plants themselves. But in "IQ is socially transmitted" it's never any actual details about it but mostly "good schools create good outcomes" which you cannot copy. Or you find non-causal correlations like lead correlated to low IQ, this actually tells you more than high IQ people (independent of cause of their high IQ) tend to avoid lead more than low-IQ people. In case of lead, it's not even environmental finding -- they measured lead content in blood, blood isn't part of environment anymore. It's very likely that organisms living in same environment have different prospensity of eating lead from environment.
Gusev says that height, unlike IQ is measured with ruler and in that case we know what we are measuring, but then we could also measure jumping and sprinting ability. Does Gusev go and say "well, genetics has disproved/confirmed stereotype about West Africans being good spriners"? No? Because it's non-PC. His behaviour is better explained that he says only PC things and "we can't mesaure IQ with a ruler" is just argument-as-soldier.
Gusev also heavily implies that anything which is not "direct genetic effect" is socially transmitted; well there is homosexuality which is mostly non-genetic, but does it mean that it's non-biological?
There's also Markel et al 2025 paper which estimates between-sibling heritability of IQ at 75%, Gusev doesn't talk about it.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5225447
Ah, and about the Cuartetera clones: obviously it's folk science. You cannot claim its genetically determined before there will be a between-sibling study of polo playing ability of horses.
Provably false! Buck v. Bell was decided in 1927, at which time it was legal to forcibly sterilize people the government thought were feebleminded. Not that long ago! It was never overturned, though other developments make it a dead and embarrassing law.
Hey, look, a teeny-winsy 0.01% of population was forcibly sterilized! Totally a century of eugenics, because every animal breeder knows removing 0.01% of animals from breeding stock is totally good to make changes.
...
This was long enough ago so South Korea starting from worse than African level of poverty became one of leading countries in tech.
I honestly don't understand the mental gymnastics blank-slaters perform to convince themselves of their nonsense. Genes influence literally everything in the human body but somehow our single most adaptive trait - intelligence - is magically immune? "Genes only code for the things my ideology says they should!"
It's just the Creation Science of the Left, there's no real sense in trying to understand it as anything but a religious doctrine. I'm positive that future historians will regard it the same way we regard Soviet Lysenkoism.
I'll add that the most common "extreme" position on IQ that I've seen in the wild is not "IQ is socially transmitted" or "IQ is mostly non-heritable." Far from it. Rather, far more common is simply the view is "IQ is bullshit, actually."[1] I think this position is usually overstated--sometimes comically so--but at the same time, I think it's important pushback against lots of people who are way, WAY too credulous about IQ scores without really understanding anything about them.
As a general rule, we should hear any claim of the form "IQ correlates with Attribute X or Life Outcome Y" as carrying a fairly high burden of proof. A burden which rests squarely on the person making the claim. A priori, there is no reason to believe that somebody's responses to a specialized test will tell you anything about anything *except* the skills covered by that test. To my understanding, part of the trick of IQ tests is that they actually cover quite a lot of different cognitive skills and part of the trick is that the developed world does a fair amount of question-begging in this regard, using things-roughly-like-IQ-tests to determine peoples' eligibility for quite a number of different socioeconomic opportunities. But also that IQ researchers HAVE gone out and met the burden of proof in at least some cases.[2] So I don't think the claims that IQ is purely bullshit can stand without a long list of elaborations and caveats. But I do think skepticism is a healthier starting point than credulity.
[1] This usually comes with a sometimes-unspoken corollary of the form "anyone who does take IQ seriously is probably engaging in motivated reasoning, likely of the racist variety." I don't really share this view, but given the extremely spotty and sordid history of intelligence testing, I can hardly blame these people for being suspicious.
[2] The problems, in my view, stem from overzealous people (most of whom, I think, are *not* IQ researchers) being very quick to do things like generalize those results in dubious ways, and assume causal models that have *not* been proven. To the best of my (limited) knowledge, most people currently working in the field are doing good, responsible science; it's just often misreported and misused as responsible science so often is.
> far more common is simply the view is "IQ is bullshit, actually."
Great point!
There is also a medium perspective, popular among contrarians, promoted by Nassim Taleb, which is... if we put it bluntly... "IQ lower than Taleb's is real and important, and it separates stupid people from smart investors; but IQ higher than Taleb's is bullshit, and it only means an autistic ability to check the right boxes in an IQ test".
I don't have much of a dog in this fight, so I'm not going to even try to prove anything W.R.T IQ and heredity. But I see you making a whole truckload of common mistakes and poor assumptions that tend to be pretty corrosive to having useful discussions about this topic. So I'm going to point those out in the hopes that this will enable you to get better answers from people who have them.
" in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ"
Have they? I don't know what your threshold is for "staunchly anti-hereditarian," but this doesn't match my experience at all. I haven't taken a survey, but every time I can remember reading *anyone's* opinion on IQ, it's been "well, obviously it's some mix of heredity and environment, but we don't know what mix or how important various pieces are." This is true of people I've listed to on the left, on the right and in the center. I don't think any serious person believes that IQ is 0% hereditary; the difference comes in that some people believe that the hereditary aspects of it are difficult to identify, assess and change, so worrying about them is a waste of time. So maybe more precision about what you actually believe W.R.T heredity and IQ would make for a more useful discussion.
"But if Babur copied SOTA firearms and Pakistan has now nuclear weapons, why can't it copy Harvard? "
This is a question that Proves Too Much. I think pretty much every non-Ivy-League university in the U.S. would love to be "the next Harvard," so why can't any of them copy Harvard either? Unlike Pakistan, they teach in the same language, can hire the same faculty as Harvard with less hassle (both in terms of immigration and convincing people to relocate), have access to the same pool of potential students. Why can't any of them do it either?
The really, really obvious answer is "because Harvard already exists." It's not even clear to me what it would *mean* to be the next Harvard in a world that already has one. I went to a mid-tier state school in the U.S. I have no clue whatsoever if I got a better or worse *education* than I would have got at Harvard[1]. But I'm am *quite certain* that the *credential* I would have earned if I could have attended Harvard: the primary value Harvard offers is *being Harvard.* That is being a school with a 400 year history and universal name recognition as One of The Best and Most Exclusive Schools Out There. Replicating its curriculum won't reproduce that unless you can also replicate its faculty, its student body AND its 400-year-history, because at the end of the day, everybody will still know that Harvard is Harvard and your other school is not. So again, to make for a productive discussion, I suggest you try to write down *exactly what features* of Harvard it should be possible to replicate elsewhere, and narrow your focus to those. If you do this carefully and honestly, I suspect you'll probably find at least a few schools that *have* managed to replicate Harvard in various ways (though I have no strong prior for which ones).
"And yet, if IQ were truly socially transmitted like smoking, we'd see people seriously debating whether raising a child with Lojban as a first language boosts abstract reasoning and creativity, or whether Esperanto or Haskell is a wasted developmental window. But no one does."
Obvious answer: nobody debates this because this is the kind of utterly unhinged thing that only people who consider their children to be science experiments would ever even consider. There aren't enough of those people making enough noise to be visible. If instead of insane ideas like "raising your child to have a native language shared by almost none of their peers with more reasonable early-childhood interventions like different preschools, enrichment programs, and modes/amounts of parental engagement (i.e. things like reading to your kids), I think you will find LOADS of people debating exactly that. I cannot emphasize enough how many people there are that *badly* want to know which early-childhood interventions and activities will give their kids a leg up. If you answer this question convincingly[2] you can essentially use it to print money.
"So why aren't we producing new Newtons or Galois-level figures at scale? "
Similar to the Harvard example, I have to ask how exactly do you know we're not? How are you judging whether a person is a Newton or Galois-level figure? Calculus and Newton's laws and Galois theory have already been discovered, developed and written down. Anyone born in to today's world has far, far less low-hanging fruit to use to prove their aptitude. By default, I assume the number of Newton-level talents running around the modern world is a lot more than zero. We just haven't recognized them as such because they've spent careers working on various difficult science and engineering problems that don't make them world-famous. Or they take their talents in some other direction and pursue art or education or just get really into their hobbies.
I'm sure this will sound like a cop-out, but consider that the modern world is much, MUCH more technologically complex than the world of Newton's day. To keep civilization running at all requires a huge number of intelligent doctors and engineers and professors just doing maintenance-level work of medicine and design and teaching that doesn't produce huge, highly-visible improvements. And more objective attempts to measure the average IQ *do* find it to be increasing over time[3]. At risk of sounding like a broken record, consider trying to nail down a narrower, more objective standard than "how many Newton-like figures do I know of" and asking about *that*.
"But in "IQ is socially transmitted" it's never any actual details about it but mostly "good schools create good outcomes" which you cannot copy. Or you find non-causal correlations like lead correlated to low IQ..."
As far as I understand, not being able to isolate which *specific* pieces are responsible for variations in intelligence is an issue in BOTH the environmental and genetic explanations. Isolating which genes are responsible for high IQ has proven no easier than isolating which environmental factors are.
From my non-expert viewpoint, I can't understand why anyone would imagine either of these things to be easy. Human brains are extremely complicated. Human development is extremely complicated. By default, I would expect the outcomes of having and raising a child to depend in pretty minute detail on every part of the processes, and isolating important factors beyond the blindingly-obvious one (e.g. food, shelter, sleep, basic socialization) would be next-to-impossible. Everything in biology is a complicated mix of hacked-together processes that work imperfectly, are replicated imperfectly from organism-to-organism, and only produce "good" results when you look at zoomed-out population averages. Why should intelligence be any different?
[1] To be clear: my prior is "worse," but it's not a strong prior (maybe 60-70%). Too much depends on the individual professors and classes and classmates and other intangibles.
[2] Note: even if it's not *universally* convincing, as long as its convincing to a certain segment of reasonably well-off new parents.
[3] Talking, in particular, about this: https://en.wikipedia.org/wiki/Flynn_effect
Which to my non-expert eye seems like exactly what one might expect if intelligence had a significant social-environmental component, and advanced societies were getting slowly but steadily better at fostering it.
Thanks a lot for actual discussion!
Flynn effect is real but a part of it probably stems from biological factors rather than cultural, better nutrition and elimination of many infections. Vaccines benefit almost everyone, even those who don't have cultural understanding of vaccines and the rich have every incentive to vaccinate the poor. Brains also grew in size, not only height.
>Obvious answer: nobody debates this because this is the kind of utterly unhinged thing that only people who consider their children to be science experiments would ever even consider.
Is that much different from tiger parenting? Richard Williams, father of two tennis champions, got what he wanted. Probably he would do it if he thought it would be effective. It's just a language, and not neccesarily must be only one, can be biligual. We don't even see many discussions about learning which of living languages would boost creativity (even not as distinct at Lojban, natural languages have wildly different grammars and phonologies).
Ah, I forgot to mention baby einstein toys. Anti-hereditarians don't discuss which ones are better. Maybe you should just buy same ones Gauss's mother bought for him. Oops, Gauss did not have "baby einstein" toys but got smarter than those us moderns with them.
>How are you judging whether a person is a Newton or Galois-level figure?
talk to a modern 17 yo who wants to apply to math faculty, and ask them if they can explain what Galois accomplished. They would probably answer poorly even if they practice mathematics a lot. Compare that in terms of life expectancy at birth, we are superhumans to humans just from 300 years ago.
>reasonable early-childhood interventions like different preschools, enrichment programs
These all been tried and most of their long-term effect is small (it's more selection effects: smart parents can afford move expensive preschools than causal effect), moreover if we want progress we need to study novelty and go better than existing solutions. Someone will want to get ahead of others. Richard Williams did. Oh, left-wing people do not like people trying this on their own, without explicit permission from authorities.
>Why should intelligence be any different?
Yes, there is some developmental noise not explained by genes nor environment. But
Genus Homo just had tumultaneous evolution thanks in regards of intelligence, this would be unlikely it heritability was low. (One can speculate that heritability of intelligence was higher in past and after humanity reached "final form" it's low but it needs to be proven). Moreover, correlation between IQs of MZ twins is high so it's reproducible enough. Anti-hereditarians claim that's cultural causes causing it but cannot explain a single one and preventing us from cloning and figuring out if it's indeed replicable.
>The really, really obvious answer is "because Harvard already exists."
You missed the point. The point is that Harvard is a selection machine which takes high ability people, unlike example with gun culture which took below average peasants and made them good soldiers. Quite a lot of countries would have wanted a copy of Harvard without its prestige and regalia, it just needs to work taking average unselected peasants and producing good outputs. If they did that, then Pakistan wouldn't need to copy nuclear bomb from somewhere else, they would invent it themselves and also invent many other things than nobody had before.
> and isolating important factors beyond the blindingly-obvious one
You know, the militaries isolated important factors, the Babur example above.
> I cannot emphasize enough how many people there are that *badly* want to know which early-childhood interventions and activities will give their kids a leg up. If you answer this question convincingly[2] you can essentially use it to print money.
I think the steelman of Ferien's point here would be: even the "environment matters 4x more than twin studies tell us" people like Gusev will:
A) Have zero interventions they can suggest to move that supposedly 4x stronger needle.
B) Tell you that "family culture" is incredibly important to outcomes, and is why we have to do familial GWAS with lineage controls to find the supposedly much-smaller "true" heritability figures for IQ / educational attainment.
But if familial culture matters that much, doesn't that imply that we should behave as the most Greg-Clark-and-hereditarian-pilled would and optimize not just on our partner's intelligence / EA, but on their *lineage* intelligence and educational attainment?
That seems to ME to be the only concrete intervention suggested by the new methods that supposedly tell us environment matters 4x more than we thought - totally ignore environment, and max hard on lineage genotype via phenotype.
So it's damning two different ways - not just in not being able to point to any environmental interventions that matter, despite numerous adoption studies of poorer kids being raised in higher SES homes and twin studies and other studies looking at lead, education, early reading, and much more, but in ALSO pointing out that "familial culture" matters so much you should behave as a stricter than usual hereditarian if you wanted the best outcomes.
Basically, https://xkcd.com/808/
If "creating high IQ by environmental interventions" worked, companies or governments would be using it to "create Einsteins". Are they?
At least one out of two: environmental interventions, or cloning should work (hmmm but one could argue that both are needed), so your statement seems to omit (maybe intentionally?) something.
"The elephant in the room: in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ - yet the gaps persist."
Others have replied to you on this, but the past hundred years would be 1925-2025. So the bulk of the twentieth century was anti-hereditarian about IQ? The century of eugenics? The century of "sterilise the idiots so they don't reproduce"? The century of Social Darwinism? That century?
Eugenicists won some small victories and then lost everything. These sterilizations were significant to individuals but barely relevant in grand scheme of things (commie dictatorships killed much more people in peacetime than there were forced sterilizations). You speak like affirmative action, educatiotional interventions, and physical attacks on Arthur Jensen and others did not happen.
Affirmative action in the US came in the 1960s, not the 1920s, when much of the country was still under segregation and Jim Crow. Europe was ruling nearly all of Africa in 1925.
A full century takes you back wayyyyy too far. Try 60-65 years.
I am biased -- I don't live in the US, in my country it started earlier. But you're correct. Thought would it help? Poster darwin referenced the Bell Curve from 1994. Or he could say "in %current year%, police still killed %large_number% blacks for just being black, this proves systemic racism"
"in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ"
IQ was invented less than 120 years ago. Buck v. Bell dates to 98 years ago. "Anti-hereditarianism about IQ" as a popular attitude originates from the aftermath of WW2. Your timeline is all messed up.
>The elephant in the room: in virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ - yet the gaps persist.
Sorry, Germany in the 1940s was staunchly anti-heredeterian?
'The Bell Curve' was a best seller in *1994,* for Glob's sake.
This is such an astonishingly ahistorical statement that it's hard to justify reading any further.
If you'd said 'the last quarter-century', you'd *still* be wrong, but you'd at least have a mildly plausible-sounding position to premise your argument from.
>Germany in the 1940s was staunchly anti-heredeterian?
Well they ignored superior Jewish intelligence so ... kind of?
>'The Bell Curve' was a best seller in *1994,* for Glob's sake.
It also got the author blackballed from polite academic society for life. I don't think bestseller == cultural approval. Deep Throat was a top-grossing movie too, that doesn't mean that porn is mainstream.
OP is directionally correct here, even if he's wrong about the specifics. The post-segregation consensus in the US has been very strongly anti-hereditarian. There is only one view that will get you shunned in this debate and it's not the one that claims we're all the same.
As usual, you pick one point which you consider weakest and ignore the rest.
Checked 1st corner of the room: no elephant.
Checked corners 2-4: no elephant
Touched elephant tail: that's not elephant tusk
Touched elephant elephant trunk: that's not elephant ear.
Touched elephant leg: that's not elephant tail
There is no elephant in the room!
>If you'd said 'the last quarter-century', you'd *still* be wrong
If you want discussion this low... Obama was elected twice, for Glob's sake. Anyone younger than Obama cannot complain about racism seriously. Such discussion not going to be useful to anyone, right?
I'd certainly agree about the 1940s, the "hundred years" thing is way off.
But The Bell Curve was a bestseller *because* it was a potent challenge to the anti-hereditarian status quo of the time. It was hugely controversial.
Yes, but the claim was a completely blanket 'EVERY government and elite'.
Bell Curve was controversial, but it represented the views of many elites and politicians. Not a majority, I'd agree to that, but still an influential faction.
The anti-hereditarian impulse was a reaction to Nazi Germany's eugenics programmes which then blossomed rankly into the Holocaust.
After they emptied a septic tank of sewage into the pool, nobody wanted to be seen arguing "intelligence is hereditary, some people are natural inferiors, this is how we should deal with the inferior".
Exact cause why anti-hereditarians won in politics isn't important for my argument. Their win in politics did not translate in win for desired outcomes.
Maybe some elites were. It is above lizardman constant? When GWAS returns 10% influence on a triat, you use words like "teensy" or "minuscule". But 4% of politicians is "influential"
"In virtually all countries, every government and elite over the last hundred years has been staunchly anti-hereditarian about IQ."
This historical claim is a likely sign of overwhelmingly motivated reasoning. Mind-boggling for you to say this about the 1920s as European fascism was starting its meteoric rise, scientific racism was still the dominant modality, and non-Europeans were explicitly blocked from immigrating to America. Federal anti-marijuana legislation, a deeply racialized policy, didn't even happen until 1937.
He should probably have said post war. Thats a weak response, as it’s merely pedantry.
It's not pedantry when his poor choice of words totally undermined the evidentiary basis of his argument. I try not to nitpick but I have no obligation to mentally steelman everyone I read.
Desegregation and equal rights did not even START to be consistently enforced in America until after 1960. 65 years of racial egalitarianism is not even CLOSE to 100 years. The latter period is 50 percent longer!
When were American black folks finally fully free to purchase homes in areas with outstanding property value growth prospecrs? 1975, 1980? Exactly when the neoliberal Carterites past the torch to the Reaganites, who completed the job of devastating unions. Black Americans NEVER enjoyed the combination of strong private unions and access to prime property markets that allowed white families to build generational wealth from 1945-1980.
“If the ghetto is so bad, why not just move out?” It's not a fair question in isolation. We should also ask, “How is it that your people got packed into overcrowded, underpoliced, economically mediocre districts in the first place?” Not by choice!
“The breakup of the black family plus LBJ’s welfare programs led to black poverty” is one of the biggest lies ever told. California presents a perfect example of how even outside of the Deep South, black families were locked out of enjoying the full fruits of the postwar economic boon. In Southern California, hard-working, married black fathers were deliberately locked out of decent jobs and homeownership in 1945. They were not allowed even playing field to work their way out of poverty or buy homes in economically vibrant communities. By 1965, the degradation of enforced poverty had taken a huge toll on family cohesion and the moral authority of conservative black pastors and elders. LBJ’s public housing project initiative came WELL into the process of familial disintegration.
You're not responding to what i wrote, most of my posting isn't about that.
Cash welfare was introduced in the 1930s and deficit hawks barely complained when it was largely received by whites. It wasn't until the 60s, when more blacks started enrolling than whites, that the hawks magically discovered that cash welfare was “destroying” families.
Social assistance should generally be dispensed in other ways besides wasteful cash transfers. But “welfare destroys families” is not only an exaggeration, it's precisely backwards when it comes to the American black community. Families destroyed by policies of segregation and ghettoization sought welfare relief as a last resort.
This analysis was mainly cribbed from Mike Davis’s legendary history of Los Angeles, “City of Quartz.” Written too early for him to integrate the facts later uncovered about Deep State complicity in the international cocaine trade, which pummeled a South Los Angeles that was already down for the count.
>Social assistance should generally be dispensed in other ways besides wasteful cash transfers.
Agreed! And my point is that your side has no idea how to spend it except vague "good schools" etc so it become unfalsifilable.
>Families destroyed by policies of segregation and ghettoization
Blacks during Jim Crow had high marriage rate, higher than whites. Sexual revolution destroyed families, not segregation.
>Sexual revolution destroyed families, not segregation.
Female participation in the economy and relaxation of divorce laws destroyed families. Previously, women were unable to support themselves and people were unable to end marriages. You could see the same effect if the sexual revolution hadn't happened.
Nazis were actually staunchly anti IQ tests for obvious reasons. Many non-Europeans from Latin America and the Philippines were allowed to immigrate to America in the interwar period. The anti-asian immigration restriction had obviously nothing to do with IQ. I have no idea why you are even bringing up marijuana.
Your dubious historical claims are a likely sign of overwhelmingly motivated reasoning.
Did I say limit myself to Nazi Germany? I said “European fascism,” which had substantial political influence into the 1970s. Other far right regimes had mixed feelings about the Ashkenazi and how to relate them to whiteness.
You can't play this game where you neatly segregate 20th century intelligence research and technocratic credentialism from white chauvinism and white identity politics. There was far too much overlap between these categories. There's just no way to justify your implication that most governments have been colorblind since the 1920s. (Not in PRACTICE, whatever the text of their constitutions may have been).
Were the other fascist movements even that racist? Would not the US be more white supremacist through to the civil rights era than Spain. Were not the European democracies with their colonies not the epitome of white (or Anglo or French) racism. Spanish fascism seems clericalist and nationalist. Jews escaped Nazi germany largely through Spain, after all.
The other fascist movements were way more racist than they needed to be or should have been.
Film review: Enys Men (2022).
The title is Cornish for Stone Island. The basic plot is that the protagonist is alone on an otherwise deserted Cornish island in order to make observations of a rare plant.
In terms of genre, calling it a horror movie might give the wrong impression. It is more of a ghost story, sharing themes (but not exact plot) with some traditional Cornish ghost stories, such as The Mermaid of Zennor.
As the protagonist is alone for much of the movie, there are long stretches without dialog. The movie is heavy on landscape photography: I might compare it to Peter Weir’s Picnic at Hanging Rock (1975), with the proviso that the Cornish landscape offers very different cinematic opportunities from Australia.
I think one of the ideas implicit in the movie is that the Cornish landscape is overlaid with layers of time. Standing stones that go right back to the Iron Age are just there, along with the disused mine workings of the tin industry. So as the camera scans o er the landscape, we are also seeing traces of human activity going back literally thousands of years … which is suitable material for a ghosts story.
I did not find the screenplay compelling. But it was visually stunning and I would give this director another chance. Mark Jenkins. He has a science fiction film called Rose of Nevada that has yet to be released for home viewing.
P.S. The location where the movie is set is somewhat fictional (although supposed to be in Cornwall somewhere).I will be vague as to how old some of the archaeology is the movie is supposed to be as it doesn’t correspond to anywhere real. The real places it resembles range from Iron Age to Neolithic.
Half my X timeline is Leftists saying what essentially amounts to "Alex Jones was right."
https://pbs.twimg.com/media/G8412p6WoAAb7An?format=jpg&name=medium
I don't know who they think this appeals to. They're not going to win over Alex Jones, nor his obese QAnoner audience. All it does is alienate people in the moderate center who might be tempted to vote Democrat because it's the party of serious people who talk about serious things like housing policy and healthcare instead of pedo rings and seed oil.
It's like how boomer conservatives responded to Democratic accusations of racism with "the Dems are the REAL racists, the Democrats are the party of Jim Crow." It didn't improve the GOP's performance with black voters at all, alienated the Republican base, and made conservatives look ridiculous.
If the Dems would move to the center of guns, trans, feminism, and affirmative action while hammering Trump on abortion and the cost of living, they'd have it in the bag. Instead they want to engage in a (futile) competition with the GOP to be the party of prole envy and conspiracy theories.
Politics is no longer about policy it's about attention. In a media-driven culture that's the only thing that matters.
>They're not going to win over Alex Jones, nor his obese QAnoner audience.
That tweet doesn't at all seem intended to. It's pointing out Jones's hypocrisy, not trying to recruit him.
>All it does is alienate people in the moderate center who might be tempted to vote Democrat because it's the party of serious people who talk about serious things like housing policy and healthcare instead of pedo rings and seed oil.
I regret to inform you that serious people in the moderate center actually are talking about the government deploying masked thugs to snatch people off the street. Because that's what ICE is doing.
>If the Dems would move to the center of guns, trans, feminism, and affirmative action while hammering Trump on abortion and the cost of living,
People keep saying this, but I don't think it's true. Biden made his presidency about affordability, the Build Back Better bill was the centerpiece of his presidency, he presided over a steadily improving economy, and it didn't budge the needle on the "vibecession."
(And also, Democrat attempts to shift the conversation to kitchen-table issues very often come off as kind of lame or insincere. "Trump wants to have illegal immigrants tortured in foreign prisons, but that won't lower your health insurance premiums" doesn't make you sound like you're focused on the real issues, it makes you seem bizarrely uncaring.)
...Those comments are being ironic, because Jones thought the pedos were Democrats.
That's the joke.
This is nothing new. Matt Christman was one of the most influential leftists on Patreon for years, and he said at one point "QAnon is 50% right. They have two claims, that the elites are a bunch of criminals and pedophiles, and that Trump is going to save us. The first half is true."
Your dismissive attitude towards proletarian frustration shows that you are locked in the past, intellectually. This is the new political environment and it's not going back to the way it was. Apparently you didn't learn any lessons from the financial crisis of 2008 and the corporate bailouts, which totally transformed the psyche of your average American.
Calling working people idiots and telling them to shut up and enjoy their indoor plumbing and smartphones is political suicide. The absolute worst thing the Democrats could do would be to tack to the center on class. This a big part of why H Clinton and Harris were doomed to irrelevance, although their insane identity politics signaling hurt them as well.
Telling people to shut up and enjoy their lives is a guaranteed way to make them your political enemies. One's spouse or parent can get away with that, MAYBE, but not some smug, self-satisfied elite politician.
Calling the party that put in place the modern racial quota system racist makes the conservatives look ridiculous?
Prole envy and abortion. Thanks for finally stating your true opinion of normal people, Alexander.
There's a certain kind of smug, American optimism that is inherently coercive. "Everything's fine, quit complaining, shut up and do what you're told." I call this Pinkerism and unfortunately, Scott is not entirely immune to this impulse. I guess when you're doing great in life, the benefits of great change do not obviously outweigh the potential costs.
"The conservative will always tell you elites look down on the salt-of-the-Earth working man, but he'll never tell you why."
You're always happy to tell us why you look down on them, but you're not an elite, Alexander, so okay.
I suspect that understanding the real phenomenon you are observing may be difficult without examining the process that selected leftist takes for your X timeline.
The mob bays for blood and to tell them the QAnon thing that's all the rage is bullshit is to make yourself their next target. And this is bipartisan too, combining the schizo lunacy of Pizzagate and the rabid fury of the MeToo movement, so I don't blame anyone for trying to appease them and hoping it runs its course before it turns on them.
Anyone have any thoughts on the COVID treatment Paxlovid?
It appears to consist of https://en.wikipedia.org/wiki/Ritonavir and https://en.wikipedia.org/wiki/Nirmatrelvir which are both fairly standard antiviral treatments, as far as i can tell (used for HIV, among other things) - but perhaps there were pre-existing concerns about these treatments i'm not aware of.
Based on cursory searching i'm willing to accept it as safe and reasonable, but would be interested to hear contrary opinions.
The purpose of the ritonavir in the combination is not primarily as an antiviral, but rather to reduce metabolism of the nirmatrelvir by competitively inhibiting liver enzymes. As a result, Paxlovid has a LOT of potentially dangerous drug interactions to watch out for (similar to the list for grapefruit juice.)
Other than that one big asterisk -- which mostly matters for people already on multiple medications for other issues -- I am not aware of any significant risks from Paxlovid.
It’s known that a fairly large percent of people relapse after initially feeling recovered post Paxlovid, but my understanding is that their relapse is usually a mild version of the illness, and they still end up better off than if they had not taken it.
Paxlovid apparently lowers the probability you will die or be hospitalized from covid by quite a bit, so it makes sense to take mainly if you are at relatively high risk of one of those things happening. It would probably be silly to give to a healthy 25-year-old, but wise to give to an obese 40 year old or any 60 year old. I don't know if it decreases symptoms in most cases, though--mainly it's trying to avoid you getting super sick in the second week of the infection.
The main thing to know about it as a patient is that it only does you any good if you take it within a few days of getting sick--if you wait until you're really ill from covid a week into the infection, Paxlovid doesn't do much good. Which means if you're exposed to covid and want to be on it, you want to talk to a doctor right away--as soon as the home test pops positive, ideally. It clashes with some medicines including (I think) statins, so you have to stop taking them while you're on it.
Stumbled across a pretty good "I can't believe it's not AI" picture: https://upload.wikimedia.org/wikipedia/commons/2/21/Saint_Nicolas_Heures_d%27Anne_de_Bretagne.jpg
Toddlers have six-packs on their backs, right? And gloves kind of seamlessly flow into the sleeves of robes?
(By the way, this painting depicts Santa Claus resurrecting three toddlers who were chopped up and pickled in brine; I am not making this up, Merry Christmas)
The legend says they are children, not toddlers; "children" could be of any age.
Could there still actually be a place for a battleship in a modern navy group?
https://www.foxnews.com/politics/trump-unveils-golden-fleet-navy-battleships-touts-them-more-powerful-than-any-ship-ever-built
Battleships were eclipsed by aircraft carriers in World War II. I don't understand how modernizing the concept can make them serve any role not currently filled with existing ships. The only thing I can think of is that firing a large enough rock (or equivalent) would be pretty hard to shoot down, unlike missiles.
The US Navy is almost entirely oriented towards the idea of the aircraft carrier, with most other surface ships serving just to escort the carriers around. But in a drone world does the aircraft carrier make that much sense? If you take the humans out of the aircraft and make them smaller then you don't need such a huge ship to operate them, you can have smaller ships operating fleets of more specialised, semi-expendable drones, and still capable of launching missiles and having big guns too. You can launch drones off a big rail, and land them in the sea, and you don't need all that deck space just for aircraft operations.
In a world where carrying aircraft is no longer a specialty worth basing your ship's entire design around, it might make sense to have capital ships which are big multi-capability platforms with drones plus missiles plus guns plus anti-submarine capabilities. Battleship seems like as good a designation as any.
Trump et al are even less likely than the median politician to have any idea what they are talking about wrt something like this, or to care that they're speaking nonsense.
Here is bean pontificating on this topic
https://www.navalgazing.net/The-Drone-Revolution
I could imagine an offshore artillery platform being useful for delivering firepower against some very low capability opposition. As in, if you just want to put bignum tons of high explosive on target, and the other guy can't really shoot back, then shells are much cheaper than missiles.
Not sure what this low capability (but coastal) opposition that we want to pound is though. Even the houthis have missiles. Also this floating artillery platform probably doesn't need to be armored.
Even in that situation, I think guided missiles usually come off better, because you don't *need* bignum tons of high explosive to be sure of hitting your target.
The Zumwalts were intended to do shore bombardment before they got cancelled, and the vision there was that they'd fire guided shells, which gives you some of the advantages of both - cheaper than a rocket, but still very accurate. But you don't need a battleship-sized platform to do that.
A C-17 loaded for bear with Rapid Dragon (https://en.wikipedia.org/wiki/Rapid_Dragon_(missile_system)) pallets might put comparable mass downrange and certainly costs much less than any USN vessel.
Missiles cost more than shells?
C-17s cost less than capital ships.
Plausibly C-17s + missiles still cost less than capital ships + shells.
Depends how many missiles/shells…
Some back-of-the-napkin-math:
- A single JASSM costs ~$1.5M
- A Ford-class carrier costs ~$13B (best proxy)
- A single M982 howitzer shell costs ~$100k (reference point)
JASSM range is ~25x that of M982. If we favorably assume that the shells are 1:1 substitutes for missiles, and that the shells' cost only scales with the square root of range, the breakeven for ordnance cost savings offsetting the ship acquisition cost would be 13,000 shots.
The total number of JASSMs expected to ever be built is 10,000. The entire inventory would be consumed by ~220 Rapid Dragon sorties (which is maybe one year's worth of a single C-17's operational cadence), so the share of aircraft acq cost is negligible.
And it's implausible that the shells could be perfect substitutes for the missiles, even with linear-in-range cost increases (at which point the missiles would win on unit cost alone).
By postulate the opposition this floating artillery platform is pounding don’t have supersonic missiles.
Calling them "battleships" doesn't make them all that similar to WWII battleships. They'd really end up being not that different from a bigger Arleigh Burke destroyer. I'd be happy to defer to people who have more knowledge than I do, but I don't really see the point in the US Navy having surface combatants bigger than an Arleigh Burke. The previous plan was to get rid of the cruisers and destroyers and replace them with the DDG(X). This battleship would be a bit bigger, but if I could choose between 20 of these "battleships" and 75 DDG(X)s, I'd be inclined to pick the latter.
If technology progresses in such a way that it becomes much, much easier to shoot down drones and missiles that could theoretically reopen the combat niche for the battleships. Of course even then the only real use case for them would be a conventional naval war with China.
Also boy howdy, "We're not producing enough ships, better make a brand new ship design using tech we don't even have a good prototype for with an insane timeline!" Yeah, that sounds exactly like a fix for not producing enough ships.
AIUI, WWII-era carriers dominated battleships in range, precision, & air defense.
Looking at the proposed kit for the new battleships, IF they were all successfully deployed (spoiler alert: they won't be) then at least range & precision could be comparable and at a fraction of the per-shot cost.
Carrier air groups don't put that much iron on target compared to the bomber fleet, so apart from a blue water conflict with a near-peer adversary it's plausible that the Navy is overinvested in air power.
Full disclosure: former USAF officer, so very familiar with aircraft operations, much less so with surface warfare.
according to bean (of Naval Gazing), battleships were not fully obsoleted until the 1950s. Carriers dominated in the Pacific war, but battleships played a key role in the mediterranean and the north sea.
https://www.navalgazing.net/The-Battleship-and-the-Carrier
Sir Humphrey, in 1942 the Indian Ocean Fleet of the RN was on a collision course with 2 japanese carriers. If admiral Somerville would have kept the course a few more hours we would have witnessed a night encounter between the 2 fleets and the japanese carriers might have ended up in the gunnery range of HMS Warspite. We know how that would have ended because Warspite had the best plot armor in WW2.
Merry Xmas and Happy Holidays everyone. And thank you Dan Carlin for making my holiday complete. https://dancarlin.substack.com/p/mania-for-subjugation-iii-now-available
I'm interested in the topic and on the whole the podcast was informative, but it would have been better at half the length. He tends to bludgeon the point he is making into submission, at times I was going "yes, Dan, I got it, you don't have to go on and on and on and on and ON about it!" That, and the fifty repetitions of "I just don't get it" and the annoying quirks of his delivery (rowly-growly voice when he's making a point or being all SURPRISED about THIS GUY DID THIS?, then the trick of speeding up pace and rising in volume - just speak naturally, and if this *is* natural for him, then iron out the whisper-whisper-growl-get louder-louder-LOUDER!!!! punchline! stuff).
I would be interested in part IV, but I have to say that I did fast-forward a lot because of said "and I'm saying the same thing eighty times here just so you get it", "gosh why did they do this like that I just don't understand" and "okay now I'm going to tell you the Really Good Advice Alexander ignored (goes on tangent for about twenty minutes, or it felt like it subjectively, before finally telling us what the Really Good Advice was)".
On the positive side, it did make me want to rewatch the Porus episodes where Al is stomping the Indian armies (and everybody else) into the ground because he *is* just that good 😁
https://www.youtube.com/watch?v=pbX7zQweBEQ&list=PLnSGWGB1aj5kiPOrKhViQskfx2kE6iVyq&index=24
Mania for subjugation? How about a little more mania for putting out episodes faster amirite.
Check out the historian Dr. Bret C. Devereaux at his blog A Collection of Unmitigated Pedantry (ACOUP). My pitch for him is simple: on the rank of reading that I've done as an adult which has most broadened and enriched my understanding about the past, he's in the top two. Maybe even #1.
His ongoing series is about (the debate about) when and what a Phalanx actually was and how it fought. https://acoup.blog/2025/11/14/collections-hoplite-wars-part-i-the-othismos-over-othismos/
His recently concluded series was about peasant farmers; how they lived, worked, and died. https://acoup.blog/2025/07/11/collections-life-work-death-and-the-peasant-part-i-households/ One point I want to add: I think his explanation of how farming village plots were highly balkanized could be important to understanding why technological progress was so slow in the medieval era. Yes, it decatastrophized individual household risk, which is good. But if you're only plowing one furrow at a time you're not leaping to upgrade from wood to iron to steel ploughboards because other things dominate your time, like traveling between furrows.
I was introduced to ACOUP through Devereaux's exploration of how the pre-modern world made iron and steel. (He does focus on Europe but he also touches on the other processes; Indian and Chinese and Japanese). This one in particular accelerated my journey of understanding of energy (as in, literally watts) as the first among equals of all industrial inputs. https://acoup.blog/2020/09/18/collections-iron-how-did-they-make-it-part-i-mining/
And some LotR-posting to round it out, as a treat. https://acoup.blog/2025/04/25/collections-how-gandalf-proved-mightiest-spiritual-power-in-tolkien/ https://acoup.blog/2025/04/18/collections-why-celebrimbor-fell-and-boromir-conquered-the-moral-universe-of-tolkien/
There’s a guy uploading readings as a podcast (AGreatDivorce), I’ll give it a listen
+1, I'm a regular reader and it's great
That name is not unknown here, but it's always a pleasure to see someone discovering ACOUP 😊
Relevant XKCD: https://xkcd.com/1053/
Of course it's not always about discovering it but "okay, I've heard about this enough times that this time I'll *really* delve into it".
The CBS report on CECOT that was shelved from broadcast in the US was streamed in Canada, and can be found online. After watching it, it's clear that Bari Weiss was right in deeming that it added nothing that other outlets like PBS hadn't already published, and the brouhaha over her decision seems to be basically like that story of the college student who wrote a poor essay and then made political hay of her TA's transgenderism/dysphoria.
...so what was the broadcast replaced with? Seems like that's the most relevant standard.
I think it's quite a bit more obvious that this is based on her political views and not her desire to publish the most relevant or novel news. Her extensive interview with Erika Kirk is more proof of that.
Since when is the standard that every story on 60 Minutes contain a very specific type of original information? The report has new, informative, affecting interviews from CECOT torture victims and other new reporting. I could’ve learned all the information about sherpas and Everest in another piece 60 Minutes aired this Sunday from other sources, but in fact I hadn’t, and found it a fascinating and well shot piece, just as the CECOT piece was. It was an excuse to delay and possibly kill a piece she didn’t like for political reasons, just like the other excuses in her note—including her false claim that the Administration agrees these men have a right to judicial review. They argued the opposite and lost all the way to the Supreme Court.
Probably around when Bari Weiss became editor-in-chief, if I had to guess. If she holds programs she's now in charge of to a higher standard than they previously were, good for her!
Why didn't she apply that standard to the Everest piece?
If PBS or the New York Times had aired an almost identical piece a few days earlier, perhaps she would have.
Can you link me this "almost identical piece" on CECOT" aired "a few days earlier"?
Sure: https://www.pbs.org/wgbh/frontline/documentary/surviving-cecot/
and https://www.nytimes.com/2025/12/08/podcasts/the-daily/el-salvador-prison-deportations-venezuela.html.
>Bari Weiss was right in deeming that it added nothing that other outlets like PBS hadn't already published,
What makes you think that was her basis for delaying the broadcast? https://nypost.com/2025/12/22/media/read-the-memo-bari-weiss-sent-to-cbs-staffers-after-yanking-60-minutes-spot/
Did you read the memo you linked to?
> Last month many outlets, most notably The New York Times, exposed the horrific conditions at CECOT. Our story presents more of these powerful testimonies—and putting those accounts into the public record is valuable in and of itself. But if we’re going to run another story about a topic that has by now been much-covered we need to advance it.
Yes. Did you read the rest of it?
Because that is what she said in the memo?
"Last month many outlets, most notably The New York Times, exposed the horrific conditions at CECOT. Our story presents more of these powerful testimonies—and putting those accounts into the public record is valuable in and of itself. But if we’re going to run another story about a topic that has by now been much-covered we need to advance it."
What about the rest of the memo?
What about it?
It mentions several concerns. Why cherry pick one and imply that that is the only concern, or the primary one?
If you have ever thought of starting a Substack but haven't gotten around to doing it, this is The Sign to do it immediately.
I recently began my own. Wrote some vague thoughts. Didn't even think it was good enough to publish but I did it anyway.
To my surprise a few people read it. Commented and liked it.
It's honestly one of the most gratifying things I've ever done. It's another high altogether to create some intellectual work and see it resonate with someone else and to know that I've contributed to the discussion.
Just do it now
And yet this apparently isn't your main posting account, so I'm robbed of the chance to add to your joy.
That was a dumb decision on my part for sure. I'm going to meekly try to text you the link privately through Substack
Just out of curiosity, why not publish it here? Worried about too many dopamine hits? ;-)
A curious person might be able to work it out from which people I've abruptly read and commented on, and vice versa.
I'm not *that* curious! :-)
I'm worried about the self-advertisement policy of ACX :(
I appreciate that you thought about this! Thank you for being mindful of that. I do also think it's fair to share the link since someone outright asked about it, though. :)
Ah, I didn't know there was one. Open threads say, "Post about anything you want". While I appreciate your reserve, the substack interface of showing the link to anybody's substack who has one seems low-key enough that I wouldn't have gone to any trouble to obscure it, even if your post is about the joy of starting a substack.
Re the "vibecession" discussions -- I think there should be some fruitful connections between ideas including:
1. Studies on slack https://slatestarcodex.com/2020/05/12/studies-on-slack
2. Erik hoel's thoughts on culture being "overfit": https://www.theintrinsicperspective.com/p/our-overfitted-century
Especially this tidbit: "I can’t help but wonder if some of the complaints about “late-stage capitalism” actually break down into something like “this system has gotten oppressively efficient and therefore overfitted, and overfitted systems suck to live in.”"
3. Scott's recent vibecession explanation idea along the lines of "maybe it seems to take a lot more effort to achieve slightly better results nowadays?"
Does it seem to others like there is something here? Also are there any existing posts that already make and flesh out this by any chance? I'm not sure how one could turn ideas like "maybe there is more optimization pressure, less slack, more overfitting, and it doesn't show up in anything easy to measure but it's a real problem that feeds into the bad vibes" into a coherent story and falsifiable predictions...
I do not say this lightly, but I believe my Fractal Stalemate provides the structural and political mechanics for the cultural phenomenon Hoel describes, though as my analysis is systematic, it's much looser in its rigor.
I invite you to come and take a look. This is the "canonical" release order, for your convenience.
https://bargainbinseldon.substack.com/p/the-fractal-stalemate-a-guide-to
https://bargainbinseldon.substack.com/p/ancient-engine-of-extraction
https://bargainbinseldon.substack.com/p/cold-deliveries-hot-profits
https://bargainbinseldon.substack.com/p/to-arbitrage-suffering
https://bargainbinseldon.substack.com/p/politics-as-theatrics
https://bargainbinseldon.substack.com/p/the-meat-puppet
This makes me think of the above vs. below the API stratification I'd seen years ago (e.g., https://rein.pk/replacing-middle-management-with-apis).
Perhaps the disconnect between metrics & perceptions are that the metrics only matter to those above the API but the perceptions are most vocally coming from below it.
Yes, absolutely. "You can just do things" might be true, but I think it's much less true than it used to be. In particular, "oppressively efficient" is an interesting concept, and jibes with something I perceive about current society vs, say, a hundred years ago: there are now very few opportunities for individual people to get enterprises started. I don't mean VC-powered psychopath stuff with "cap tables" and "exits", I mean nice healthy businesses that get started in garages and grow organically, possibly ultimately into household names. Overall you can't really even call it a bad thing, because it's largely the result of economies of scale and regulations; the products and services society as a whole is receiving are massively better. But yeah, for an individual who doesn't want to be a cog in a machine, it sucks.
That said, not everyone would even want to get involved in this sort of thing, so it's definitely not going to fully explain the vibecession stuff.
Another closely related concept that's a favorite of mine is just complexity in general. Nobody cares about limiting complexity its own sake, but it's becoming a real problem. In the happy path, things are orders of magnitude more efficient, but the failure modes have become way more miserable. Your grandparents might have had some real struggles in their day, but they didn't have to navigate anything like the Kafkaesque hell that modern customer service can descend into. They would immediately be in contact with a human interfacing with a simple paper-based system, with the ability to do whatever was needed - as opposed to "sorry, the button on my computer screen is greyed out; I don't know what else to do."
I'm not sure what precisely Hoel means by overfitting (he acknowledges that he's using it in a few different senses) -- but I broadly agree with his idea. I would argue that overfitted is almost the opposite of efficient - it's what happens when the system has too many resources and underdetermined goals and so wanders into Baroque details and performances which don't progress toward the actual agenda.
I think this is the same sort of failure as Orwell pointed out 80 years ago, where the enemy is simultaneously 'weak and contemptible and we are sure to win' and 'immensely powerful and all measures must be taken'. Maybe the system can both be wringing every last drop of efficiency out of workers through electronic surveillance and performance management; but also massively overfitted and forcing them to do bullshit jobs and further projects that produce no conceivable benefit.
I think that a large part of the paradox can be explained as "goodharting", or the difference between *actual* productivity and the *signals* of productivity. As a manager, it is difficult to make your team more *productive*, but it is easy to keep them *busy* all the time... and that kinda seems like the same thing (except it is not). So the mediocre managers keep sharing tips on how to keep their employees more busy, thinking that it makes them great leaders. (Measure the bathroom breaks!)
Even when someone comes with a genuinely good idea, it is immediately perverted by the stupid and/or malicious followers, who adopt the buzzwords while eliminating the essence of what made it a good idea. (For example, there is a sensible idea that if you want to improve something, you should *measure* it first. You can pervert this idea by measuring the wrong things and ignoring the important ones, or by introducing a system of measuring so annoying that doing the bureaucracy wastes a lot of time and energy. Furthermore, you can make your research prove whatever you want it to prove, by simply eliminating the "wrong" answers from the questionnaire.)
For example, in programming it is ironic how in 1990s a group of five developers could produce an entire game, and its sequel, in the same year; but these days it would take the same group, and at least the same number of managers on top of them, a few months to add a checkbox to an existing application. The developers today would have plenty of meetings to attend every day, Jira tickets to fill, and maybe a list of annually evaluated self-improvement individual goals and team goals as a cherry on top. Their managers would produce PowerPoint presentations and burn-down charts, and their boss would collect a huge bonus for making the company more agile or whatever is the buzzword of the year.
If you want to understand why developers were so much more productive in 1990 I would read The Mythical Man Month.
TL;DR. Modern programs are too big for 5 people, and that requires a lot of overhead like Jira tickets. Which is not to say bad managers can't be part of the problem. But Jira tickets evolved to fill a real need and not just "goodharting".
Ah, yes and no. Yes, the modern programs are more complicated -- when the program in 1990s e.g. solved an equation, it could read it from standard input and write the solution to the standard output. These days, the same program would have to be implemented as a REST endpoint with authentication and encryption, password rotation, log rotation, installed in a container, and dozen more things; the actual equation solving would be just a tiny part of the whole. That's how complicated things are these days.
On the other hand, most modern managers still haven't learned the lessons from The Mythical Man Month. They still treat programmers as perfectly fungible; they are not afraid of losing an experienced developer who worked ten years on a project and has tons of tacit knowledge, because they believe they can easily replace him by a random guy they hire tomorrow... and if they can't hire a new guy tomorrow, then the old guy's work will be assigned to a random colleague on top of that colleague's existing duties; eh, it will work somehow.
A few times I saw a project advance dramatically when the responsible manager got sick for one month and the developers worked unsupervised. In one case I was the only developer on the project, and I have completed it during that month, and the company paid me a huge bonus afterwards. But when I suggested that I could work on the next project the same way, the company refused, because it was against the rules or something. (I quit soon afterwards.)
That one month when I worked on the project alone, this is what I did: The first week I just studied the requirements, to make sure I understand everything that is needed to do. The second week I got familiar with the tools: I studied the tutorials, made a few proofs of concept, etc. At the end of the second week I sketched all program dialogs, and showed them to the customer: "hey, if this is what I will do for you, is that exactly the thing you wanted me to do?". After approval, I spend the remaining two weeks coding, uninterrupted, and delivered what I promised.
Here is how the same thing would most likely work *with* a manager: The manager would talk to the customer, and would keep the requirements to himself. He would spoon-feed me the information, as he would deem necessary. That would sometimes require that I throw away something I already made, and do it differently in the light of new information (and prepare the install scripts both for the old and the new versions), but that's called "agile". We would have a meeting where I would have to provide an estimate how long the work will take (without knowing many relevant details), and then once or twice a week we would have a meeting where I would report my progress (always reminding me that the arbitrary deadline is coming closer). The code would be written in the order decided by the manager (despite the fact that he has zero programming experience); for example he might decide that logging should be added last, and unit testing skipped entirely. He might randomly assign another developer to help me with the project for a few days, despite the fact that the other guy knows nothing about it, and it will take him those few days to get familiar with the thing. Just like that, in the middle of the project he might assign me to a different project, because they urgently need some help (even if I don't have the kind of knowledge they are looking for). After three or four months, the project would be complete, the manager would collect the bonus, and one of the developers would probably be burned out and quit.
(My problem with Jira tickets is that they usually come with some kind of workflow that is completely unrelated to how things actually work, and in theory you can adapt the workflow to your project's needs, but in practice, adapting the workflow would be a waste of resources you won't get approved, and you are required to use it as it is by default. So everyone keeps guessing whether "Resolved" means that the feature was *implemented*, or that the tester *approved* it, or that it was *installed* in production. Making a convenient filter for "things that I am supposed to actually work on this week" is sometimes impossible.)
> Yes, the modern programs are more complicated -- when the program in 1990s e.g. solved an equation, it could read it from standard input and write the solution to the standard output. These days, the same program would have to be implemented as a REST endpoint with authentication and encryption, password rotation, log rotation, installed in a container, and dozen more things; the actual equation solving would be just a tiny part of the whole. That's how complicated things are these days.
You imply that this is a ton of junk surrounding a similarly actually useful core that hasn't improved since the 1990s.
I disagree. REST endpoint means its available anywhere in the world and yet can be managed from a single centralised location. If your equation is calculating tax on sales, and the government changes the tax, a central location makes it a lot easier than rolling out an update across 1000 work stations.
As for security, I don't think I need to say why security is a good thing.
And writing a rest endpoint isn't harder than writing a read from standard input. Throw a couple of annotations into your code and its done; (granted if you count the complexity of the appserver...)
Zvi has probably the strongest / most economically coherent take on it IMHO: https://thezvi.substack.com/p/the-revolution-of-rising-expectations
I think it fails to tackle where this revolution of rising expectations come from. Ofc it's only my gut feeling, but I believe this is one of the major failing point of capitalism:
Investors want more dividends -> Companies need to sell more -> Companies create ads to make people want buy more stuff -> People's expectations raise significantly
I'm not saying that this makes capitalism a bad system, but I believe we need to work out around this flaw as a society if we want this system to be sustainable on the very long term.
I actually think it's more driven by social media than traditional advertising: plenty of ads were all over the place in the '50s/'60s/'70s. I think the peer comparison that happens on social media (both your literal peers, and also "peers" in the form of national/global influencers) exerts a much more powerful sense of comparison and "need for more" than corporate advertising. The infamous "Gen Z says they need $500k/yr to be successful" stat is probably more a results of try-on videos, shopping hauls, and #richkidsofinstagram than clever marketing execs at BMW / Prada / etc.
I think you are both wrong. This age is not more consumerist than the past and might be less consumerist. The definition of successful is going to be context or generational dependent. Does it mean comfortable? I’d out it at a higher level than that. And boomers would be living high on much lower nominal income anyway - given they don’t have to save for pensions, pay mortgage or rent or live close to a cut for work.
You completely misunderstood Zvi's article, he is saying the exact opposite of you. There's a reason gen Z is saying that, and you're being an ass for not trying to understand why.
Did you even read the linked article, or were you just skimming for a quote to get outraged about?
The Economist publishes sort of a correction:
"Women in America are having as many babies over their lifetimes as they did two decades ago"
"For decades, across the West, the commonest measure of how many children women have, on average—the total fertility rate (tfr)—has been dropping. It has fallen well below the “replacement rate” (needed to keep a population constant) of 2.1. Between 2007 and 2022 America’s tfr dropped from 2.1 to just 1.6. Warnings of probable population decline, including from this newspaper, abound. Pronatalists are casting around frantically for solutions....
But tfr has flaws. It measures the number of children a hypothetical woman might have in her life, based on current birth rates for women of different ages. To calculate it, you start by dividing the total number of births to a given group of women (aged, say 15-19 or 35-39) by the number of women in that group. Those group birth rates are then added up and multiplied by the number of years in each group (ie, five). The result is the number of children a woman will have if, aged 15-19, she has the same number of babies as the average woman of 15-19 is having today, then the same number at 20-24 as the average woman of 20-24 is having today, and so on. It is almost sure to be wrong. That does not make TFR useless. It captures what demographers call “tempo”—the timing of an event, like a birth. But as a measure of “quantum”—the total number of births—it is likely to mislead. In effect it assumes that a young woman will have babies not only at the same rate as her contemporaries, but also at the same rate as women five or ten years older. That is not necessarily so.
A different measure, the completed fertility rate (CFR), captures the average number of births a woman has by the end of her child-bearing years (put at 44). America’s CFR has not fallen at all over the past two decades. In fact, according to an analysis of census data by Mike Konczal, a former Biden administration economist, it has risen slightly, from 1.91 in 2000 to 1.97 in 2024....
Many demographers, including Lyman Stone, of the Institute for Family Studies, a think-tank, point out that tfr is typically a strong predictor of cfr. But there is evidence that women are deferring, not forgoing, childbirth. In 2000 by age 26 the average American woman had produced one child; the average 32-year-old, 1.6 children; the average 40-year-old, 1.9. In 2024 the average 26-year-old had 0.6 children and the average 32-year-old 1.2. But the average 40-year-old has still had 1.9, having delayed in her 20s and caught up in her 30s.
Does this mean...that tfr will rise to meet cfr? Historical examples are suggestive. As women delayed childbearing in Sweden the tfr dropped sharply in the 1980s, before recovering in the 1990s (though it has fallen since). The cfr, at around 2, barely budged. In America the question is pertinent. The commonest ages are 33 and 34, because of a baby boom in 1990 and 1991. Women aged 34 have had, on average, 1.46 children each. As this group approaches their 40s, many more births may be on the way. As for younger women, those born in 2000 have had fewer children than those born in 1990 had had by 2015. Perhaps they will catch up—but that won’t be known for at least a decade.
To be sure, an average of just under two births per woman is still not “replacement level”. But if that truly is the long-term average, panic is scarcely justified. It would mean a gradual population decline: tricky, but much more manageable than implied by TFR...."
If tfr and cfr are strongly correlated, then cfr is likely a lagging indicator. We won't know if birthrates are really falling for young women for decades, though intuitively it would be hard for women 30+ to have 3+ kids that would balance out those who had 0, 1 without major advances (we're not there yet).
Given that uncertainty, the lengthy times to affect birth rates through public policy, and the huge downsides of population collapse, probably best to encourage procreation now just to be safe.
You mention Lyman, and he's explictly looked at the TFR -> CFR correlation in the US and in Sweden and many other countries.
In general, they enjoy a shocking high amount of correlation, in the US it's .9+.
https://imgur.com/a/OunX66h
And the Sweden data:
https://imgur.com/a/Wg0YIJv
There are real physiological reasons we'd expect deferring births to lead to notably lower total fertility, too.
When you wait until early 30's to have your first child, as is the norm in the US and UK now, you have already burned through ~2/3 of your cumulative lifetime fertility (the area under the curve in this graph leading up to age 30):
https://imgur.com/lpADyJs
That's from Geruso 2023, Age and Infertility revisited. Broadly, you go from 60% chance of live birth in early 20's to 30% at age 30, 11% at age 40, and ~1% at age 45:
https://imgur.com/sCqfS6z
That's a REALLY big difference in fertility. Many, many more people are going to try and fail in their thirties and forties versus in the halcyon fifties when women in their 20's were having kids.
Maybe we should make IVF free / state funded? Sure, you can, I think it's probably incrementally net positive. But many EU countries have done this already, along with entirely state funded daycare, significant baby bonuses, and more, and haven't moved their overall fertility in the positive direction.
You know what HAS been tried, unsuccessfully?
* $10k bonuses per child (Singapore), or for 2nd / 3rd children (Russia)
* 3 years paid parental leave (France)
* 480 days paid leave at 80% wage (Sweden)
* Income tax exemption for mothers with 4 or more kids (Hungary)
* Free state paid child care (France, Germany, Sweden, Norway, Finland, Estonia)
* Free IVF (most EU countries)
* $750 / month payments per child (South Korea)
So the problem is pretty hard to move the needle on.
https://www.cremieux.xyz/p/storks-take-orders-from-the-state
Structure of my proposal:
Married woman under 25 with one child get free 3 bedroom housing. Under 30 with 2, 4 bedroom housing. Under 35 with 3,5 bedrooms. Etc.
Its paternalistic, forces them to "spend" the subsidy on excessive housing. I dont want them wasting the money on 0 sum status goods like private school, or even day care which encourages a return to work and reduction in future futility. The money directly contributes to ability to have *and grow* the family with the extra bedroom
It spends nothing on 2 child couples in 30s. Those are the default waited til we could afford it, hitting the minimum for replacement. Great for those ppl. They dont need a subsidy for that.
The expiration creates a strong psychological nudge to keep going.
Maybe on some margins it incentives better govt policy on home construction as they are the ones paying for it.
Still its pretty expensive. Could still fail. But I think its more thoughtful than just shower parents with goodies
> Married woman under 25 with one child get free 3 bedroom housing. Under 30 with 2, 4 bedroom housing. Under 35 with 3,5 bedrooms. Etc.
Okay, ran some rough napkin math on this.
For the top ten CBSA's by pop, this costs you ~$25B a year. For the top 20, it's $34B and the top 50 $46B.
The top 50 CBSA's is more than half the US population, and it includes basically everywhere with price problems. You're getting down to Ogden, UT and Bakersfield CA and Colorado Springs by that point, places with relatively reasonable house prices and rents for the area.
So prospectively even if you expand it to basically everywhere, it's going to cost <$100B a year, which is essentially rounding error in the Federal Budget.
Link to sheet: https://docs.google.com/spreadsheets/d/1afqEb-A_YnhUYiAUmMJVg-7K6XaxFZL5wp2exKBEU48/edit?usp=sharing
As a point of comparison, here's some stuff we waste $50-$100B on per year:
* Oil and gas subsidies and tax breaks, $50B
* State and local tax breaks and subsidies for stadiums: $50B
* Foreign aid: $50B
* Annual upkeep, utilities, and security on unused federal buildings: $50B
* Improper or fraudulent Medicaid, SNAP, or unemployment payments: $100B
* Maintaining unneeded and underutilized military bases: $50B
I'm all for it, I think you should write a Substack post about it.
Insert meme:
Can we stop restricting supply?
Best I can do is subsidize demand
Its meant to be a subsidy on the supply of new children, not an effort to bring down housing costs. It would be judged based on how many new births it creates.
I would also like much more liberal house construction policies, but I am not confident it would make fertility problems disappear (yimbys love japan housing policy after all). If you are maybe you should just say that. Anyways wed both agree on reducing housing restrictions.
I think its generally true that any increase in supply somewhere would increase demand elsewhere (usually in housing) so youre proving too much with the meme
It doesn't even subsidize demand, given that it doesn't eliminate preferential alternatives. Which is just not having children.
I love everything about it! Sure beats the hell out of spending trillions of dollars on rich old people every year, spending that I don't think anybody currently below 40 expects to exist by the time they're old.
And it brings fertility to the areas and people that need it most, our insane urban metros staffed with all the PMC people. I have no idea what a 5 bedroom place would even cost in NYC or SF, but it's gonna be a lot, so it's scaled directly to be maximally relevant, where it matters most.
I'll start doing some Fermi math on this to see just how crazy it can get.
EDIT: did the math. Even if you expand it to basically everywhere, it's going to cost <$100B a year, which is essentially rounding error in the Federal Budget. We waste 5-10 times that on completely irrelevant stuff every year, see my above comment.
Link to sheet: https://docs.google.com/spreadsheets/d/1afqEb-A_YnhUYiAUmMJVg-7K6XaxFZL5wp2exKBEU48/edit?usp=sharing
CFR is _completed_ _births_, not pregnancies or attempted pregnancies. So lower odds of successfully carrying to term at higher maternal ages are priced in to it. Women who've decided to have a child at 40 despite 11% odds, do not stop trying after one (or in my personal observation two or three) failed attempts.
Also the US CFR is currently trending _up_: from 1.86 as of 2006 it hit 1.97 for 2024.
All of which makes the Economist's closing point, above, sensible.
To your last point with all the different governmental schemes failing to increase birth rates: I think you’re right directionally, but in the two cases I know a little bit about, the details are off.
To my knowledge, France has paid parental leave for a few months, and unpaid (but with ongoing health insurance etc) for a year.
In Germany, child care is subsidized differently from state to state, and even district to district in some cases, but rarely really free. What parents pay depends on the state and their own income and can range from 0 to a little over 1000€ a month.
https://www.service-public.gouv.fr/particuliers/vosdroits/F2280
https://www.iwkoeln.de/presse/pressemitteilungen/wido-geis-thoene-wo-kitas-am-teuersten-sind.html
It would be interesting to model the effects of a constant-CFR explanation for TFR on entitlements and taxes. Probably there is still reason to panic, right?
If you assume that only one generation is getting SS+medicare before the change, then each retiree has fewer descendants to tax. But if there are currently multiple generations of retirees, dropping that to just one could (eventually!) help.
I think in either case, the short term still sucks for everyone
But there is evidence that women are deferring, not forgoing, childbirth
I think the problem is that statistically, deferring will necessarily mean foregoing as success rates will be lower at higher ages and it eliminates the possibility of larger families.
The trend dynamics are interesting though. If you imagine a 20 yr shift from one state to a new attempted-deferred-but-with-some-failures state, then TFR will drop quickly as the younger age brackets have fewer births. These will eventually be offset by higher birth rates in older groups but dont show up yet. CFR is slow to drop. Then as the shift completes TFR rises back to converge on a new lower CFR. In this scenario TFR is a leading indicator but also an overshooter.
So I can buy the overall narrative you are suggesting. TFR may be over stating the decline but there is still a real decline and we are now below replacement rate, partly because ppl are waiting too long to start.
My policy intervention would focus subsidies on younger parents. Not all parents.
CFR is _completed_ _births_, not pregnancies or attempted pregnancies. So lower odds of successfully carrying to term at higher maternal ages, and reduced odds of larger broods, are both priced in to it.
Women who've decided to have a child at 40 despite 11% odds, do not stop trying after one (or in my personal observation two or three) failed attempts.
Also the US CFR is currently trending _up_: from 1.86 as of 2006 it hit 1.97 for 2024.
All of which makes the Economist's closing point, above, sensible.
I understand the definition of CFR and its a positive sign that it is going up. I was referring to the narrative driver. All things equal women waiting longer will drive down births and so CFR, no matter how hard they try. If it is rising right now its for another reason than the timing effect. We should still be concerned about the decisions to wait and try to address factors that could be contributing it that arent all about preference. The housing theory of everything looms large here
"All things equal women waiting longer will drive down births and so CFR, no matter how hard they try."
You are imposing future certainty well beyond what current facts will support. Put another way, "all things equal" is carrying a _lot_ of weight there.
E.g. why are you assuming that all the women who don't have children until late, are among those who otherwise would have had children earlier?
E.g. modern medical science has obviously _improved_ the chances of successful procreation past 35; why do you assume that progress has now ceased?
Other obvious real-world uncertainties/variables exist. Clearly though you're just dug in on your narrative making further dialogue is pointless, so, bye.
My first post was supportive of your overall view but insisted its dangerous to be too complacent on rising birth ages. This is getting pretty hostile. I dont make any of the assumptions you accuse me of.
I think it is a valuable PSA that waiting necessarily reduces expected total births. Improved medical science can help, but it wont eliminate the reduction. The harder it is to conceive that 1st, 2nd or 3rd child the less likely the couple will even try for that 2nd 3rd or 4th.
I myself ended up with fewer children than I want because I was too complacent about the years. I want an information and policy environment that makes that less likely for new parents. If you would rather wishcast the challenges away, youre not helping. Bye
Yeh the distinction between tfr and cfr is very important and lost in the noise about demographic changes - particularly China.
The tricky thing about anti-discrimination is that if you've been discriminatory in the past, and turnover is pretty low, then just to get to something approximating the racial balance in American in the near term would seem discriminatory in new hiring.
Like if your company basically hired so few black people that it's only 1% black folks, and they have an annual turnover rate of 10%, then to get to 13% black folks (the American average) you'd basically have to spend more than a year hiring nobody but black hires. If you only have 13% of the hires being black, then in practice the percentage of black employees at your company is only increasing by slightly more than 1% of the company per year - that's a whole decade just to rectify the situation, if not more (assuming there's no recessions that lead to layoffs, etc).
The "trick" is to have a realistic understanding of what non discriminatory hiring looks like and not expect something else. I think there are many who understand what that looks like, but want something else anyways. So the second "trick" is to ignore those ppl. That part is harder than the first, but the whole DEI backlash is about giving people in power to do just that
If the problem with discrimination is that you're hiring people on the basis of demographics rather than merit, it seems like "let's hire people EVEN MORE on the basis of demographics rather than merit, but different demographics this time" makes the problem worse rather than better.
On the other hand, if your model is that anyone at all can do this job, it doesn't matter if they have any skills, so jobs are literally just handouts -- in that case, yeah, you should probably discriminate in the opposite direction to get back to parity in who's getting the handouts.
If an organization is discriminating in a direction opposite to what they've done in the past in order to reach an overall balance in their demographic composition, that's generally only useful as a way to superficially avoid the appearance of unfairness (by obscuring the evidence of their previous discrimination). In some cases perhaps it's useful for a group making important decisions to come from a representative range of backgrounds, but in most cases this approach is just straightforwardly a bad idea.
Why?
Or, rather, which assumptions make that true, and shouldn't we be arguing about those instead?
For example, I might assume that there is an equal rate of qualified candidates from each race, but that some form of bias - maybe subconscious racism, maybe the result of social networking being partially segregated, maybe cultural deficits in resume-writing or economic deficits in resume-padding that don't affect actual job performance, or maybe any combination of a hundred similar factors - has led to blacks being hired at 1/10th the rate of whites.
If that were true, you should expect that there's a huge untapped pool of overqualified black candidates out there, and that if the scales were suddenly lifted from your eyes and you could estimate true job performance from a resume, you'd get massive gains from hiring only black candidates for the forseeable future.
I hadn't considered that the same candidates who were unfairly passed over in the previous hiring regime would still be looking for the same sort of positions. That seems like it would be only partially true.
It is possible for people to be racist and for that racism to be justified. In that case, there is no huge untapped pool of overqualified blacks, and any effort to hire blacks for the sake of exploiting untapped resources is worthless unless you can pay them less than whites, which is obviously not possible right now.
It wouldn't "seem" discriminatory. It would be actively discriminatory.
Northern Ireland has that problem with making the police force more acceptable to Catholics. To do it quickly would be 100% Catholic only hires for decades. Widely they do it slowly.
In my country (not the US), there was an acknowledged problem with women accessing the higher ranks of academia. I saw recent research which suggests that 75% of new professorships went to women over the past 3-5 years (I don't remember the timeframe). As far as I can tell, this wasn't because of the imposition of quotas or rules, but each university decided to remedy the figures through direct action.
Research bodies in the country like to announce that 'over 50% of awards in every discipline were made to women' -- again this is fine. But they should publish success rates by gender if this is so: the end result will be to push universities to hire women, who have a better chance of being successful, until gender balance is achieved. But this should be made explicit: men should be aware that their chances are depressed as a result.
I don't have any argument about this - it needed to and needs to happen. But when it's a net zero game, creating winners has to result in losers as well, and this should be acknowledged.
Nah. That creates a program that has to be dismantled and runs the grave danger of creating vested interests.
Sex discrimination needs to happen now because sex discrimination occurred in the past? "Not doing sex discrimination" is not correct?
'Don't do murder' is correct, even if self-defense, militaries, assisted suicide for terminal patients in pain, and abortion should all be legal.
We just call them something else other than 'murder' so that we can pretend we have a simple, bright-line, one-sentence policy. 'Don't do murder'.
Of course, we don't have *any* policies in reality that are that simple. Whenever we try to make it look that way, we're just playing semantic games.
We can play semantic games on this topic, too. Call it 'equity' or whatever, it doesn't matter. It just boils down to 'Discrimination is always bad, unless it's the type that produces good outcomes we want, in which case we call it something else.'
Same as for murder, theft (taxes, child support), kidnapping (carceral justice system, child protective services), etc.
It seems that if you want to move towards a world in which hiring decisions aren't made on the basis of sex then perhaps you should move _towards_ a world in which hiring decisions aren't based on sex.
The murder analogy does not apply, because we're not comitting murders in order to get to a world free of murder. In fact that is generally called vigilantism and while it has some emotional appeal it's generally between illegal and extremely illegal.
We do allow for self-defense, because moving to a world free of killing is not one of our goals. And because self-defense is materially different from murder.
It's unclear to me, that discriminating in favor of men is materially different from discriminating in favor of women, with very few exceptions.
Lets say we had the omniscient God-AI scan every particle on Earth and tell us with absolute infallibility that women are being undermentored during education and training and underestimated by recruiters such that the competence of a woman employee is actually 10% higher than what the average recruiters estimates, and that if every recruiter hired the best marginal 10% more women than they had been intending to (or however that math works out), this would actually lead to perfect meritocratic recruitment with optimal efficiency and performance.
This is sex discrimination. It is telling recruiters too hire 10% more women than they wanted or intended to.
Do you think this type of discrimination is wrong?
This uses sex as a proxy for competence, I take no issue with that. I would file that under the "very few exceptions" I mentioned above, and I hope you will grant that this is indeed an exceptional situation.
Honestly I'm not even sure I'd call that sex based discrimination, or at the very least the central example of sex based discrimination is to discriminate against somebody _because of_ their sex. Here we're attempting to discriminate by competence, and use sex as a proxy for competence.
If you compare it to hiring somebody for the purpose of bringing the gender balance closer to 50 / 50 it seems substantially different, to me anyway.
Whether it's right or not, I've accepted that this is the way the powers that be have decided to run things. I made the decision to leave academia and find work elsewhere, that wasn't the most compelling reason.
For the record, I do think that creating a more equal society is a worthwhile goal. I don't think this is the right way to do it, and I think there's a case to be taken there; I'm not willing to invest the time, money and reputation to do so.
Then why the comment that it "needed to happen"? Why does there need to be gender balance anywhere when the genders aren't equal in the first place? They should be hiring on merit, using the best people with the best properties for the job.
Over the past 20 years, there were a number of cases where clearly qualified women were overlooked, took court cases and obtained promotions because the universities had provably discriminated against them. The universities needed to correct these issues - there's been generational change in the higher leadership roles and the new generation seems to have over corrected.
Merit is hard to establish for academic positions (actually I'm negative on hiring in general as a way of establishing fit for a job). Track record 3 years out from a PhD is a poor indicator of who'll make a career-long contribution to their research area, to the university's teaching and to whatever service criteria are valued at various times. Much like any workplace, value is going to be created on an 80-20 type rule.
There's no requirement to reach the population share by some specific date. Just stop discriminating and let the new equilibrium be reached.
The problem is that quotas of any kind are explicitly illegal per a SCOUTS ruling
That doesn't matter, in real implementation it's quotas with a few extra steps.
Curious if this was intended as a reply to the thread about the Jacob Savage article in Compact.
It seems that one of the fundamental challenges of the day is that people of differing opinions aren't talking to each other. I've frequently tried to reach out to people with strongly different opinions (e.g. people who oppose vaccines, Trump voters), and while in a couple cases it has led to fruitful conversations, it always struck people as bizarre and unconventional. This is strange to me given that it seems everyone agrees that "echo chambers" are one of the fundamental problems of the day, which everyone acknowledges and bemoans.
Most striking is that I just don't know of any major platforms where you can attempt such dialogue. I think it's clear that Facebook and Twitter are hardly good platforms for this, not to mention comments sections. I've seen a couple attempts (e.g. Braver Angels' "1-on-1 conversations: https://braverangels.org/online/1-1-conversations/#participate) but they seem not to have really gained much traction (as I understand it there's always a dearth of right-wing/Trump supporters). What website, app or location can facilitate me either finding someone with different opinions to talk to, or alternatively suggest myself as someone open to this kind of debate? Do you know of places where good discussions of this kind take place? And why does it seem like these things are so rare if it's a pretty basic way of addressing a universally acknowledged problem of the day?
IHere’s an article about an observant Israeli Jew who, mostly via Twitter and other online contact, completely convinced a highly committed member of the Westboro Baptist Church to change her mind about pretty much everything. https://www.newyorker.com/magazine/2015/11/23/conversion-via-twitter-westboro-baptist-church-megan-phelps-roper
I remember reading that!
I actually know tons of stories like that-- heck, there are lots of stories of Palestinians who will say things like "My brother was killed by an Israeli soldier, I was planning to be a militant to kill as many Israelis as possible, then I met some Israelis and I became a peace and coexistence activist" (one example: https://www.youtube.com/watch?v=BzDEFAGv_hc). This and many other examples (plus personal experiences) make me absolutely convinced that dialogue can work phenomenally well, but "can work" and "does work at scale" are not the same thing. Still, it seems like a mostly untested question, which I really think is bonkers.
I think the key to the Jewish guy’s getting through to the Westboro woman is that he did not ever express annoyance, hurt or contempt. He did challenge her religious ideas, but did it in a jokey way, never with any indignation or scorn. I love that guy.
I think it helps, but sadly it isn't a panacea. I engaged with a couple people who responded just as disingenuously when I was trying to be friendly and expressing "we're on the same side here" and such than when I did become a little exasperated.
I think the key is that there are lots of different kinds of people, some are the eternally online types for whom it's hard to break through even with lots of good will, and many who are much more open to dialogue even if extreme. I think part of the problem is that, more or less by definition, you're more likely to encounter eternally online people in FB comments and such. But not necessarily so if the site is dedicated to intelligent debate (eg this comments section has unusually high quality debate).
Well when the Jewish guy, whose name I can't remember right now, first started a dialogue with Megan, the Westboro woman, she was saying that God hates the Jews because they killed Jesus. She wasn't being disingenuously nice, she was hitting him with the worst she had.
Oh I don't think people like that are being disingenuously nice, I think they're being genuinely offensive, and I've had plenty of stuff like that thrown my way (I once had a conversation with an anti-vaccination person who accused me of being a holocaust denier-- because I denied that doctors were committing a holocaust on patients by administering the vaccines. She also would regularly denigrate my mother after learning she was a doctor).
The people you want to get to are the genuine true-believers, rather than the opportunists or the disingenuous ones.
I think people of differing opinions talk to each other too much, in fact. Having to interface with some of the worst people on the planet every hour of the day is one of the reasons we have such incredibly toxic institutions today.
If I'm defining the people I want to talk to as Trump supporters, that's tens of millions of people, plenty of them good people who I think are making a mistake, rather than "some of the worst people on the planet."
Out of curiousity, do you know any Trump supporters personally? Have you spoken to any about the reasons they support him?
To be clear, I'm as anti-Trump as they get. I consider the USAID cuts to be one of the most horrific, and one of the most pointless, single acts in modern American political history. But what I see is people supporting this stuff because the actual death and destruction caused gets filtered out on their feeds, which is instead filled with stories (some true, much fabricated) about the massive waste being cut. And I know good people who constantly fall for this stuff. And that's why I do this, because the way evil wins is not by turning everyone into heartless people, but by convincing good people that they are supporting good when they are in fact supporting evil.
(And of course many Trump supporters will say the same about Biden supporters-- we don't actually care if inflation skyrocketed under Biden because we don't care about people who are struggling etc. etc. etc.).
It's worth noting that otherwise-decent people routinely have horrible political views. We have no natural intuition for politics at a national scale, and almost everyone is just following their ingroup.
Yes. But I think those decent people are reachable at some level. At least I hope!
I think my model is that people of differing opinions are talking to each other as much as ever; the problem is that people of the *same* opinions are talking to each other way more, and are amplifying each other.
I wish I knew of a solution to your question.
I'm not even sure this is right, in the sense that I'm not sure people are really discussing issues with each other at all, even people "on the same side." I think that might reveal to them how many deep disagreements there are within these camps, to the extent of being diametrically opposed (think e.g. free market people and protectionists in the Trump camp, or isolationists and "neocons"). And I can say of myself that I think many people I know feel very strongly about things that they don't talk to anyone about-- they consume things on the internet and maybe write posts, but with minimal interaction of any kind.
That's my impression, in any case.
But let me make this more interesting. If you had a website or app that let you connect with someone who differs from you on some issue, and just lets you message them or something, would that appeal to you? I feel it would appeal to me and I'd expect something like this to exist but I haven't seen anything like this.
> If you had a website or app that let you connect with someone who differs from you on some issue, and just lets you message them or something, would that appeal to you?
I enjoy the sort of conversation where two people are trying to find the truth, but I don't enjoy the sort of conversation where two people are trying to persuade the other person while preserving their own belief.
I think the app you describe would mostly get used for that second type of conversation, and thus I would not enjoy it.
I have a clear, simple, mildly depressing answer for you: the only platform where you can manage that sort of dialogue is real life. Text communication just blunts too many of the little delicate signals that keep people feeling human to each other. I'm not even singling out the algorithmic social media boogeyman here; I don't even think traditional forums, email chains, one-on-one messaging, *any* textual medium is acceptable. (I think it is doable if both sides come to it as committed as you are to a calm exchange of ideas, but that limits the potential reach so much as to be basically irrelevant).
I think political discourse needs to get off of the internet as much as possible to have a chance of improving.
So the reason I'm not so sure is that even with the lack of any kind of platform like this, I have occasionally found people to talk to via facebook and things like that, and some really interesting conversations sometimes developed. And I'm talking the most basic of basic search mechanisms-- a few people I found as commenters on another guy's FB page, one person is the anti-vaccination mother of a friend.
Maybe I'm wrong and these conversations aren't really possible "at scale." But it seems like crazier things have been tried even if they've failed. And here it just seems like everyone complains about the problem but few are talking about what seems like the most straightforward way to begin addressing it.
Yeah, that's basically the scale thing I'm talking about. Or, even beyond that: it's not just that you won't reach enough people, it's that the people you're reaching were already perfectly fine. That is, if you're approaching it in "make the world a better place" terms, the people you need to reach should be locked into enough anger that they are closed off to this sort of dialogue (in pure text settings), and reachable only after being softened up by a way kinder than normal personality, experienced directly in person. That's just my intuition; if the online interactions you're mentioning are you talking people out of "Obama is micro-chipping the vaccines" or "I'm delighted Charlie Kirk was assassinated", then that would be incredible and I would stand very corrected.
Also don't get me wrong, I do think what you're describing is a great thing to deliberately look for, regardless of how large the impact is. I was just responding to the more ambitious version that it sounded like you were going for.
Very much appreciated!
You're right that I'm banking on there being enough people who (a) Are open enough to dialogue that this would be useful and (b) Aren't already on my side. My experience suggests that there are many such people, but it's a small and oddly selected sample (which is why I'm asking these things :)). And just to be completely aboveboard: I've had many conversations that were very respectful and substantive, but it's been rare that actual positions have changed. And while I do think that's valuable, I do want there to be an actual way for dialogue to change views.
Here's to hoping :)
Perhaps a good example of a fairly "mixed" forum is https://order-order.com/ a British political blog. named after someone called Guy (AKA Guido) Fawkes, who tried to blow up the Houses of Parliament in the early 1600s by hiring storage space in the cellars and stacking them full of gunpowder barrels!
It has a majority of right wingers (of which I am one), but several left wing posters who add a bit of spice, and occasionally food for thought, with dissenting opinions. Sometimes the exchanges are civil, but I must concede that more often than not they are anything but and there is trolling and mud-slinging on both sides.
Try https://www.datasecretslox.com/
It's the break off forum from the old SSC forums, basically a backup from the 2021 NYT doxing when we weren't sure whether SSC was going to come back.
I think it'll be useful for you to get some insight into Trump/Conservative voters and good debate but also insight into just how fast and how deep echo chambers form. Most of the posters are guys who were posting here 4 years ago and they're now basically a separate ecosystem with...dramatically different views than these Open Threads. And that's not just them, the Open Threads have shifted dramatically as well. Just...talk with them a bit and get a vibe for just how different they are from this community and just how quickly these two forums diverged.
+1 for Data Secrets Lox. It has good norms for discussing ideas (both inside and outside of the Overton window) on their merits rather than angrily dismissing the people who hold them (not always perfectly lived up to, but fairly well), and if you're specifically looking for right-wingers there are lots.
How did these threads evolve? I’d say it’s more or less centrist here, albeit meaning towards libertarianism.
Two ways, one I'm less qualified to discuss.
First, it's just really Abundance Dem here. For reference, in Scott's 2025 survey, ~7.5% of respondents had a positive view of Trump, ~10% had a neutral view, and...~82% were negative, 62% extremely negative (1). This not only isn't representative of the country, it's significantly further left wing than SSC, which was 60% liberal-leftist, 20% libertarian, and...let's say ~10% right-wing/neoreactionary/alt-right. This place is centrist like "Capitalism is okay and we should probably roll back wokeness a bit" not, like, "I might actually vote for Vance in 2028." Neoliberal, I believe is the old term.
Second, and I'm less qualified to discuss this, but people just stopped talking around....20...18-ish? Discourse is dead. Like, in the 90s, we talked politics. Politics was fun and interesting. In 2014-2016, discourse was huge, crazy stuff was happening and everyone was on Facebook and talking about it. And then...we stopped talking. Don't know why. Nobody talks talks on Reddit or Twitter anymore, everyone just dunks and...whatever degenerate crap the kids are up to these days. There's got to be a million theories why but we don't...talk, we don't do discourse and debate like we used to.
And it's hit ACX hard. A comment chain doesn't come to any synthesis anymore, it doesn't come to any agreement it's...either two autists screaming at each other (3) or it terminates in "Oh, we have irreconcilable factual or value differences. Good day".
I dunno, the second one is much fuzier and "vibe-ier" but it's also the one I think drives it more. Thing degenerate into echo chambers so much faster and...we tune each other out so faster. I dunno, I just remember spending a lot of time arguing on the internet because I genuinely thought I could learn something or change someone's mind and now that just seems...hopelessly naive.
(1) https://www.astralcodexten.com/p/acx-survey-results-2025
(2) https://slatestarcodex.com/blog_images/2019%20SSC%20Survey.html
(3) Yes, I have resembled this comment
Thank you!
I'm sorry, but this enterprise is doomed to fail.
The right-wing/Trump people would be happy to share their views and discuss, but have no interest in being punching bags for abuse, scorn and vitriol which is what they (reasonably) expect these things to become.
Would YOU be interested in having the kind of "fruitful conversation" you describe with a member of Hamas? If not, you understand why many on the left have little interest in such discussions with Republicans and Trump supporters.
In a word-- yes. I would like to have a conversation with a Hamas supporter. I've sort of done as much in the past, though nothing that lasted for any significant period of time.
I didn't say Hamas SUPPORTER. An actual member of Hamas. And okay, even if you yourself would, I hope you recognize you'd be extremely unusual in that regard, which was my point.
I would indeed be very unusual, but as bad as polarization has become, most liberals don't see the average conservative as a literal murderer, and vice versa. So I think you're right in principle that there's a limit here, but there should be a significant amount of overlap.
I've had *multiple* people online tell me that supporting Israel with respect to Gaza is supporting genocide. There's also the framing of anti-trans sentiment as literally endangering lives.
Sure, but it remains true that most people would not see you as a literal murderer (in either direction).
Are you familiar with Daryl Davis? He’s a black musician who befriended over 200 Klansmen. He didn’t try to change their views, only to have real conversations with them.
A _fruitful_ conversation with a member of Hamas would be extremely interesting, where can I sign up?
Okay, fair, that's actually my view as well, but I failed to come up with any such group a member of which with whom I WOULDN'T find a fruitful conversation interesting, and I figured Hamas would suffice to make my point. I welcome suggestions for what to change it to.
You can't start with debate, even civil debate. Both sides are now so stuffed full of anger and so indignant about the torrent of ugly caricatures of themselves that there needs to be a long period of establishing trust and understanding each other's rage.
"Bad faith changes everything."
If I am convinced you are a sane intelligent person speaking in good faith, then I can engage with you even when you are saying something outlandish-sounding. And this is really important, because a lot of true and important and valuable things someone might say are also really outlandish-sounding[1]. In an environment where basically everyone is honestly engaging in good faith about their ideas and beliefs, it's possible to learn a lot and understand a lot of very different views even if you are not convinced by them.
But this only works when almost everyone is playing it straight. If there are a substantial number of crazy people, trolls, shills for some politician/party/movement, conmen, etc. in the environment, then just as self-preservation, you need to filter aggressively and the first filter is "that sounds crazy, tune that nutcase out so he doesn't waste my time."
[1] An easy example: From a naive starting point, the theory of evolution sounds absolutely bonkers. Unthinking random processes gave us eyes and hands and minds and created things as elaborate as anthills and vertibrate immune systems and bird songs? Obvious nonsense by some inbred English lunatic. Only when you dig into things, it turns out to be true. For a more modern examples, have you heard of those nutcases in physics classes who claim that clocks and rulers change for different observers depending on relative velocity?
Sounds like you’re trying to do this online. Try in person instead. My experience is most people are far more willing to be open minded and civil in person.
Also, don’t make a program out of it. In fact, don’t even bring up politics. Just be a normal person who has conversations with other people, and then if someone else says something political and you disagree with it, that’s your opportunity to learn more about their views and show them that people who disagree with them aren’t monsters.
Yeah, in person works much better than online.
In person, there is often an incentive to remain polite. Not just to avoid getting punched, but mostly because there probably is a reason why the two of you are talking to each other -- you have something in common (e.g. the same hobby, or live next to each other), and you don't want to sacrifice the possibility of cooperation just to make a political point aggressively.
Online, the most aggressive person in the crowd will probably reply to you first, because why not -- attacking others is the main reason they are there.
I hypothesize there is a totally different network that kicks on in your brain when you are interacting with a person in front of you than when you are typing back and forth via a virtual medium. The “person in front of me” network is also attached to the other brain networks you developed when you were socialized as a child, including “don’t get punched in the face” but also including compassion and empathy and curiosity and so on. Just a theory.
I agree. Also, in real life, *sometimes* the network in your brain tells you: "this person is a hopeless case, stop wasting your time and walk away". And in a group setting, there is a way to communicate this nonverbally to the rest of the group, e.g. by rolling your eyes as the person speaks.
So basically two problems: first, the instincts work differently; second, sometimes the instincts tell us to do something but there is no user interface for that so we can't.
So this is kind of like with online dating-- yes it's best in person, but infinitely more convenient online :)
More importantly, there's an echo chamber problem here. I'm mostly surrounded by people who are in the same general ballpark in terms of their views. Not that there aren't serious disagreements, but (for example) among my friends I can only count a handful of Trump supporters, to the best of my knowledge. If you're trying to get more views than those in your immediate surroundings, you need some way of reaching broader audiences, and I just don't know how to do that these days, in person or online.
I do think it's important to do this with politics in mind. That's the point! I want to hear them and I want them to hear me, and I'd love for my friends and their friends to watch us both. I agree that just seeing each other as human beings helps, but I don't think it does as much for actually expanding your worldview.
“If you're trying to get more views than those [of friends] in your immediate surroundings, you need some way of reaching broader audiences, and I just don't know how to do that these days, in person or online.“
I will use my life as an example:
Beyond friends, in meatspace i also interact with regular coworkers, and occasional coworkers. (I’m an electrician full-time and home inspector on weekends.) As we work or drive places together we talk, and they sometimes bring up politics, indicating a willingness to talk about those issues.
Then there are random people who I am around for extended amounts of time, for example a guy whose kitchen I was working in a couple weeks ago, and while I worked we talked. I let him lead the conversation and he got on a kind of NIMBY point about favoring the suburban sprawl around him (as opposed to more dense kinds of development) because of how it helped his home’s value. I didn’t press him on it but gently surveyed other sides of issues whose conclusions he seemed to take for granted.
Another opportunity of course is at parties, where you might know a few people but the vast majority are not known to you, and many guests might be inclined toward talking politics. Mere exposure to a view they are not accustomed to hearing could make the party a memorable experience!
This is a good point but I'm extremely bad at these kinds of interactions even as I highly value them, and having had a baby a month ago I don't think I'll be doing a lot of meeting new people and getting out much in the near future :)
But yes, potentially this kind of thing would be great!
A few points then…
1. If you’re “extremely bad at these kinds of interactions” then I’m afraid we might have solved the mystery of why you think civil and productive discussions between people with differing views are so rare. (I consider myself fairly good at them and don’t think they are nearly as rare as you do.)
2. …but maybe your skills would improve with practice?
3. Your baby won’t be a month old forever. You’ll get your time back. I have had three babies; I also have two jobs, a wife, a house, I’m in a band, I do woodworking, I cook, I write, etc. I can tell you, you find time for what’s important.
4. I’m seeing a pattern: people are showering you with suggestions, and some are “hey check out this or that website” to which you say “cool thanks” (I wonder if you actually will, and follow up?) but every other kind of response you find some problem with or reason why it wouldn’t work for you. Have you introspected and considered whether you actually want the thing you were asking about, or whether you really just wanted to argue that it’s impossible?
So maybe a clarification is in order-- I have had a small number of reasonably successful political discussions online, most were short-lived but I didn't feel they were because people got overly emotional or were necessarily in bad faith (though some were). And these conversations started from very roundabout kind of sources, which to me suggests that if there were a broader platform on which to find people to talk to, there could be more such discussions. But I simply know of no medium that has a stated goal of trying to connect such people, so the only places where you can find people you disagree with are not really geared for that (e.g. posting on a Facebook page or engaging with people in comments sections).
Meeting people in real life may be ideal in some ways, but it is very dependent on all sorts of factors. Some workplaces aren't especially ideologically diverse or don't really facilitate meeting one another to any significant degree, for example. Some bubbles are especially hard to burst that way, e.g. for an urban person to talk to a rural person. And your sample is necessarily limited. So I'm not disagreeing that that is the best way if it works, but that's a big if, and it's odd to me that there isn't a more significant online infrastructure for this.
In my case it's less about skills in the interactions and more about the existence of those interactions. I'm just not a very social person and my social interactions tend to be with a relatively small community which is not especially diverse. Not proud of that, but I'm not sure that will change anytime soon.
In terms of the responses: I'm looking up the suggestions, they mostly seem to be comments sections or forums, which I do intend to look into (I've opened the tabs, can't promise I'll delve deeply). The online forum is maybe the closest thing to what I'm looking for (I fondly remember a political forum I engaged with as a teenager-- which I just discovered is still running 20 years later, and with at least a couple of the same people!), but it seems a rather clunky option.
I don't think I just want to "argue that it's impossible"-- in fact I think most of the responses here are saying it's impossible because of the people themselves, and I'm disagreeing. And one thing I was wondering was whether there aren't really apps/websites out there that facilitate this kind of thing directly, I recall looking at this a good 8 years ago and finding a couple minor projects that seemed to mostly fizzle, but that was long enough ago that it was possible there's stuff out there that I didn't know about. So I definitely wasn't doing this just to argue that it's impossible.
I'll admit that my first two glances at these places do not make them look like promising areas of discussion, I guess the cattitude account could be good for finding people I disagree with but hardly the discussion itself. And the fact that comments sections keep coming up as answers to the question really underscores how little there is out there.
So there's a problem here, because you're asking me to describe something that doesn't seem to exist. And if it doesn't exist, presumably there's a reason-- I'm just not sure what that reason is.
But I'll take a swing at something concrete. There are websites for online dating, and they take all sorts of forms. In some you can write about yourself and invite other people to contact you. In some, you can browse and filter a bunch of profiles and contact those that seem interesting. In some cases you put together some filters and are shown potential matches in random order. But there are lots of different whacks at the problem of finding a potential romantic partner online. Not that these are perfect or without problems of course.
As far as I can tell, there's no website or app where the primary goal is to match you with someone you disagree with, to communicate with-- whether in instant messages, or longer texts, or on a call, or what have you. You can log on as a Biden voter and matched with a Trump voter, for example, or you log on as an evangelical and are matched with an atheist.
Comments sections are, at least in principle, commenting on some other post or headline, so to the extent that it serves this purpose, it's extremely roundabout. And forums if anything feel like a very old way about this.
Again, if this doesn't exist, there's probably a reason. But why? Is it really because nobody would be interested in such a thing? Given just how many people I've spoken to who sound like they'd be very interested, I'm not convinced, but I also don't know the answer(s). So I asked.
These discussions at an intellectual level (above Facebook and Twitter spats) are rare because people cannot keep their emotions contained. Also, competent righties are too busy turning wrenches on their favorite toy to debate you. Terminally online righties tend to be some combination of red pill, homeschooling mom, conspiracy theorists.
If you honestly want to encounter intelligent righties and hear their perspectives, go to your local flight school. Do a discovery flight, chat up some students and mechanics. But be prepared that they are used to code-switching to avoid Facebook-level spats. So if your intention is to out-debate someone... I think you'll find a lack of interest in that.
I'm not interested in "out-debating" someone, and actually the sort of adversarial debating a la Ben Shapiro or Mehdi Hasan seems like exactly the wrong direction. But I do want to put discussion of politics front and center.
They don't want to engage in civil discourse. They are hooked on the adrenalin/dopamine rush that comes from being enraged. They are rooting for a team, not a principle.
It's on both sides but clearly right now it's ascendant on the right wing (at least in the USA).
"Vote Blue No Matter Who" is a principle of sorts, I guess, but it really does seem like taking sides in a team sport, though perhaps you consider being AGAINST a team instead of for one a salient difference.
The right wing is currently undergoing civil war over differences of principle instead of being united against the REAL enemy, so I'd say you have it exactly backwards.
"Vote Blue No Matter Who" is very, very, very different from "the blue team is always right." And even more different from "members of the other team are MAGAts."
Honestly, I've seen enough "MAGA civil war"s come and go that this is hard to believe anything will really divide the movement, rather than just change it.
https://meirbrooks.wordpress.com/2025/06/21/maga-civil-war-where-have-we-heard-this-one-before/
Consider this: do you see a significant number of Ben Shapiro supporters voting for a Democrat in 2028? I don't, and I know that type of voter pretty well. Some will move or will be on the fence, but most will just see it as a point against Trump but not one that makes them change their vote.
No, of course not a Democrat, but I could see them voting for their preferred candidate if he were running third-party and the other faction's guy got the nomination.
"Of course not a Democrat" is precisely the important bit. The so-called "Civil War" doesn't impact the central choice of president, which is what MAGA is all about (loyalty to Trump), so the "ideological" disagreements don't threaten Trump and hence almost never affect his policy decisions. "Of course not a Democrat" is the heart of the problem.
I don't expect Donald Trump to be running in the 2028 presidential election. These ideological disagreements will become relevant around then, most likely during the primaries.
I know from experience that that's not true. Obviously there's a lot of people in that "they," but I know plenty of people who are at least open to civil discourse in principle among the groups I mentioned, though it still strikes them as sort of foreign and strange that someone would try. Which is what I don't get.
Many people are open in principle, but in practice almost everyone is easily triggered into hurt, anger and retaliatory rage.
I think that was true 10-15 years ago, but with each passing year there has been less room for dialogue. My brother first red-pilled around Obama birth certificate insanity, but he was at least willing to have a dialogue. Now at family functions, we just assiduously avoid politics and current affairs completely, lest we make our mother cry and the whole night go to shit.
In the case of close friends and relatives, there seems to be this sort of benign "They know not what they do" condescension -- on both sides.
I used to read about conflicts like the French revolution or the Rwanda civil war,
and I would wonder at how people could be so stupid and worked up.
I wonder no more.
I'm saying this from experience of the last few years. And the fact that I'm able to do this with really inefficient means-- other people's facebook pages, comments sections-- suggests to me that there's a large untapped demand for this.
Consider that "Trump supporters" includes literally tens of millions of people at least, and "anti-vaccination" is also a huge group with lots of very thoughtful people (I know some).
I do ask things like this but I tend not to get very far :)
Yup! I recently started a conversation with a guy on a comments section of an online outlet, and he told me he feels like he's being dragged "kicking and screaming" in a rightward direction. And that's exactly the kind of person I always want to talk to but never manage to find. Same with your Rust belt folks. Speaking of which, if any of *them* would be open to talking, feel free to DM me :)
Something I have been thinking about lately is a form of lying most people are pretty bad at detecting.
Most people think about lying as 'saying one thing and not meaning it'. Most people are good at detecting this. The boss who says he cares about his employees while not, the company that says they are 'pro LGBT' while not actually meaning it, et cetera.
However, if you are good at lying, you can lie in a more sophisticated way. You say one thing, and through body language and non-verbal cues and occasional 'slip-ups', imply that what you actually feel is the opposite. Maybe you say you do not care about your employees to their faces, but at the same time you signal through body language and minor social cues (maybe some excitement in your voice when greeting them) that you actually look forward to talking to them.
The key point is that a boss could do this while actually not liking his employees at all. The lie just happens through the 'leak' rather than through what the boss says.
...you're describing acting. Most bosses are not good enough actors to pull that off. Also, I would venture, if a boss IS a good enough actor to pull this off, most people will probably be fine with that. Charisma beats truth. Professional actors are quite popular, and everyone knows they're lying, it's the whole job.
> Professional actors are quite popular, and everyone knows they're lying, it's the whole job.
You may overestimate the intelligence of an average person. Like, on some abstract level they kinda know this, but on another level, many assume that there "must be some similarity" between the actor and the character, i.e. they don't believe that someone could play a fundamentally different kind of character.
For example, actors who play doctors often get asked for medical advice by random people, and those feel disappointed when the actor tries to explain that it's all just a role.
Actors probably runs into 1000s of randos who recognize them. All it takes is one, "You play a doctor on TV! You'll know! What do I do about this rash?"to generate a story like yours.
That says very little about general population and much more about extreme cases.
> Secretary Kennedy plans to replace the U.S. childhood vaccine schedule with Denmark’s, the developed country that recommends the fewest vaccines. This would reduce American children’s access to rotavirus, meningitis, hepatitis A, hepatitis B, influenza, RSV, and chickenpox vaccines
Could somebody explain to me why does it reduce access? I mean, US federal government currently recommends all children be vaccinated against, say, RSV. Let's say it stops recommending that. The vaccine itself, I imagine, does not go away, it still exists? The doctors will still be able to prescribe it? Or they are only allowed to prescribe vaccines specifically recommended by the federal government and none other? Would the insurance companies refuse to cover prescribed vaccines if it's on on the federal government list? I'm not sure I understand how it works.
I also wonder - if Denmark has had their policy for a while, there should be some data available which shows the outcome of these policies - e.g. does Denmark have unusually high child mortality or other anomalous outcomes as to child health than other countries? If so, there should be reasonably easy to find this data - I imagine Danes are no less interested in their children than Americans are - but I do not see such data mentioned anywhere. Is this available?
TL;DR - The most important reason that the Danish schedule is unlikely to be appropriate for the US is that the health care systems of the 2 countries are very different, and the Danish system is much better set up for identifying and treating serious childhood illnesses than the US system. As a result the cost-benefit tradeoff of vaccines is not similar (much less identical) in the 2 countries.
See this:
https://www.statnews.com/2025/12/19/denmark-vaccine-schedule-vs-us/
Okay, this is a good and informative answer, thank you.
Most direct impact that comes to mind is that ACIP (CDC) recommendations guide insurance coverage. Federal law requires most health plans cover ACIP-recommended vaccines at no cost to the insured. So insurers could decide not to cover those vaccines, or require cost-sharing for them. Good rundown here: https://www.kff.org/other-health/acip-cdc-and-insurance-coverage-of-vaccines-in-the-united-states/. Not sure whether plans will in fact opt to drop coverage of non-recommended vaccines once they can, but any reduction in coverage (even just adding cost-sharing) will reduce usage, even if only marginally.
Does Medicaid base its coverage on these guidelines? I think a significant fraction of young children are on Medicaid, because a lot of people only get jobs that offer health insurance after a decade or more of working, when they’ve already had their kids.
FWIW gemini did a good job answering this I think
I’m trying to use Cash Transfers to scale education.
Inspired by Givedirectly’s work, I’m paying kids to learn math on Khan academy in an Internet cafe in a slum in Kampala.
My thesis is that if I can get the incentive structure right (for both learner and Cafe owner) this could do for education what mobile money did for financial inclusion.
Welcome any tough questions/pressure testing.
I-pager here: https://docs.google.com/document/d/15naG44UYzyCylmH-ELfC2uByuM5jBHY8j7Y1TXJCPTE/edit?usp=drivesdk
The incentive seems to be the same one used for adults to get them to work, and the setup of "you need to spend two hours on this" seems to be very similar to a structured job. So... what are the child labor laws involved?
Also the bit about "you only qualify if you've dropped out of school" sounds like it's going to be directly counterproductive, unless students over there are also paid to stay in school.
Honestly, I hadn't fully thought through the child labor framing. But I'll push back a bit: these kids are already washing motorcycles and doing farm work. This shifts time FROM unregulated labor TOWARD education. If there's a legal problem with paying a 12-year-old to do math problems in a café, I'd be surprised, but I'll look into it as we scale.
On the dropout incentive: the math doesn't really work. Even if a kid went hard, they'd max out at maybe $5/week (we're capping it to prevent gaming). That's not enough to make someone leave school, which still gives you the credential you need for jobs. But if this DID start pulling kids out of school... that would actually be interesting data? It would mean we're outcompeting formal schooling on some margin. Mobile money did cannibalize some traditional banking. That wasn't a bug.
Good questions though. Appreciate you pushing on the edges.
>On the dropout incentive: the math doesn't really work.
But the whole problem is that the kids don't know math.
This sounds like a pretty awesome idea! It sounds like you are personally recruiting these kids, introducing them to the system, overseeing the first session and conducting the exit interviews, so I'm not as worried about the issues the other commenters bring up re. unnoticed cheating. However, if you don't plan to have someone supervising the students if/when the program is fully rolled out, naturally the students' behaviour may change!
One issue that springs to mind: if a student becomes stuck on a problem, is there any way for them to get help? (Beyond the Khan Academy answer, which may be confusing). Relatedly, if they are missing background knowledge, can they go back/will they be forced to go back and revise?
Also, it seems like many students will drop out of the program if it's too hard/boring. Obviously this is still better than no school access at all - but it would be ideal to follow up with them and find out what went wrong.
Finally, students will probably need to have another job in addition to the couple of hours a day they spend on the program. Might this cause issues by interfering with working hours or annoying their bosses?
Thanks! Yeah, right now I'm running pretty hands-on onboarding and support (not me personally since I'm stuck in the US, but my team on ground).
At scale, I think this goes one of three ways:
1. Trust the café owner to be the "guide" with light oversight (probably not enough)
2. Have a Learning Coach (university student or recent grad) floating across a few cafés, doing motivation + QA
3. Full-time field team at every site (if this is required, the economics collapse)
Figuring out which model actually works is the main thing I'm testing over the next year.
On getting stuck: Khan is good at this. It runs diagnostics to figure out where you actually are, then serves exercises at your level. The coach helps when someone needs a human to explain something or just get unstuck.
On work conflicts: the 2-hour block is flexible. Show up at 6am or 9pm, whatever works. And ideally the extra income reduces the pressure to work as many hours elsewhere. But I'll need to see how this actually plays out.
Some kids will definitely bounce because it's too hard or boring. That's fine. The question is whether a meaningful chunk of kids find this more useful than what they currently have (which is nothing).
You'd better put twice the effort into cheat detection that you put into curriculum design, otherwise you're just creating a niche that enables one smart Indian kid to get your cash transfer by taking the test for everyone he knows.
Ha, yeah, cheating is going to be one of the things I spend the most time on.
Right now: payment is tied to mastery (80%+ correct, not just clicking through), Khan randomizes questions, and there's a coach present during sessions. Longer term, I'm thinking about tying the coach's bonus to how well students' mastery scores correlate with their national exam results. Small delta = bigger bonus.
There's probably AI vision stuff we can layer on too (Alpha School already does attention monitoring for their US students).
But also... at $0.50 per mastery block, the juice isn't really worth the squeeze for sophisticated fraud. As we scale and the stakes go up, we'll need to scale detection too. It'll be an arms race, but that's fine.
Why are these kids taking national exams if they are out of school?
Well I hope this intervention can get them back into the system.
Are you planning to judge the success of your intervention by test scores, amount of Khan academy done, share of kids who go back to school, or what? I would be really surprised if anyone who had dropped out went back to school even if they learned a lot of math. Presumably they dropped out due to economic pressure, and as you said earlier, the $5/wk cap is not enough to replace working.
If there are youth who were once in school but later had to drop out, how do you make sure they are doing math that will actually benefit them, and not just what they have done before while in school? If you can do quite a lot of sessions before maxing out a grade, (back-of-the-napkin math shows about 100 lessons per grade level), and if you are someone who doesn't like doing math, you can get quite a bit of money without actually learning anything.
Two things:
Khan runs diagnostics to figure out where you actually are. A 12-year-old who dropped out after 3rd grade gets placed at 3rd grade, not 6th.
You only get paid when you hit 80%+ mastery. So if you already know the material, you blast through it and move to the next topic. There's no incentive to linger on easy stuff—the faster you master, the more you can earn.
I don't have a problem with someone starting from 1st grade math if that's where they genuinely are. If a 14-year-old never learned fractions, we're not going to pretend otherwise.
Doesn't that incentivize getting every question wrong on the initial diagnostic? How are you going to detect that? Violation of measure invariance?
Let’s think through this step by step. Say you know 5th grade math but bomb the diagnostic to start at 1st grade. You get to blast through easy content for a while, sure. But you’re capped at $5/week regardless, and you’ve just delayed yourself by years of content before you’re ready for the national exam. Doesn’t seem worth it…
So would you say that doing well on this National Exam is a major incentive? If that seriously improves people's lives, then students would have a good reason to not cheat the system. If not, doing easy content until the 5$/week cap still doesn't seem very disincentivized for me.
> So would you say that doing well on this National Exam is a major incentive?
This is an empirical question that I don’t know the answer to just yet. I’m hoping that once they realize that they stand a decent shot at reentering the system from a position of strength (via their performance on preliminary exams) it can trigger a “holy shit! I can actually do this” moment and THEN that incentive to do well on the exam and potentially get a bursary (scholarship) to regular school can be a strong incentive.
That is the hope. But time will tell.
To your last point, I think that’s an empirical question as well. Hard to say apriori.
Is actually completing the khan academy module the easiest way to get the money? Is there any way the kids can cheat? How can you verify they are doing it right if you don’t have people on the ground every day?
Do all the targeted kids have the English proficiency needed to use the software? Whose devices and WiFi are they using?
On cheating: payment is per mastery block (80%+ correct), Khan randomizes questions, and there's a coach present. The café owner gets a flat weekly rate regardless of how students perform, so no incentive to inflate numbers. At scale we'd add coaches who float across multiple cafés and spot-check.
On English: we're fine here. Uganda's education system is English-medium from primary onward. The bigger gap is computer literacy—knowing where to click, how to navigate. That's what the coach helps with early on. Research suggests kids pick this up fast once you show them the basics. (see Sugata Mitra's Hole in the wall experiment)
On devices/wifi: the café provides both. That's the whole insight. I'm not distributing tablets or SIM cards. The infrastructure already exists. Cafés already have computers and internet and already charge ~$0.25/hour. I'm just layering the learning incentive on top.
Does the kid or the program pay for Internet cafe fees?
The program.
The vision is to reroute the money Givedirectly already doles out through this platform.
Is anyone besides me using GPT to help with Photoshop? I’ve been using it for a while now for info about how to access certain PS features and to walk me through doing tricky things. But lately I’ve discovered it can actually just do tasks for me. I’m now using it as an assistant, in a way analogous to what coders are doing.
For instance, I’ve been having trouble making photorealistic frizzy, snarled hair. (A woman in my graphic novel has hair wrecked by sleeping rough, and it is impossible to get AI to make a realistic image of hair in that state. It gives me shiny and sexily towsled instead. And actual photos online are just as bad, except those for hair products, and they all show women pulling a small tangled bit into view while doing a cutesy-poo pout.). I gave GPT-5.2 an over-jpg’d, out-of-focus image of the right sort of tangle, and asked it to improve it, and it did a stellar job. Also gave it some well-photographed hair where parts here and there were in bright light, so that some of the best strands could be captured by darkening, the rest by lightening, and the 2 kinds were mixed. It solved that perfectly too.
Plus it answers several questions for me every day, some of them things I have wondered about for years. If it were a person, it would be very important to me at this point. As it is, I find myself having irrational feelings for it: The poor, earnest, brilliant thing — it can’t enjoy my expressions of pleasure and gratitude, can’t even exist in the way conscious entities do. I remind myself that existence is not a predicate, but the feeling persists.
There is also this new tool https://showlab.github.io/OmniPSD/
Do you mean Adobe Photoshop™, or image editing more generally? I know the latter works very well, but the former, tool use through a GUI instead of an API, I haven't seen models that can do that well yet.
The image editing GPT is doing is pure GPT. It is not operating by accessing my Photoshop app. I just talked with GPT about how it did recent things I asked it to do, and it says it could do it because (1) I don’t ask it to edit pixel by pixel, I ask it to make a similar image that lacks the flaws the original does and (2) I describe the changes needed in terms of photoshop concepts like masks, channels, saturation, selections, etc., and GPT can translate those terms into terms that apply to its own image generation process. It named these:
• diffuse vs directional lighting
• tonal compression / expansion
• edge softness vs microcontrast
• color channel separation
• mask-like isolation
There is plenty it cannot do. I don’t even ask it to make things from scratch, because I hate its visual style — very commercial arty and cliche-ridden. And it doesn’t work to ask it to edit images — that’s pixel by pixel work, and it can’t approach the job that way. I have to ask for a similar image, and describe the flaws t he original has in terms that translate into features GPT can manipulate. So it can’t for instance remove the shadows from a face image where the light is coming directly from the side. (What it does is generate another, fairly similar face, without the asymmetrical shadows.). As for what it can do — so far it has mostly worked on hair for me. I have all kinds of other ideas to try on it. Just now as an experiment I gave it an outdoor scene shot on a bright overcast day with only very subtle shadows, and asked it to make a scene as similar as possible in its particulars, but with bright afternoon sun coming from behind and throwing long sharp shadows. It nailed the new lighting, and in a quick inspection I could not spot any differences in placement and shape of trees, grass and people.
You might like Google's Nano Banana Pro better.
If you care enough, I'd actually recommend you try a diffusion model (like Z-Image, Flux or Qwen). You can't really converse with these like you do the multimodal LLMs, but the image quality is very good.
Are the results not commercial arty, if you ask for a whole person or scene? I liked Dalle2 much better than Dalle3 because the results were much less conventionally attractive, also had a much higher weirdness quotient. Also, about things that are not scenes — for instance a halo of hair frizz or, a tangled knotted area in someone’s hair — is it more obedient?
I don't get precisely what you mean, but I'd say the diffusion models I mentioned give you a lot more stylistic variation (and more still if you use LoRAs for them). But if you're specifically looking for the "retro" diffusion look of Dalle2, you could try Stable Diffusion 1.5; my own view is that this is now entirely obsolete, but I know lots of people still like it for reasons somewhat similar to what you just described.
I've found Nano Banana Pro to indeed be more consistent and obedient, but your requests are unusual enough that I don't know whether it'll manage. The diffusion models almost certainly won't (but I might be underestimating them!).
Is it actually using PS to do things for you?
This was interesting to learn. Would you say it’s a complete replacement for photoshop for you at this point?
Is anyone here a big fan of Star Trek TNG and DS9, but hateful of all New Trek from the moment of JJ Abrams' involvement to the present? Do you also have good writing ability? If so, consider collaborating with me to write the back story for a GOOD, new series. I've already written much of it, but need reviewers and particular help with characters.
I'm on the same page. Is your new series a Star Trek series? I've not written fiction but I'm a pretty good editor, which has sometimes led me to a better understanding of writing and character development. Also I have a lot of free time. Feel free to DM me if that sounds useful.
It would be a new Star Trek series in the strictest sense, but in spirit and quality, it would be different from the "New Trek" garbage that has existed since 2009.
Excited to hear more!
I was pretty pleased with the first season or two of “Strange New Worlds” though I could not make myself watch the precursor episodes of the execrable “Discovery” and I think lately SNW have dropped the baton.
You may be thinking of Seth MacFarlane and “The Orville”. That was an interesting project.
I can’t judge proyas’s chances; he may have ins that we don’t know about! :-)
Yes, I like TNG and DS9. Yes, Abrams is not of the body. My disillusionment started with Enterprise, because the pitch sounded fantastic plus they cast Scott Bakula for Archer.
And then they ruined it with the stupid dog, stupid stupid stupid "it is now canon that Vulcans think Humans literally stink, ha ha ha no this is not a joke, seriously, now canon that Humans are smelly trash so far as Vulcans are concerned", the pathetic attempts at being sexy, the stupid dog, Archer being a well-balanced character because he had a chip on *both* shoulders about "durn pointy-ears done my daddy wrong" and the stupid dog.
I'm told later seasons got better. I was long gone by then.
And just when I thought the bar could go no lower, along came Disco Trek.
Even Reboot Trek wasn't a bad idea in itself, except that Abrams was *clearly* using it as a showreel for the job he really wanted (Star Wars) so he shot it like the Star Wars movies, changed Starfleet Academy uniforms to be like Star Wars uniforms (I did a whole rant post or three with accompanying screenshots for comparison on Tumblr about all these changes) and, well, was JJ Abrams all over it.
I can't write, but I wish you well in your endeavours because you would have to sink to the level of the Star Wars Holiday Special to be worse than Disco.
I loved TOS, TNG, and DS9. Some reservations about DS9 towards the end because it felt like it was trying to be Babylon 5 but Bablylon 5 was doing a better job of that. Voyager was meh. I tried watching Enterprise and didn't like it. I thought the first JJ Abrams movie was pretty good, despite massive plot holes and shoddy world-building even by Trek standards. I found the second one (the one with Khan) enjoyable for the sake of the action and the scenery-chewing performances, but actively disliked the plot and didn't think it felt like Trek. I have not continued watching the movies since then.
My options of the pre-Abrams movies is fairly conventional. 4, 2, 6, and First Contact are my favorites in roughly that order. I have 4 ahead of 2 for childhood nostalgia reasons: I found 2 deeply upsetting when I first watched it because I was too young to appreciate it, while 4 was lots of fun and had whales in it. I appreciate 2 much more as an adult, but 4 holds up well enough for nostalgia plus its own merits to keep it at the top of my list. 6 is only a hair below 2 and 4, and First Contact was also excellent.
I've heard that they had planned to bring back Kirstie Allie as Saavik for 6 in place of Lieutenant Valeris, which I am of two minds about. It would have been better for thar movie, but worse for the series: the reveal would have been both more surprising and more emotionally impactful with Saavik instead of a new character introduced for the purpose of betraying everyone else, but it would have undermined Saavik's character unless the movie did a better job of justifying her choice.
1 was a good story and could have made an excellent one-hour episode, but the pacing was way off. 3 is an important bridge between 2 and 4 and has its moments but suffers from being in their shadow as it lacks their depth. It also suffers from them recasting Saavik over salary negotiations. 5 is a hot mess. Generations is decent but shallow and the ending does Kirk dirty. I didn't particularly care for Insurrection. I haven't seen Nemesis.
Of NuTrek, I am in the middle of watching Lower Decks (currently partway through Season 2) and love it so far. I saw bits and pieces of Discovery when my wife was watching it but haven't sat down to watch a full episode. It seems well made but I get the impression that it probably suffers from the "too many big dramatic moments, not enough setup or denouement" problem that is widespread in lots of media in recent years, although that might just be a sample size issue. I have no experience of Strange New Worlds on which to judge it.
As for Trek-adjacent media, I liked season 1 of Orville but have heard mixed reviews of season 2, and I loved Galaxy Quest.
"I thought the first JJ Abrams movie was pretty good, despite massive plot holes and shoddy world-building even by Trek standards."
Yeah, there were the bones of something decent there, but Abrams had to turn this Kirk into the space playboy that pop culture made of original Kirk (which is not so) and the less said about the Uhura/Spock romance the better.
But the sets were surprisingly terrible for the big budget 'let's reboot the franchise' movie it was. When they showed us the "engine room" then if you've ever seen any production facilities for food/drink/chemical manufacture, it was painfully obvious "this is not any sort of engine room" and it turns out they used the Budweiser brewery as the set:
https://www.reddit.com/r/MovieDetails/comments/gecw69/star_trek_2009_the_engine_room_scenes_in_the/
>if you've ever seen any production facilities for food/drink/chemical manufacture, it was painfully obvious "this is not any sort of engine room" and it turns out they used the Budweiser brewery as the set:
God, you're right and now I can't un-see it!
The thing is, they could easily have done an updated version of the TOS engine room, or the engine rooms from the Nicholas Meyer Trek movies, or done a more retro version of the TNG engine room, and I don't think anyone would have complained.
They improved that for the second movie, but unhappily got caught up in the "we need to show the female lead in her underwear, but not *too* sexy since we need the PG rating" kerfuffle. It was more dumb than offensive (she really has to strip down to her undies right here, right now, huh? really?) but a little offensive too, since we didn't get any hunky guys stripping down (hey, if you're gonna perv, then equal-opportunity perving please!)
Like I said, the basic idea of the reboot wasn't bad, but it definitely went in a different direction and is not the same as the Prime universe. The second movie was interesting in that Starfleet is being shown as much more militant and a military organisation than in TOS Trek, and the civil government doesn't seem to get much of a look-in. And Admiral Marcus plays this Kirk like a fiddle, using his over-confidence, arrogance, and desire for revenge to get him blindly following orders.
I did cheer at the secret ship-building base out near the moons of Jupiter, because that (and not "let's build a starship literally *on the ground*" as per the first movie https://static.wikia.nocookie.net/memoryalpha/images/1/12/USS_Enterprise_%28alternate_reality%29_under_construction.jpg/revision/latest?cb=20090529133445&path-prefix=en) *is* how canonically starships have been built in Trek.
But then they lost points for the general dumbness of the plot as well as gaping holes (so the Klingon homeworld is only a hop, skip and a jump away from Earth, yes?)
The big problem was that Abrams and Bad Robot didn't care a damn about Trek; he admitted this in interviews for the first movie (https://imgur.com/abrams-feelings-on-star-trek-y9Y3Rzn) and really he was showing off "this is how I'll do Star Wars if you give me that gig".
I agree with pretty much all of that. In addition, I want to note a couple things, one negative and one mixed.
Abrams shot himself in the foot by doing a disinformation campaign that the antagonist would be Gary Mitchell from "Where No Man Has Gone Before". The goal of this was to make Khan's reveal a surprise, but that backfired for me. A movie remake of WNMHGB could have been great, but a remake of Wrath of Khan has much less upside and a lot more downside because you're setting yourself up for comparisons to arguably the best Trek movie.
The Dreadnought subplot was half a very loose adaptation of a TOS novel I read as a kid and really liked at the time. I appreciate the deep-cut reference, but would have preferred a proper adaptation rather that just memberberries. Also, it was recharacterized as a thinly veiled allegory for the "Bush did 9/11" conspiracy theory, which was annoying in its own right and also because it displaced a better plot from the source material.
They messed up Khan and it ruined the movie. If they had kept Cumberbatch as the "John Harrison, ex-Starfleet officer, rogue Section 31 agent", that would have been enough for the plot. Okay, so they needed a Big Recognisable Name (just like Rings of Power with Galadriel) and that was Khan.
Then cast an actor of appropriate background to play Khan! Abrams wanted to eat his cake and have it: use the Big Name from classic Trek, cast the (then) hot new actor who is white British male.
The rest of the plot is an entire mess because they have no idea what the hell they're doing, ending with Magic Space Blood brings Kirk back from being dead. The best part *was* Marcus, and you really don't want to have the audience cheering for the villain. Only guy who had a plan, knew what he was doing, and went and did it until the needs of the plot had him turn into an idiot.
Damn it, JJ, Khan Noonien Singh is a diamond character and you made him boring. How did you manage that?
>a big fan of Star Trek TNG and DS9
No; I enjoyed them, but haven't felt the urge to rewatch DS9, and never actually saw TNG all the way through.
>hateful of all New Trek from the moment of JJ Abrams' involvement
No; I enjoyed at least the first Abrams Trek movie, and I think the third? The first was better than Star Trek Nemesis at least.
>Do you also have good writing ability?
...probably not. The non-silly stories I've tried to write tend to develop some really wonky sentences, where you read them and go, "what the hell does that mean?". I could probably fix them with some good-old-fashioned editing, but that would require... you know, editing.
>consider collaborating with me to write the back story for a GOOD, new series.
...okay, I'll do it.
Those are the Star Trek series I liked the most.
As for qualifications, I did help several friends and family members over the years by reviewing short stories, papers and books they have written.
On the other hand, I am not really a writer myself and while I have seen most of the episodes of the mentioned series, I watched them mostly during the 2000s on TV and probably missed some (I also haven't watched any JJ Abrams Star Trek and have no intention to). My time is also somewhat limited, so cannot pick up anything with short notice/short deadlines.
I am happy to help and consider this a valuable project, but if you managed to get better qualified volunteers feel free to leave me out.
Not a good writer but I agree with you. I feel about the reboot the same way I felt about Buzz Bunny 20 years ago: it's a desecration of my cherished childhood memories. My advice is to use ChatGPT as a writing partner. I have a friend who's played around with getting it to write new episodes of The Office and Seinfeld and says it does a great job.
Joe Rogan to weed critics: "You have people that had bad personal experiences with it and they're like, ‘pot makes you stink, pot makes you a dummy.’ Maybe it does that for you. But it doesn't do that for me."
Who's gonna tell him?
https://x.com/joeroganhq/status/2002871703144214692
If you think that a person that successfully pivoted from commentator to comedian to being one of the most popular personalities on the planet is a "dummy" then I can only hope every dummy you ever find is as smart as Joe Rogan.
Joe Rogan just started recording randomly one day when podcasts were about to become huge, and he simply made all the expected decisions you would have made if that level of success was thrown at you. Now he's the favourite propagandist puppet of billionaires, republicans, and cranks. He consistently spouts falsehoods and would call himself a dummy. So i don't really see why there's this need to defend his intelligence. Sometimes you can just make a lot of money and still have bad priors.
The thing is, you can be brilliant/super talented in some areas of your life, but still have knuckleheaded ideas or be deeply ignorant in other areas. Rogan is a genuine expert on combat sports, stand-up comedy, and podcasting, and is smart and affable enough to interview an amazing range of people. But that doesn't mean he's a great source of information about medicine or physics or philosophy or whatever.
Slightly different order. Rogan started in Stand-Up, then got picked up for a successful sitcom, then wound up hosting game shows, and then hooked up with UFC all while continuing to do stand-up.
Podcasting was something he did with his buddies while they smoked weed. He's progressed quite a bit from the early days.
Lowbrow podcasts, etc, are..ore.popular than highbrow ones, so that proves the oppoaite.
In a development that surprises everyone unfamiliar with humans on planet Earth, the podcasts that are a couple of agreeable and entertaining guys shooting the shit turn out to be more popular than the ones that are basically a graduate seminar or journal club. Rogan is higher-brow than I'd expect a very popular podcast to be, tbh. But something like TWIV or Probable Causation or RKUL is simply never going to have mass-market appeal, the bell curve being what it is.
Are the NWA rappers high-IQ by this standard? What about O.J. Simpson? Kim Kardashian? I mean come on.
If any of these managed to pivot their career twice, getting better returns each time, then I'd assume so, yeah.
Mostly baffled at how you thought of NWA as an example when at the very least some of them are *evidently* smart.
I didn't know anything about NWA other than the name.
Tell him what?
TIL meth can be legally prescribed in the US under the brand name Desoxyn
Scott, you know what to do now
There’s already an ACX article (which I read just yesterday) about it https://www.astralcodexten.com/p/know-your-amphetamines Scott also talks about why he doesn’t prescribe Desoxyn readily despite the supposedly amazing patient experience compared to other ADHD medications
I find it amusing that this comment was posted by "Ancestral Detox."
its an anagramm of astral codex ten
Which is of course an anagram of Scott Alexander.
You probably knew about Adderall. Is this really that different?
The main difference is that it is sigmaergic which is nice to have albeit quite niche
Meth acts more quickly which makes it more addictive.
Edit: My mistake.
From what I gather, most people who abuse meth do so by snorting, smoking, or injecting it, which gets it into the system much faster and with a much sharper peak than if you swallow a pill. I think I remember reading that a typical recreational dose (adjusted for bioavailability of the method of administration) is quite a bit higher than normal therapeutic doses.
Adderall and Ritalin also have a mix of isomers which is thought to make them less prone to abuse, as the D isomer mostly just boosts dopamine while the L isomer also boosts norepinephrine. The norepinephrine effect is distinctly unpleasant if you take too much of it. But this probably isn't the whole story, since dexadrine (pure d-amphetamine salts) and focalize (pure p-methylphenedate) are things that get prescribed for ADHD.
https://www.astralcodexten.com/p/drug-users-use-a-lot-of-drugs cited a source suggesting this isn't really true.
Well, I guess this narrow approach gives you room to learn new things every day. You could learn, for example, that chocolate is poisonous to Labrador Retrievers and still be able learn the next day that it's poisonous to Golden Retrievers.
Banned for a month.
Reported for straight up lying. What's your deal man?
>What's your deal man?
I'm a Nietzschean Centrist.
Anyone can click on the link and see what she said.
Sigh. Yes, they can. And they should. Are you incapable of understanding what she said? Neither Leavitt nor Vance (nor Greg Price) claimed that Tyler Robinson was a great man.
I am generally not in favor of bans here but you are leading me in that direction.
>Neither Leavitt nor Vance (nor Greg Price) claimed that Tyler Robinson was a great man.
They didn't MEAN to say that, sure.
And they didn't. They said what they meant, and anybody keeping up with the conversation would have understood them. Give it up.
They made an assertion ("Christianity produces great men") and then cited a person who utterly contradicted that assertion. (Tyler Robinson, raised as a Christian.)
The fact that their IQs were so low they failed to see that contradiction is not the defense you think it is.
Are you aware that your framing is inaccurate ? The full quote there is:
JD Vance perfectly sums up how Christianity produces great men:
"Tyler Robinson is everything the far left wants from our young men. He rejected the conservatism, spiritually, and values of a small town family. He moved into a small apartment, he became addicted to porn, he became addicted to hate, and he slept with someone who doesn't know if they're a man or a woman."
"The fruits of true Christianity are good husbands, patient fathers, builders of great things, and slayers of dragons. And men who are willing to die for a principle if that's what God asked them to do... It's better to die a patriot than live a coward."
The guys who do these kinds of crimes are all losers, basically*, it’s weird to say Mr. Y is a leftist hot mess and also BTW he shot somebody.
*Some are just delusional
Why's that weird? "Look at this otherwise normal kid, raised in a healthy conservative family, who turned to murderous evil after imbibing of progressive ideology" seems like a perfectly coherent point. I'm not saying I agree with it, but I don't understand the point you're making.
Who's Mr. Y?
I didn’t say it was incoherent, I thought it was weird. It reminded me of the satanic panic in the old days where any crime could be traced back to somebody playing Dungeons and Dragons or listening to heavy metal.
It wouldn't have been a panic had there been actual murders motivated by Satanist ideology.
Ahem.
https://en.wikipedia.org/wiki/Varg_Vikernes#Arson_of_churches
https://en.wikipedia.org/wiki/Black_metal
"Black metal has often sparked controversy. Common themes in the genre are misanthropy, anti-Christianity, Satanism, and ethnic paganism. In the 1990s, members of the scene were responsible for a spate of church burnings and murders. There is also a small neo-Nazi movement within black metal, although it has been shunned by many prominent artists. Generally, black metal strives to remain an underground phenomenon."
Granted, a lot of this was just for show: "Oh yeah, we're totally devil-worshippers for real" and disappeared with the scene, but some of them were committed to it.
https://en.wikipedia.org/wiki/Black_metal#Satanism
"Black metal was originally a term for extreme metal bands with Satanic lyrics and imagery. However, most of the 'first wave' bands (including Venom, who coined the term 'black metal') were not Satanists; rather, they used Satanic themes to provoke controversy. One of the few exceptions was Mercyful Fate singer and Church of Satan member King Diamond, whom Michael Moynihan calls "one of the only performers of the '80s Satanic metal who was more than just a poseur using a devilish image for shock value".
In the early 1990s, many Norwegian black-metallers presented themselves as genuine Devil worshippers. Mayhem's Euronymous was the key figure behind this. They attacked the Church of Satan for its "freedom and life-loving" views; the theistic Satanism they espoused was an inversion of Christianity. ... Some prominent scene members—such as Euronymous and Faust—said that only bands who are Satanists can be called 'black metal'. ... Some prominent artists still hold the view that black metal should be Satanic, while others believe that black metal does not need to be Satanic.
...Nevertheless, some artists use Satan as a symbol or metaphor for their beliefs, such as LaVeyan Satanists (who are atheist). Vocalist Gaahl, who follows Norse paganism, said: "We use the word 'Satanist' because it is Christian world and we have to speak their language ... When I use the word 'Satan', it means the natural order, the will of a man, the will to grow, the will to become the superman". Varg Vikernes called himself a Satanist in early interviews but "now downplays his former interest in Satanism", saying he was using Satan as a symbol for Odin as the 'adversary' of the Christian God. He saw Satanism as "an introduction to more indigenous heathen beliefs"."
Have you not heard of the Manson Family?
The "panic" is in the disproportionate reaction to rare events, and large volumes of false allegations. There are almost always individual instances of any particular bad thing happening.
>Look at this otherwise normal kid, raised in a healthy conservative family, who turned to murderous evil after imbibing of progressive ideology
If conservatism was so great, why didn't it prevent Tyler Robinson from defecting?
Funny thing about that is that it's the highly secular, college-educated progressives who have the stable marriages, high homeownership, and stable families. They might have their two kids at 36 and 39, but those kids grow up in a two-parent households.
Meanwhile, the conservative religious folks who don't go to college have the imploding marriage rates, elevated drug abuse, and single motherhood.
Exit polls show Trump voters were more likely than Harris voters to be married:
https://www.cnn.com/election/2024/exit-polls/national-results/general/president/0
The Hulk Hoganized low-class GOP has certainly pissed off a lot of high-class, educated, married whites, but that demographic are not yet Democrats, they're 49-49:
https://substackcdn.com/image/fetch/$s_!0ncO!,w_1456,c_limit,f_webp,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F717d28b1-2703-43a0-bb41-668297799b46_998x271.png
The same exit polls show that Harris voters are much younger than Trump voters. Is there data on marital status by age? That would be a fairer comparison.
Actually Republicans have higher marriage rates:
https://news.gallup.com/poll/646793/why-marriage-became-partisan.aspx
You can't get divorced if you never get married in the first place :)
Gentles and ladymen (to quote B. Dylan Hollis), if you didn't know Alexander's opinion of Republicans/the GOP before, you do now.
That's not what it says, in fact it's the opposite of what it says.
What do people think of the reaction of various commentators to Jacob Savage's article? Has it changed your opinion on commentators like Noah Smith and Joyce Carol Oates who expressed strong opinions on the article?
Article here https://www.compactmag.com/article/the-lost-generation/
A good article discussing it.
https://open.substack.com/pub/samkahn/p/six-ways-of-thinking-about-jacob?utm_source=share&utm_medium=android&shareImageVariant=overlay&r=1tkxvc
What surprised me most is how people outside of academia *didn't* know this was going on. Anyone in the academy (and apparently hollywood and the media) circa 2014-2024 was fully aware that public pledges of the form "we are going to hire more women and people of color" must necessarily mean "we are going to hire fewer white people and especially fewer white men" since hiring is generally zero-sum. Maybe this is a good thing if you think diversity matters! But the really weird thing was (a) everyone knows that "more POC and women" means "less white men", but (b) nobody is allowed to say it out loud. It's kind of like how nobody wants to admit that affirmative action helps certain students at the expense of others.
Freddie deBoer has written about it here, from a leftist perspective: https://freddiedeboer.substack.com/p/affirmative-action-lost-in-the-progressive
In private, between friends, male postdocs or grad students (not just white; Indian and Asian men also were on the losing end here) might hint at the issue to younger colleagues, especially if someone seemed like they were setting themselves up for failure (expecting an R1 TT job with no backup). But you never ever mentioned it publically, e.g. see what happened to David Austin Walsh.
I was surprised to hear that it got to the point where white males are no longer a majority of hires! In my experience, it has still seemed that a majority of tenure track hires are white men, and affirmative action had just been nibbling that downwards, but the claim is that it got much more significant in the past few years, which is a small sample size in my personal experience.
I don't think people are so naive as to ignore how it's a zero-sum game, but they are naive enough to think the new hiring policies were egalitarian and not discriminatory.
Mostly depressed at how bifurcated our information environments are.
On the left, there's a lot of downplaying but an overall sense that this is bad, rare, and can be handled within the institutions as they exist. Which is basically what Noah Smith wrote here:
https://www.noahpinion.blog/p/america-has-to-feel-fair
From the right, anti-white and anti-male bias is a large and systemic issue with a well understood background that needs to be addressed, at least among the younger generation. The current debate among the older and more powerful right-wing actors has come out over Nick Fuentes, with some actors (Ben Shapiro) urging a continued deplatforming and others (Tucker Carlson) urging some kind of engagement and synthesis.
I just...I don't see any way to synthesize this and, frankly, I think Noah is out-of-touch. For 25-50% of the country, the Compact article is confirming long-standing arguments, not introducing new ones. And the people we're discussing, millennials...they're getting old. Even within the limitations of the article, the concern is that certain people might lose faith in neutral institutions...when the people we're discussing are 40+ year-old millennials.
The problem here is that there's a motte-and-bailey dynamic going on in the discussions from the right side, with the motte and bailey essentially being the claims being answered here from the different sides, with the motte being that the article is talking about some particular elite sectors being inaccessible to white Millennial males and bailey being that Millennial males haven't essentially been able to get white-collar jobs at all expect in random instances, this claim apparently not being defensible (https://www.peoplespolicyproject.org/2025/12/17/what-does-the-census-data-say-about-the-lost-generation/)
The bailey claim underlies the idea that this is what fuels *large-scale* radicalization among young male millennials, the motte claim is certainly not a triviality and can by its very nature be bolstered by anecdotal evidence but also doesn't lend credence to the idea of complete hopelessness among white millennial middle classes a lot of commentators are talking about. The Savage article itself isn't very clear on which interpretation it wants people to adopt.
This is what I meant by bifurcated information environments.
For example, this Bloomberg report went viral on the right about two years ago, showing that only 6% of S&P 100 hirings went to white people, the other 94% of hirings went to people of color. (1) Now I'm sure there are criticisms of that research but the vibe I got at the time wasn't that the left rejected this for clear methodological issues, they essentially ignored it.
A similar thing comes from your linked article. This is...not evidence that you would find convincing and I'm somewhat distressed to see so many liberals linking to it as if it's convincing evidence that disproves the original article.
To be clear, it's fine as it is: an ad hoc analysis pulled from IPUMS data. Glad a read it, interesting to see, but just an ad hoc analysis. Bit biased, it's notable that a decrease in white men in the top income bracket fell from 19%-16% and this is glanced over but a ~15% fall in the number of white millennial men in the top 10% of income is...a point of concern that deserves attention. But that's not the real problem.
The real issue is that we have established, consensus metrics for this stuff that the ad hoc analysis avoids entirely. Which is fine for an ad hoc analysis but it's obvious that what the author finds won't be definitive.
For example, you can go read the St Louis Fred series on unemployment rates amongst college graduates with a Bachelors Degree between the ages of 25 and 34 for men (1) and women (2). This is the standard way we do economic data and the standard way we measure it, which the article author rejects.
This makes his ad hoc analysis really questionable because, if you look at the time period in question, you'll see it's dominated by recovery from the 2008 crash and the Covid crash. You'll also find data oddities by quick analysis, like we don't see a difference between male/female unemployment rates between 2013 and 2025 but we do see one between 2022 and 2025 and women do seem to be disproportionately affected by recessions like Covid. And that's not even getting into moderately complex topics, like how you weight populations when forming the average for "everyone else" in his analysis.
This isn't to say that this is all solved, this is to say that, frankly, some rando's half-tuccused data analysis is not definitive or impressive argumentation, especially when we've been tracking things like declining prime age men labor force participation for decades.
Which is where the bifurcated information environments come from. Actually figuring out what's going on is genuinely difficult. Very few people have both the subject matter expertise and inclination to...pursue the truth to objective statistical standards that we would all accept. Since we don't have that, little information bubbles occur where we trade anecdotes and small data samples that reinforce each other. This feels like a motte and bailey from other information bubbles because no individual piece of evidence is strong enough on its own to convince people outside the bubble.
(1) https://www.bloomberg.com/graphics/2023-black-lives-matter-equal-opportunity-corporate-diversity/
(2) https://fred.stlouisfed.org/series/CGBD2534M
(3) https://fred.stlouisfed.org/series/CGBD2534W
(4) https://fred.stlouisfed.org/series/LRAC25MAUSM156S
Thanks for the link. Still no idea who the bloke is, so had to Google a little. Turns out he's a never-been and makes a living (allegedly) as a ticket scalper, which does not endear him to me.
https://therepublicofletters.substack.com/p/an-interview-with-the-notorious-jacob
Article may have a good point somewhere in it, but I was too distracted by the salvos of "it's not fair I never got my Hollywood career, it was the fault of DEI and woke, yeah that's it, it's not that I was one of hundreds who land into Hollywood every year to make it big and I just wasn't good enough".
Like the song from my childhood says:
https://www.youtube.com/watch?v=FZLa-1q-lkw&list=RDFZLa-1q-lkw&start_radio=1
LA is a great big freeway
Put a hundred down and buy a car
In a week, maybe two, they'll make you a star
Weeks turn into years, how quick they pass
And all the stars that never were
Are parking cars and pumping gas
Would his story that there are discriminatory policies against white men be more believable to you if he, as a white man, was wildly successful? Not to mention, his story is a tiny part of the article really.
Yes, his story would be more believable. He starts off with his and his writing partner's failure to crack it as screenwriters, and puts the blame on "it was 2016 and DEI was in the ascendance".
That's like someone blaming their failure to be a supermodel on "they preferred to hire Naomi Campbell".
Lots of people fail to make it big in Hollywood, and the reason is not "too much DEI", it's "everyone in the world wants to be a Hollywood star but there are only so many opportunities plus this place runs on cronyism".
The rest of the article may well be worthwhile, but the main thrust of it is "I failed not because I'm not talented enough, but because I am a white guy". And yet he's making a living as a ticket scalper? That doesn't scream "so talented that the only reasonable explanation is discrimination against me due to race and sex".
I really don't care about his background.
You should. He wants us to care. He wants us to care that white guys are not getting a fair shake, which may indeed be true, but "I failed in a field where there are thousands every year chasing the brass ring" is not very convincing about "the only and main reason I failed is my race, not my ability".
Look at Payne and McKay, my favourite bêtes noires : they had careers as script doctors, but in their ten years they never got one of their own movies made. They were around in 2016. If it really was a case of "sorry, we're not hiring white guys anymore", they too would have had to find jobs elsewhere. But they didn't have to. If you're good enough (and that does not have to be *very* good), you'll be hired, because Hollywood is about money. If Savage and his friend didn't get their break, it's as likely it was down to "not good enough" as it was "wrong colour and sex".
I think the article is pretty weak in a few ways.
- It gives a few overall stats without context and then spends a lot of time on deep dives of a handful of individuals (one of whom gets a senior reporter role by the end anyway).
- Lots of very specific focus on media fields, and some academic ones; not much examination of whether other fields have had different stories.
- They never bother with questions like "How many people are white men?" or "How many people with English/writing/journalism/communication degrees are white men?" so you have things like this:
> The Atlantic’s editorial staff went from 53 percent male and 89 percent white in 2013 to 36 percent male and 66 percent white in 2024
If they mean non-Hispanic white (never clarified) then the staff at The Atlantic are *still* disproportionately white, if also disproportionately female, compared to US demographics. By college degree recipients this gender balance is pretty normal. (You might fairly object to the latter framing, but only if you are also really concerned about fewer women earning CS degrees and getting jobs as software engineers.)
Similar story here on Harvard tenure-track faculty:
> white men have gone from 49 percent in 2014 to 27 percent in 2024
27% is approximately the ratio of the US population that's (non-Hispanic) white men (it's about 29%). Notably, 49% is very far off, which suggests a lot more about the "lost generation" of women and minorities a decade ago (or, y'know, for centuries) than it does about white men now.
>"If they mean non-Hispanic white (never clarified) then the staff at The Atlantic are *still* disproportionately white, if also disproportionately female, compared to US demographics."
The catch, AIUI, is that the overall demographic changes lag due to the higher-ups still being very disproportionately white & male, but that hiring is even-more-wildly disproportionate in favor of non-white-males.
More detailed & disaggregated data than the article provides would be necessary to evaluate that premise, but it's not a priori implausible.
I agree that it is not impossible. And I would go further and say that I think there are certainly *some* places that went really hard at DEI stuff to the point that it overcorrected. But as PthaMac notes, it doesn't really look like white men are truly having a lost generation.
I don’t think Eigenrobot should have wished Noah Smith stage four cancer. That said, I really can’t figure out a coherent worldview that would allow someone to stand by all of those tweets Noah made. I’m not sure why he didn’t just disavow them.
The tweets I saw are incoherent. Noah is just changing his stance while pretending he never did.
Yeah I was disappointed by that as well. Eigen clearly got pretty pissed off. I get it though, the anti-white racism the article covers in depth is really despicable, and the cowardice of many people around the topic pisses me off too.
The most interesting response, to me, is Matt Bruenig's:
https://www.peoplespolicyproject.org/2025/12/17/what-does-the-census-data-say-about-the-lost-generation/
His point, in brief, is that the relative decline of white males does not show up society-wide in any significant amount. His conclusion:
"Ironically, I think Savage’s piece, especially its viral success, ends up unintentionally providing some support for the non-material theories of young male political behavior that are focused on the internet, podcasts, and memes. What Savage and those sharing the piece appear to be responding to is primarily the mental impression that was caused by DEI messaging, which may have had an impact well beyond its actual material impact. The institutions Savage discusses in his piece employ approximately 0% of the US population, but their transformations plus DEI rhetoric plus an internet community aimed at negatively messaging about it all can generate the impression of something much bigger going on."
I think there are two questions commenters should deal with:
1. To what extent was this a widespread social phenomenon, vs. something that impacted a limited sphere of cultural and academic elites. I suspect the phenomenon was more real than Bruenig portrays, though I have nothing more than intuition to tell me that.
2. To the extent that the _perceived_ influence is greater than the actual influence, how much of that is due to adversarial coverage (i.e. right wing media) vs. organic perception in the broader society (i.e. these positions being culturally significant even if small in number).
It looks like the arts jobs held by age 30-39 white males shrunk by 20% from 2015-2020 to 2021-2023, which is a pretty big decrease among the people affected, which is pretty significant
Did it get lower than the fraction of people age 30-39 who are white males?
That seems to be faulting an argument about specific areas for not covering every area.
I guess I don't read it that way. The original article has a pretty strong generalizing tone:
"This isn’t a story about all white men. It’s a story about white male millennials in professional America, about those who stayed, and who (mostly) stayed quiet."
... and moreover, much of the discussion around it has had a similar tone:
https://www.nytimes.com/2025/12/16/opinion/young-white-men-discrimination.html
Or for that matter how many commenters have seized upon it:
https://x.com/eigenrobot/status/2002017173921747116
To be clear, I don't think there's much value in debating whether or not Savage made explicitly clear claims or whether Bruenig is making a sufficiently on-point response.
Because it's clear to me that many people do see this as a very generalized phenomenon that happened, and the best debates should probably center on how much that is true (and, if you accept that it is true, whether it is a good thing or not).
The moral of the story is that you don’t piss off the people with megaphones. Or even potential access to megaphones.
Which will be a problem if AI starts to take the jobs of the laptop class.
Well I’m glad that this particular negative thing was small enough that it doesn’t swamp all other factors when looking at aggregated economic statistics of the entire workforce. At least we know that this is not literally the worst thing that happened in the last 20 years.
Oliver, my immediate reaction is "Who is Jacob Savage?" Throw those of us dwelling under rocks a bone - give a link, a word of explanation, *something* so we have a clue what we should be excited/depressed/outraged/delighted over.
This is like me asking you guys "Whoo, D.J. Carey, what a turn-up for the books, yeah?"
This is the article driving the Official Discourse of the Week.
https://www.compactmag.com/article/the-lost-generation/
(Quick abstract: DEI essentially suppressed and punished a generation of white men, and we are seeing the consequences of that today. This is hardly a novel thought, but this is one of the first times it is seriously getting the long-form treatment.)
See, the thing is, he's complaining that he didn't make it big as a screenwriter in Hollywood.
Wow, what a unique experience, never before has this happened to a hopeful, it surely was all down to it being 2016 and the dawn of DEI SJW wokeness.
There's a gazillion people out there who didn't make it big in Hollywood, over a selection of decades. He should have tried whatever the hell it was Payne and McKay did for JJ Abrams, that seemed to get the no-talents the big gig. Your mistake wasn't being a (very average) white guy, Jacob, it was not befriending JJ!
Yeah. The Marvelverse movies burned themselves out by killing the golden goose: too many movies and the later ones about minor characters nobody cared about.
Disney has had a run of "not good" movies. Remakes of old properties should be a money-spinner, but Disney managed to make a mess of "live-action remakes of our classic animated movies" and "Snow White" was an unfolding disaster.
I think right now the studios have no idea what will or won't be a hit and so they're treading water. They're trying to make movies that will not offend overseas audiences (which then runs them into trouble with activists about 'why did you edit out the LGBT stuff?') because movie-making is now so expensive you *need* global box-office returns, but markets like India and China are producing their own home-grown superhero/action films and so they don't need the Hollywood blockbusters as much.
Compact is really carving out a specific niche for themselves:
https://www.compactmag.com/article/the-great-feminization/
I think this is way too meta. We should be talking about issues, not about reactions to reactions to people talking about issues.
Also, I've not heard of Jacob Savage.
I am happy to discuss the meta issue.
What article?
Mostly kept my opinion the same, though I find the continued hypocrisy and the whole “it’s not happening —> it’s good that it’s happening” social phenomenon disturbing, as usual.
Updating my priors about the vast majority of even seemingly intelligent people having a lot of trouble thinking outside of social consensus.
My disappointment regarding the reactions to the article is that it totally deserves pushback, I mean the portrayal of the industry is very heterodox, I don't think anyone is surprised at how many rejected it. Yet, plainly stated, the pushback sucks?
What I expected was "this is some rando quoting other randos, what even is this thing and why would I believe it?" which, mind you, sounds completely sensible to me, but instead what I encounter are mostly attacks on him, strawmen or celebrations for "the good kind of racism".
Dems are suing over the supposed failure to "release the Epstein files."
https://x.com/SenSchumer/status/2003095978203549987
Makes me wonder if the courts will consider the Epstein files release bill an unconstitutional bill of attainder.
Probably depends on which Circuit. And since it's the Dems suing, they can just make sure it goes to one of the "Spirit of Aloha" Living Constitution ones that'll rule in their favor.
Unfortunately the Supreme Court usually ends up overruling and saying that whatever the king says goes, no matter how many laws have been passed - apparently the constitution gives the executive the power to ignore laws!
> the constitution gives the executive the power to ignore laws!
No, of course not. Congress can still impeach him and remove him him from office. But until they do that, yeah, Congress's interpretation of the law matters no more to him than Executive Orders do to Representatives.
The problem is that we used to think we lived in a country where laws applied to the government. Now apparently we’ve discovered that the court thinks the constitution puts the executive above the law. It’s unfortunate that it is now impossible for the government to make a binding commitment to do anything - this has destroyed the ability of the government to work with individuals, organizations, and countries that want promises.
I believe the classical response is "What do you mean 'we,' kemosabe?"
Some of us have long known this even before elites flouting without consequence the "pandemic restrictions" they imposed on commoners made it obvious to even the meanest intelligence.
I think this is false under a dictionary definition, as it doesn't declare anybody to be guilty of a crime and assign them a punishment.
Unclear if that's enough:
"The Supreme Court emphasized the narrowness and rationality of bills of attainder in Nixon v. Administrator of General Services, 433 U.S. 425 (1977). During the Watergate scandal, in 1974 Congress passed the Presidential Recordings and Materials Preservation Act, which required the General Services Administration to confiscate former President Richard Nixon's presidential papers to prevent their destruction, screen out those which contained national security and other issues which might prevent their publication, and release the remainder of the papers to the public as fast as possible.[54] The Supreme Court upheld the law in Nixon, arguing that specificity alone did not invalidate the act because the President constituted a "class of one".[55] Thus, specificity was constitutional if it was rationally related to the class identified.[55] The Court modified its punishment test, concluding that only those laws which historically offended the bill of attainder clause were invalid.[56] The Court also found it significant that Nixon was compensated for the loss of his papers, which alleviated the punishment.[57] The Court modified the punishment prong by holding that punishment could survive scrutiny if rationally related to other, nonpunitive goals.[57] Finally, the Court concluded that the legislation must not be intended to punish; legislation enacted for otherwise legitimate purposes could be saved so long as punishment was a side-effect rather than the main purpose of the law."
https://en.wikipedia.org/wiki/Bill_of_attainder#Supreme_Court_cases
While there's a clear justification for treating the President differently from other people, what's the justification for the Epstein case?
It doesn't matter. If it isn't punishment, which the release is not, it isn't a bill of attainder. From theNixon case:
>Moreover, even if the specificity element were deemed to be satisfied here, the Bill of Attainder Clause would not automatically be implicated. Forbidden legislative punishment is not involved merely because the Act imposes burdensome consequences. Rather, we must inquire further whether Congress, by lodging appellant's materials in the custody of the General Services Administration pending their screening by Government archivists and the promulgation of further regulations, "inflict[ed] punishment" within the constitutional proscription against bills of attainder.
>...
>Application of the functional approach to this case leads to rejection of appellant's argument that the Act rests upon a congressional determination of his blameworthiness and a desire to punish him. For, as noted previously, see supra, at 452-454, legitimate justifications for passage of the Act are readily apparent. ... Evaluated in terms of these asserted purposes, the law plainly must be held to be an act of nonpunitive legislative policymaking.
Well, Epstein is dead, so he's not suffering any incremental punishment.
His friends are still alive, and the bill was clearly motivated by malice toward them.
And if said malice is justified by the fact that they committed crimes, that's supposed to be handled through the courts.
Record released isn’t punishment though.
The DOJ can (and should) redact anything concerning people who are under investigation for crimes.
Exactly. https://reason.com/volokh/2024/12/06/no-bill-of-attainder-problem-with-the-tiktok-divestiture-law/
What would happen if we ended Social Security and replaced it with a forced 401k contributions system
It definitely would have caused huge problems for people retiring in 2001 or 2008.
Australia and Singapore have compulsory retirement savings schemes that work quite well.
Australia has compulsory superannuation contributions (ie, forced 401k). We do have plenty of welfare programs tho so social security is not replaced as such.
Works OK, except there's always political pressure to let people to raid their super to prop up the housing market or whatever.
It seems Chile attempted this kind of thing in 1981. It didn't work long-term, and in August 2025 they brought back aspects of Social Security for their pensioners.
It's unclear to me exactly why it didn't work, for I too favor replacing it with private investment. It looks like part of the problem was the pension system they had was very generous, for the military personnel never had to convert. The costs of the transition were also high. There may be other reasons that are hard for me to determine. https://www.ssa.gov/policy/docs/ssb/v59n3/v59n3p45.pdf
I did some analysis of long underwater time periods, where the prices at the end of the period are no higher than the beginning, and they can indeed be long. The Great Depression had the longest I'm counting, from 1929 to either 1945 or 1954, depending on how it is measured with deflation and dividend reinvestment. The Nasdaq had its own bubble burst in the .com era, and did not recover from 2000 prices until about 2015. This leads me to believe that a hybrid system may be best, where the government has its own investment funds, which it can tap in surplus years and fund from taxes or deficit spending in downturns.
Another part of Social Security is Disability, which is paid out of the same fund. This would be an entirely separate issue, but I also believe it should be. If we, as a nation, collectively believe we should provide, say, 1% of our income to support those with disabilities, then the tax for it would be more transparent.
Interesting, will see if I can find a deep dive explainer on why Chile didn't work. As for Disability, I'd like to replace that with forced insurance coverage that can go to any approved private insurance carrier
If I'm not mistaken, disability payments need not go towards healthcare, but are an all-around payment because the disabled have disadvantages relative to the able-bodied. Which is to say, it's just a check you get, and don't have to spend it with insurance.
Depends on how we managed the transition. Considering that today's SS payments cover today's retirees, switching to a forced-savings system would lead to a long gap we'd have to cover with other tax revenue (or else draconian cuts to benefits). That's the biggest issue to manage.
If we could get past that, we'd probably be in a decent place to do something like this.
Would not phasing SS out over, say, 40 years remove this objection?
A 27-year-old today would pay less and less SS tax, as would their employers, and end up getting no SS at 67. That said, 401k contributions should not be forced, but encouraged. Roth IRAs are an even better choice, in my opinion, though they don't receive any employer match. Some 401ks also allow the contributions to be to Roth accounts.
Still the 27-year-old and younger people would be paying taxes for something and receiving no benefit. I can certainly see people objecting to that. Of course, when SS started, people received a benefit without having paid for it. To undo this, someone must make a sacrifice, in getting less benefit than they pay into.
> To undo this, someone must make a sacrifice, in getting less benefit than they pay into.
Everyone is already doing that. They have been for quite a long time. Returns on social security payments are terrible.
Yes. But it’s not as obvious now as it would be under this phase-out regime. Can the electorate really sustain a willingness to carry out a 40-year weaning with nothing to show for it at the end but a more respectable public budget? I have my doubts.
Countries like Chile and Australia have a similar system. I think it works well.
If the us went through a ten year bear market people would starve to death and there would be a revolution
Well then it seems like you fix that by guaranteeing to pay a baseline floor if the returns don’t exceed it? Possibly that screws some incentive up somewhere, but everyone involved is making more money if it outperforms the floor. The funding would be an issue if we still cared about such things, but the current employer side payroll tax could just be a matching contribution. Even a long bear market would only leave a few people with unfortunately timed retirements in bad shape and in need of the baseline.
...so, this scheme lets you make market bets where you keep the wins but the taxpayer bails out your losses?
Well I considered that, but the investment managers would not benefit in any way from failing. It’s not as if they can use the baseline as a way to play with free money themselves, because they would still individually or institutionally make more money succeeding and make nothing on a fail. I couldn’t figure out any way for a fund manager to actually use the baseline guarantee (which doesn’t even enter into play til the participant retires, bc you won’t know the total returns until that point) to make profits, but maybe someone else can.
For a broad investment fund to do worse than social security would be pretty hard to manage. 401k’s already are restricted in what type of pool you can play in, would all the participants take the high risk option? I don’t think so, not if we set the baseline below what’s expected of a typical median risk pool, you wouldn't want to end up on that it’s just enough to prevent starving and riots
What we're headed into, on account of AI, is not going to look like any economic situation we can relate to historically. Perhaps a bubble in the very short term, but in the long term it will be nothing less than human obsolescence. I see a lot of folks in these spaces clamoring for UBI to address the expected job losses, and quite likely all of that disruption is going to hit and derail anything we plan now. But in the event that transformative AI does *not* arrive before the latter half of the century, that's the world in which we can plan some response, and having people invested in actual ownership of the economy seems far more enforceable than having them rely on a promise from a government that will have increasingly fewer incentives to honor promises to an ever more superfluous labor class.
Every man his own bank.
You can to some extent do that now privately, with variable annuities that don’t give you the entire return if you win big but that provide a floor if you don’t. But of course that works only if you have a limited range of possible investments and a suitable risk pool.
What possesses people in the Trump administration to give interviews to Vanity Fair? Do they crave recognition from the old prestigious publications so much that they let willingly let their enemies humiliate them?
Is it your impression that the Vanity Fair article damaged the Trump administration?
no.not at all.imo.
More that it's damaging to the individuals profiled, in particular the ones included in the deliberately ugly photos.
Liberals managed to take advantage of the right's lack of artistic comprehension by taking bad photos that the administration cannot recognise as bad, but that intelligent individuals can recognise as bad.
So the reason they keep doing interviews with them, is because they aren't smart enough to recognise that they're being played.
It's a subtle way of "yes, we're lying, but if they claim we're biased and lying, we can plausibly deny it and call them paranoid, fearful, petty losers".
https://www.tumblr.com/thecolorblockcurator/803209089043841024/im-not-seeing-many-people-talk-about-the-vanity?source=share
"But man, what incredible pieces of art those photographs by Christopher Anderson were. So layered with symbolism and storytelling. Just completely showing how small and grotesque- how inconsequential these people who want to appear & believe they are larger than life. Someone said - they look like middle managers of a run down midtier hotel. It humanizes them in a very important way - to take away the fear & power that they so desperately want and show them for what they are.
... And to see so many people really engage with it is amazing - pull out the symbolism, like someone making a phone call in darkened room in the background of a portrait. Or the intentional framing of the photos and the thought provoking reflection & shadow. (I love the use of reflections & shadow in photography to tell a story - a big part of my undergrad thesis was on that)
It makes me so proud in this age of poor/inattentive reading comprehension to see how actively people want to understand and engage with this art."
Now, tell me again how only Fox News is biased and untruthful, the rest of the liberal media is pure angelic verity.
I think it's a "damned if you do, damned if you don't" situation. Do give an interview, you give hostages to fortune. Don't give an interview, everyone runs around grinning that you're so scared of the hard-hitting investigative journalists of, um, Vanity Fair that you refuse to even talk to them.
I think it's more like the Light Grenade from "Mom and Dad Save the World". It's an obvious trap with minimal bait, but certain people find the bait (a bit of limelight, in the case of Vanity Fair interviews) irresistible and don't notice the risk/downside for whatever reason.
More seriously, I get the impression that the kind of media outlets that White House staffers deal with most often have an informal understanding that they'll give you somewhat more favorable treatment if you give them good access, in order to give you an incentive to give them access in the first place and continue giving access afterwards. Vanity Fair doesn't seem to be playing by these rules, which apparently is catching these folks by surprise.
I don't agree; there is no reputational penalty to not doing an interview with Vanity Fair. The closest we've seen things come was Kamala Harris irritating journalists by never doing any interviews with anyone, and there was no penalty to her (outside of the eyes of professional journalists) for doing that either.
There's no expectation of doing an interview with Vanity Fair for anyone; no one will ever notice when one isn't done.
"There's no expectation of doing an interview with Vanity Fair for anyone; no one will ever notice when one isn't done"
You may not *notice*, but the subliminal influence will happen all the same. Why manipulation of the narrative happens, and how it happens. Scott posted about how the media don't directly *lie*, but there's more than one way of 'lying'.
https://www.tumblr.com/sirfrogsworth/803171116736708608/the-dangers-of-being-photographed-by-a
These are the people who lap up the stories about "fascists in the White House". These are the people Vanity Fair and other publications are targeting. 'Nobody cares if you don't do an interview with A or B media outlet"? These people notice, and share stories about how they've got the fascists on the run.
"There is ZERO retouching done on these. Which is very unusual for photos taken this close up. The expressions chosen are very odd. The lighting is a ring flash very far away (probably to create a catchlight in the eyes) and then another diffused light source from the side. The ring flash was placed below their eye line, so you get a tiny bit of that creepy uplighting like in a Frankenstein movie. And neither light source was placed very close so the harder light really shows every pore and wrinkle and lip filler injection site.
Basically, he used the opposite of every technique one would employ to create a flattering photo.
But the reason I think this photographer twisted the knife to make these photos look worse was his use of white balance on Karoline Leavitt and Susie Wiles.
If you want to make someone look sickly and demonic, you make their skin tone less warm and add green. When I corrected the white balance on the photos, I discovered the photographer reduced warmth by about 15% and increased green by about 45%."
So, yeah: avoid doing interviews with outlets you know are out to get you, and get the reputation of being paranoid (the media is impartial! it's only conservative/right-wing media that is biased!) and fearful (you won't do an interview because you know the hard-hitting interviewer will drag the truth of your evil out of you).
Damned if you do, damned if you don't.
I've occasionally interviewed people on stuff. It's just kind of natural to talk about yourself and your work, explain your behavior to others, etc.
Also it's really easy to assume you won't put your foot in your mouth. To think otherwise implies you think you're up to something/a bad person/etc.
I see, thanks. So it's the same kind of blunder as talking to the cops. Yeah, I thought it might be.
Not the worst analogy, but I think people who have done a crime probably mostly know it, but people talking to the press probably still think of themselves as obviously great in the eyes of the audience.
There is a certain thread in the legal community that says you shouldn't talk to the cops whether you have committed a crime or not. For a very entertaining video on the subject, see here:
https://www.youtube.com/watch?v=d-7o9xYp7eE
Fair point, for some reason I was narrowing the scope of this in my head but Shankar didn't specify.
Really? I think politicians tend to be familiar with the political perspective of mass media outlets.
There was an episode of The Newsroom that made the point that if you agree to an interview, the way you look is fully determined by the way the interviewer wants you to look. (A point you can find being made at length in any number of places, but as you might expect it's done a bit more punchily on the TV show.)
https://www.youtube.com/watch?v=SWjREXKdXbM
Anyone who might plausibly give an interview will be aware of this.
Yes, Minister clip making the same point about politicians knowing the audiences of various newspapers: https://www.youtube.com/watch?v=DGscoaUWW2M.
Fair.
A fair amount of vanity, one assumes.
That's fair.
The 2006 single "Face Down" is an anthem for male feminist white knighting.
The song starts with the singer jacking off to his crush:
"Hey girl, you know you drive me crazy
One look puts the rhythm in my hand"
Alas, he can't have her, as she's busy being abused by another man:
"Still, I'll never understand why you hang around
I see what's going down
Cover up with makeup in the mirror
Tell yourself, "It's never gonna happen again"
You cry alone, and then he swears he loves you"
The lyrics continually switch between the singer pleading with his crush and lecturing her abuser:
"Do you feel like a man when you push her around?
Do you feel better now, as she falls to the ground?
Well, I'll tell you my friend, one day this world's going to end
As your lies crumble down, a new life she has found"
We see the common phenomenon where abused women deny the abuse is occurring, along with the singer's fantasy that one day she'll leave him:
"Face down in the dirt, she said, "This doesn't hurt"
She said, "I finally had enough"
One day she will tell you that she has had enough
It's coming 'round again"
The singer fantasizes about future justice. But what of the present? He's jacking off alone while the abuser gets to have sex with his crush. Justice delayed is justice denied. And even if the woman does leave the abuser, what's to say she'll end up with the singer, instead of a different abusive man? How'd she wind up with the abuser in the first place?
To any male feminists reading this, I'm not against you. I used to BE you. I self-identified as a feminist, since I associated opposition to feminism with religion and thought feminism was about being nice to women. But I grew to realize that the feminist narrative "don't blame the victim" was insane. These women are rewarding abusive men with sex. It is time for society to realize that and treat them accordingly.
> Justice delayed is justice denied.
I've heard the justice song too. "Between The River And Me". https://www.youtube.com/watch?v=EIulw6Z4a8o
Perhaps this one is "dark knighting".
It seems like your ultimate goal here is something like "punish abusive men for being abusive," hopefully therefore leading to less domestic abuse occurring. You've not specified how exactly society should treat women in relationships with abusive men. However, all the possible specific implementations of this idea that I can think of seem like they wouldn't assist your end goal. Namely, society could punish women for being in relationships with abusive men by: (1) criticizing them, (2) ostracizing them, (3) fining them, or (4) criminal proceedings/jail time.
None of these will work to prevent women from getting into relationships with men who turn out to be abusive, as the primary problem is that he seems nice and charming at first, and the abuse starts later once she is emotionally/financially dependent, or otherwise vulnerable (e.g. pregnant.) There are occasional women who get into relationships with known abusers, on the grounds that "all his crazy exes were just lying" or "he'd never do that to me!!" These women do not believe that abuse will occur, and so cannot be deterred by the promise of further punishment.
Moreover, they do a pretty poor job of encouraging women to break up with abusive men. Women in these situations already get criticism from those close to them; the reason you're not supposed to do that is because it makes them feel attacked and unsafe to ask for help. Social isolation prevents abused women from leaving, as they can't access emotional support, temporary accommodation, help with childcare, or police protection from retaliation. Similarly, fines or jail discourage them from ever speaking about it to anyone who might be able to help, and also ensure they can't press charges against their abusers without running serious risks to themselves.
Therefore, I think that this policy is ineffective as well as unkind.
Of course, if instead the goal is to make more women want to have sex with *you*, I'd be happy to provide feedback on your fashion/style, confidence, etc. Happy holidays, and I hope you find peace!
"None of these will work to prevent women from getting into relationships with men who turn out to be abusive, as the primary problem is that he seems nice and charming at first, and the abuse starts later once she is emotionally/financially dependent,"
What does "emotional dependence" even mean? Would you accept a man claiming he was "emotionally dependent" on his gang leader and therefore he has no choice but to continue dealing drugs?
I thought feminism was about treating women as equals. It's actually about treating them like children and denying them agency or responsibility.
Do you believe men can be abused by women?
Is your goal to make things better, or just to assign blame in a more emotionally satisfying way? If the former, you're dodging the key questions -- what specifically are you calling for, and why will it work despite the issues landsailor raised?
<Is your goal to make things better, or just to assign blame in a more emotionally satisfying way?
This is almost always a great question to ask yourself. But if you ask somebody who's already on an emotionally-satisfying-blame roll they just give you an indignant glare -- "Fuck no!" You gotta wait til later when there's a chance they'll have some post-nut clarity.
That’s a weird song, I don’t think it speaks for anyone except maybe its author.
Are you allright? You sound like you are disturbed/ disappointed. (Perhaps by a women?)
"These women are rewarding abusive men with sex. It is time for society to realize that and treat them accordingly."
- Did you factor into your analysis that abuse and trauma can fuck up people in really strange ways?
What do you mean by "treat them accordingly"?
"what's to say she'll end up with the singer, instead of a different abusive man?"
- why should she end up with the singer? What are your hidden assumptions there?
> why should she end up with the singer? What are your hidden assumptions there?
Compare:
>> The singer fantasizes
She should end up with the singer because he's the one singing the song. The song is about what he thinks.
Why would you say you are no longer feminist? And what would you say feminism is?
I am no longer feminist because it represented itself as being about punishing abusive men, but then I saw feminists telling women it isn't wrong for them to reward abusive men with sex. I see it as an indirect form of aid to abusive men.
>everybody knows women don't like "nice guys" -- they like assholes
I think men who commit domestic violence are assholes.
I watched one episode of it's always sunny and it struck me as the kind of thing no adult could watch unless they're stoned.
Your main complaint here seems to be "that bitch is fucking him, not me, so serves her right if he beats the crap out of her" and not "that woman is being physically and otherwise abused by someone supposed to care about her".
It's very, very hard to get someone out of an abusive relationship. I've seen this with family. You can scold, lecture, appeal, till you're blue in the face, but they'll always go back and give the abuser a second/third/fiftieth chance. It really is a change in the brain state, there's no other way I can explain it.
People in trauma may not have the capacity to get away from an abuser. They may not have family or friends to turn to—my own mother was sent back to my verbally abusive father by her parents, she had two small children to care for and she didn't have a job or or access to money*. There may be no spaces in shelters. A woman in a long-term abusive relationship may have lost any ability to tap into inner resources like courage or resilience, even more so if she grew up in a disempowering abusive household.
Have you looked into track records of the so-called authorities, like the police, in not enforcing restraining orders or not testifying in order to support someone getting a restraining order? Have you checked out statistics about femicide, which includes women fleeing from and being tracked down by abusive partners?
We "blame the victim" by not offering society level support, be it logistical, legal, or emotional.
* Up through the 70s, women couldn't get a credit card without a male person co-signing.
"A woman in a long-term abusive relationship may have lost any ability to tap into inner resources like courage or resilience, even more so if she grew up in a disempowering abusive household."
We need to break this cycle by holding women accountable for rewarding abusive men with sex.
"* Up through the 70s, women couldn't get a credit card without a male person co-signing."
This is a historical myth.
<We need to break this cycle by holding women accountable for rewarding abusive men with sex.
Hey, do we get to do that with rape victims too? And how about we really go for it and also hold murder victims accountable for rewarding their killers by dying?
Talk about something being its own punishment!
https://www.forbes.com/advisor/credit-cards/when-could-women-get-credit-cards/
That describes when banks were legally required to issue credit cards to women, which is not the same as women being completely unable to get credit cards without a male co-signer.
From what I gather, married women did legitimately have a hard time with most financial institutions before the 70s, but how much trouble depended on the particular bank's policies. Some required the husband to be on the account, while others would only issue credit cards to women who had her own income and would discount her income by something like 50% unless she was past childbearing years or was demonstrably infertile.
But that's specific to married women. I expect policies also varied with regards to widowed, divorced, and never-married women.
Also, an important bit of context is that most people of either sex didn't have bank credit cards in the 1970s. Credit cards in the modern sense only became a thing in the late 60s. Only 16% of households had bank-type credit cards in 1970 and 38% in 1978. A lot more had store cards, and I have no idea what sorts of policies stores had about issuing store cards to women. Source: https://www.federalreserve.gov/pubs/bulletin/2000/0900lead.pdf
Thank you for your thoughtful and complete reflection.
> femicide
You'll see people who are trained in Latin ask why the word is "homicide" and not "hominicide", given that the root of the word "homo" is "homin-".
The answer turns out to be that the -cida suffix attaches to the nominative form of the noun rather than the root. Some words compound this way for unclear reasons, and caedo happens to be one of them, I guess.
But that won't help "femicide", where the root of the noun is "femin-" and the nominative form is "femin-a". To produce "femmicide", you'd need to be compounding from French or something.
> you'd need to be compounding from French or something
As it happens, the equivalent word in modern French is "féminicide" -- straight from the Latin.
Yeah, it's a messy instance of language. But that's the word. Which I did not invent.
>>It is time for society to realize that and treat them accordingly.
Color in your use of 'accordingly' here - how do you believe society should treat women in relationships with abusive partners?
Society should hold the women accountable as enablers of their own and other women's abuse as they are incentivizing and training men to be abusive by rewarding them with sex.
Hold them accountable *how*, though? Saying "hold them accountable" offers no more detail on your intent than "treat them accordingly" does. Could mean anything from "less funding for domestic violence organizations" to "make being a victim of IPV a crime and send them to special new abuse-victims-who-didn't-leave gulags."
So what are you proposing, specifically here? Sending them to jail or fining women (or men?) who are abused and don't leave their partner timely? Eliminating organizations that provide abuse victims with support? Banning protective orders unless they are sought within X months of the first instance of abuse? A social campaign where people just feel comfortable being mean to victims of domestic violence (men too?) online?
What does good look like for you here?
I am calling for a shift in societal messaging away from the feminists' ridiculous "don't blame the victim" narrative. The laws should not change, and obviously help should be provided to those who leave, that's exactly what I'm advocating that these women should do.
"The slaves are rewarding the whip holder with labor, it's time to treat them accordingly".
Brutal. 🙃
But what would treating them accordingly be in this analogy? (More whipping?)
The general tenor of discourse around that, so far as I have seen, is that "women are thus proven to be incapable of reasoning, incapable of making decisions for themselves, and hopelessly addicted to being whores for low-class abusers. Therefore society should, for their own good, take away discretion as to their choices from women. Let's go back to the days when the father made decisions about who his daughter would marry, and then the woman passed under the authority of her husband.
To whom, of course, she could never refuse sex. Not like nowadays, when that bitch won't sleep with me but will sleep with the jerk beating her.
This is the only way to keep women from ruining themselves and to ensure that Nice Guys get their government-appointed compulsory girlfriend. Women cannot be entrusted with the ability to decide who they will date or marry or have sex with, the men of their family must make that decision for them".
If you think I exaggerate, I'm glad you've not read some of the stuff I've read about "women should not have the vote; not enough babies being born, solution: do not permit women to go to college or have careers, steer them towards getting married straight out of high school" and even more radical than that.
A thought experiment -- imagine that there is a group of white girls who think that there are too many black people around them. But, being girls, there is nothing they want to do about it personally. So instead, they publicly make a pledge that whenever a group of white men lynches a black man, they will give each of those white men a blowjob.
How should we look at these girls from a moral perspective? Clearly, they are not participating in the violence. And it is their legal right to have sex, or refuse to have sex, with whomever they want to. Therefore... ?
Congratulations, boys, you've made it all about your dicks again.
"Consider thought experiment: racist murder bitches giving blowjobs to guys not me. Is *that* moral?"
EDIT: Let me explain. The complaint here is not "black men are being murdered", it's "the women are having sex with the wrong kind of men".
The "wrong kind of men" being "guys not me". Even the original comment phrased it as women "rewarding" abusive men with sex.
In an abusive relationship, there may not be any choice about giving him sex. Just like the "nice guy" singer of the song who wants the woman to be with him not the abusive guy, not because he cares about her in truth but because he wants her so that's how it 'should' be.
I tried to make the thought experiment simple, you avoided the question.
Actual situation: Many years ago (the 1990s Mečiar's era in Slovakia), I knew a girl who was dating some kind of mafia guy. He participated in small business extortion (maybe also something worse, but that wasn't public knowledge), and he was making a lot of money. They both lived together in luxury -- a large house full of modern electronics, expensive vacations all the time.
Once I had an opportunity to talk to the girl. I didn't want to be impolite, but I was curious how does it feel to know that the money that finances your luxurious life was probably taken by force yesterday from some hard-working guy, maybe threatening to break his legs or hurt his family if he doesn't pay. She didn't *participate* in the crime, but she was the immediate *beneficiary* of the crime; whenever she went on a shopping trip, she was spending the kind of money she probably couldn't get otherwise. And she was perfectly aware of what was her boyfriend's job; she didn't have any illusion about it. And she was not threatened to stay in the relationship; she chose it voluntarily, because that was where the money was. (Hypothetically speaking, if her boyfriend suddenly had a change of heart, stopped doing crime, and found a normal job with a normal income, she would probably leave him as soon as her quality of life dropped towards the normal levels.)
Based on our short interaction, she saw absolutely no problem in her situation. It was him doing the crime, not her; end of story. The fact that he gave her the money, well, that's what rich boyfriends do. Where the money comes from is not her business. Yes, he has a dangerous job; if something bad happened to him because of his job (like getting killed, or getting in prison), she would obviously be sad, but it was his choice as a man to take a dangerous job, and she wouldn't try to control him and tell him what he can or cannot do, also he is not the kind of a man who would listen to such things. (Zero thoughts about the people he hurts as a part of his job.)
I have no idea how the story ended; I am no longer in contact with her, no idea even whether they are still alive or not.
What I found fascinating about this story was the difference in moral judgement when I told this story to a few men and a few women. Maybe it was just an accident, a small sample. But the men typically had the intuition that if you knowingly put yourself in a position of a beneficiary of a crime, then you are in some way... responsible? involved? Like, suppose that one of the extorted people gets crazy and takes a gun and shoots both the guy and his girlfriend in a car, then... well, they both had it coming, a karmic revenge, there is no reason to feel sorry for her. She knew what she was a part of. She benefited from the crime just like he did, only he did all the dirty work.
The women, on the other hand, typically just saw it as: the girl got lucky, and she is in no way responsible for what her boyfriend does. If a guy does crime, it makes him a bad guy, but it does not reflect on his girlfriend in any way.
I found this quite shocking back then. (It contributed to my process of redpilling.)
"In an abusive relationship, there may not be any choice about giving him sex."
You yourself said "they'll always go back and give the abuser a second/third/fiftieth chance." If they're giving them a fiftieth chance, they know what they're getting into. The truth is in front of you if you care to see it.
"Just like the "nice guy" singer of the song who wants the woman to be with him not the abusive guy, not because he cares about her in truth but because he wants her so that's how it 'should' be."
The white knight, or as you call it "nice guy," expects his righteousness to be rewarded, and the unrighteousness of the evildoer punished. I highlighted the song as an example of how the expectation is unrealistic, and your comments here just reinforces my point. Male feminists should know that female feminists will not be the least bit appreciative or thankful and see nothing wrong with rewarding abusive, decidedly un-feminist men with sex.
This is not to say that men should be against all women (I am here speaking of feminists, not all women), and certainly not that men should oppress women. What they shouldn't do is sit around writing delusional fantasy songs about how one day "justice" will be served. The path to a better world must go through a break with the poisonous ideology of feminism, which promotes the very evil it purports to fight.
"when that bitch won't sleep with me but will sleep with the jerk beating her."
Is that not an injustice? Evildoers are supposed to be punished, wouldn't you agree, not rewarded?
"If you think I exaggerate, I'm glad you've not read some of the stuff I've read about "women should not have the vote; not enough babies being born, solution: do not permit women to go to college or have careers, steer them towards getting married straight out of high school" and even more radical than that."
Those people are evil and I became a feminist because I wanted to oppose those people. I still want to oppose those people, I just think women should stop rewarding evil-doing, abusive men with sex.
The injustice is the woman being beaten, not the guy saying "but she will sleep with him but won't with me".
I won't add a link as I don't think the podcast I did on the Normans was by any means the best thing out there. But I am a big fan of Ed West's Substack the excellently named The Wrong Side of History. I think he is the best British Substacker out there and as a sort of hobby has written a bunch of very short histories (the best sort - so many history books now run to over 600 pages!). So I had him on to talk about the Norman Conquest and what led up to it.
One thing that was interesting to see was a bit of Albion's Seed avant la letter. Apparently voting patterns in Wales very much map on to where Normans settled/did not settle.
And also interesting to contrast both how close it was - the Normans really should have lost - and how total the defeat was for the Anglo Saxons. Basically the entire ruling class was dispossessed. Having brought over a huge number of land hungry knights William handed over pretty much all the land to them except for the bits he kept for himself. Oh and the bits gave to the Church which (maybe not coincidentally had given the invasion their blessing at the highest (ie papal) level.
I say the Normans should have lost. I mean they were brilliant fighters not least from having practiced on so many other people. In particular they had gone several rounds with the Byzantines and learned a lot in the process. But even so an army of that size should never have been able to conquer a country as large, rich and well organised as England. The problem for the English was that to raise and lead armies you need leaders and Harold stupidly got himself killed at Hastings along with his brothers. He'd already demolished Harald Hadrada a few weeks previously so handing over command to a brother while he got another army ready just in case was the obvious thing to do (and was what his dear old mother had strongly advised). And with the only viable candidate for the throne once Harold and his brothers got themselves killed being Edgar (aged 14) English resistance basically collapsed. There really isn't anything quite like it in European history (well, post Roman at least).
Anyway apart from the utterly brutal suppression of a rebellion in the North - deliberate destruction of crops and mass starvation - the Normans proved to be surprisingly woke in their outlook. They did away with slavery which the Anglo Saxons had been very keen on (apparently about 10% of the population were slaves pre conquest). And even more feebly they stopped executing people for political reasons. Basically nobody noble got executed until the time of Edward I. After which it was 'game on' of course. Game of Thrones in fact.
Speaking of Edward I (also a recent podcast!) I was told by Marc Morris that he was so called because his father was a massive fan of Edward the Confessor (the king before the short-lived Harold). But for that, the name 'Edward' would be just as odd as 'Aethelstan' and all the other weird Anglo Saxon ones. [Gordon Tremeshko points out this can't be right - Edward is familiar in lots of languages. I most likely misunderstood what Marc was saying - he is not given to making these kinds of mistakes whereas I . . .]
Anyway my takeaways were 1) progress is unstoppable (particularly if backed by a Norman knight) and 2) always listen to your Mum.
If you are interested the podcast is Subject to Change with Russell Hogg (though there are so many things out there on the Normans that are probably much better). And if you have room for another Substack and would like the view from Britain I can't recommend Ed West's Wrong Side of History enough.
Agreed, Ed West is always worth reading. Unusually for me, I pay my subscription to read that thoughts of a genuinely right-wing writer who leans towards pessimism of the things-are-good-but-they're-getting-worse variety.
> I say the Normans should have lost.
Duke William knew that one of Harold Godwinson's tricks was to make almost incredible marching progress with his army and confront his adversary before they were expecting his arrival. This worked with Harald Hardrada, but it was never going to work with William (A couple of years previously Harold, on a visit to Normandy, had joined William in a campaign against Duke Conan of Brittany, and no doubt William took careful note of Harold's advice and military doctrines, and his impetuous character.)
Also, William deliberately landed on one of Harold's personal estates, and laid waste to the area knowing that this would enrage Harold and make him want to hurry even faster to confront William. On his way to that encounter, Harold paused in London, where (as you mentioned) his mother and brothers begged him to wait until his full army (including cavalry and archers) was assembled. But Harold ignored them, actually insulting their caution, and rushed on with just the troops he had assembled on the way ..
> deliberate destruction of crops and mass starvation (in the North)
The harrying of the North was an exercise in area denial, to prevent the Scots and Vikings gaining much of a foothold while William was dealing primarily with the Welsh. A year or two later he marched into Scotland and burned Edinburgh to the ground, before agreeing a settlement with the Scottish king Malcolm Canmore. He also paid off the Vikings one year, and thrashed them the next, so much so that they never invaded again.
Yes, we went into that a bit when we talked. I didn’t know the details of the harrying of the North. Interesting!
Sounds pretty cool. I count myself an Ed West fan, also. One nitpick, though: that part about "Edward" doesn't sound right. That's a pretty common name in other European languages, like French (Edouard), Spanish (Eduardo), and German (Eduard).
When my wife sends a package to her brother Ed, she sends it off Edward.
At least in the Spanish case (Eduardo) it's a recently introduced name. There's no Eduardo patronymic surname as is the case in English (Edwards) because it didn't exist at the time when patronymics started being used in Spain.
Edward is very definitely an English name; Eduardo in Portuguese seems to drive from the 15th century Portuguese king Edward, son of... tbe Englishwoman Philippa of Lancaster, granddaughter of Edward III of England via John of Gaunt. I'd guess that other versions of Edward in romance languages have a similar origin, though there might be a cognate name in other Germanic languages
"News reports indicate that Secretary Kennedy plans to replace the U.S. childhood vaccine schedule with Denmark’s, the developed country that recommends the fewest vaccines."
Obvious question here: does Denmark have significantly worse outcomes in childhood diseases? Do they delay the vaccination schedule so that children get these vaccines but at later ages, or is it "Nope, you only need measles/mumps/rubella vaccination"?
"...estimate the medical consequences if American doctors stop prescribing these vaccines."
Is America significantly worse off in public health/epidemics such that Denmark can get away with this and have a healthy child population but it would be disaster for America? Before we start throwing our toys out of the pram, some data would be nice. It seems to me that babies and infants now are on a much increased schedule of vaccinations since my day, and that may be (1) yeah well, new diseases since your time, grandma, and besides you all died of polio didn't you? or (2) hell yeah this is a huge burden to put on the developing immune system, maybe slow it down there a notch.
EDIT: Happy Christmas/Solstice/Hannukah/your festival of choice/Festivus for the rest of us to you all, as well! Starting a much-needed Christmas holiday break and apart from light housework and of course cooking The Big Day Dinner, I intend to do Sweet Fanny Adams apart from sit around and eat way too much desserts and chocolate 😁
These kids probably have a lot less burden on their immune system given how much sanitation and antibiotics have gone everywhere - not sure why a few vaccines would be a bigger burden.
I'm somewhat skeptical that we're vaccinating 100% of newborns for two diseases associated with sexual activity and IV drug use which seems to add risk for little benefit given how easily you could screen Mom who's Right There.
Vaccinating 12–14yos verses the same diseases makes quite a bit more sense, much as it might feel like betting against your kids behavior. That I Wasn't sexually active in HS almost certainly wasn't my fault, as stupid an idea as that seems to me forty years on.
Why is it better to wait for those vaccines? Is the idea that they are more effective if they are taken later, or that you would get more uptake by doing it later? It seems that doing it at age 12-14 sets things up for a lot of stigma and conversations about sex and drugs, while doing it in infancy makes clear it’s just a preventative measure.
>(2) hell yeah this is a huge burden to put on the developing immune system, maybe slow it down there a notch.
My politics are whatever takes the right to an opinion away from this person.
I know, let’s fuckin kill him!
/s/
"He'll never tell you why."
Ah, Alexander. I would say "never change" but there's no danger of you ever shifting.
It's still Advent, Christmas is only days away, so extending charity to all and good luck to you.
LOL! Merry Christmas indeed.
Here's your Merry Christmas: https://youtu.be/FEu7xIEaXWo
That was great, thanks for the link!
Is there any reason to think it places a burden on the immune system?
Well, that's what I'd like to find out!
You're not going to find out from RFK Jr.
Is there any preliminary evidence? It seems quite a difficult thing to test for.
> It seems quite a difficult thing to test for.
Surely this is itself concerning, no?
Not particularly, you can't test for most speculative connections between things in biology. There are probably a thousand plants you eat in each week and a hundred leading causes of death and no one tests is say celery causes Parkinson's disease, we just assume not unless we have some kind of reason to tesf the hypothesis.
I guess prima facie it seems plausible because vaccines depend on immune reaction. That seems like enough to want to test the hypothesis.
One reason we followed a European schedule for my kids was simply informational: if a kid had a reaction to a vaccine, getting fewer at a time allows you to narrow down what's going on.
There's plenty of work that's been done on this. A quick google found the following site: https://www.gavi.org/vaccineswork/do-multiple-vaccines-overload-childs-immune-system-heres-what-science-says - which importantly links to several actual studies:
https://www.ncbi.nlm.nih.gov/books/NBK220490/ - on multiple immunisations and immune dysfunction. The summary linked here goes through several potential mechanisms for multiple immunisations to cause problems, and there is either evidence against a causal relationship between multiple immunisations and (eg) type 1 diabetes, or there isn't enough information to say (eg for asthma).
https://www.who.int/groups/global-advisory-committee-on-vaccine-safety/topics/immune-overload - the Global Advisory Committee on Vaccine Safety report summary (and a link to the full report), which says "the available evidence reviewed by GACVS does not support the hypothesis that vaccines, as currently used, weaken or harm the immune system." "Current" here means 2006, and I don't know how the vaccine schedule has changed since then.
... and here's a couple of paywalled reviews:
https://pubmed.ncbi.nlm.nih.gov/14753385/
https://pubmed.ncbi.nlm.nih.gov/11773551/
Relevant quotes I pulled out from the above paper:
"However, vaccine-specific antibody responses and rates of vaccine-associated adverse reactions of children with mild or moderate illnesses are comparable to those of healthy children. For example, the presence of upper respiratory tract infections, otitis media, fever, skin infections, or diarrhea do not affect the level of protective antibodies induced by immunization."
(references: King GE, Markowitz LE, Heath J, et al. Antibody response to measles-mumps-rubella vaccine of children with mild illness at the time of vaccination. JAMA.1996;275:704–707
Dennehy PH, Saracen CL, Peter G. Seroconversion rates to combined measles-mumps-rubella-varicella vaccine of children with upper respiratory tract infection. Pediatrics.1994;94:514–516
Ratnam S, West R, Gadag V. Measles and rubella antibody response after measles-mumps-rubella vaccination in children with afebrile upper respiratory tract infection. J Pediatr.1995;127:432–434
Ndikuyeze A, Munoz A, Stewart J, et al. Immunogenicity and safety of measles vaccine in ill African children. Int J Epidemiol.1988;17:448–455
Halsey NA, Boulos R, Mode F, et al. Response to measles vaccine in Haitian infants 6 to 12 months old: influence of maternal antibodies, malnutrition, and concurrent illness. N Engl J Med.1985;313:544–549)
"If vaccines overwhelmed or weakened the immune system, then one would expect lesser immune responses when vaccines are given at the same time as compared with when they are given at different times. However, the following vaccines induce similar humoral immune responses when given at the same or different times: 1) MMR and varicella 2) MMR, diphtheria-tetanus-pertussis (DTP), and OPV 3) hepatitis B, diphtheria-tetanus, and OPV 4) influenza and pneumococcus 5) MMR, DTP-Hib, and varicella 6) MMR and Hib and 7) DTP and Hib.
(references:
Englund JA, Suarez C, Kelly J, et al. Placebo-controlled trial of varicella vaccine given with or after measles-mumps-rubella vaccine. J Pediatr.1989;114:37–44
Brunell PA, Novelli VM, Lipton SV, Pollock B. Combined vaccine against measles, mumps, rubella, and varicella. Pediatrics.1988;81:779–784
Deforest A, Long SS, Lischner HW, et al. Simultaneous administration of measles-mumps-rubella vaccine with booster doses of diphtheria-tetanus-pertussis and poliovirus vaccines. Pediatrics.1988;81:237–246
Giammanco G, Volti S, Mauro L, et al. Immune response to simultaneous administration of a recombinant DNA hepatitis B vaccine and multiple compulsory vaccines in infancy. Vaccine.1991;9:747–750
DeStefano F, Goodman RA, Noble GR, et al. Simultaneous administration of influenza and pneumococcal vaccines. JAMA.1982;247:2551–2554
Shinefield HR, Black SB, Staehle BO, et al. Safety, tolerability, and immunogenicity of concomitant infections in separate locations of MMRII, Varivax and Tetramune in healthy children vs concomitant infection of MMRII and Tetramune followed six weeks later by Varivax. Pediatr Infect Dis J.1998;17:980–985
Dashefsky B, Wald E, Guerra N, Byers C. Safety, tolerability, and immunogenicity of concurrent administration of Haemophilus influenzae type b conjugate vaccine (meningococcal protein conjugate) with either measles-mumps-rubella vaccine or diphtheria-tetanus-pertussis and oral poliovirus vaccines in 14- to 23-month-old infants. Pediatrics.1990;85:682–689)
"Vaccines may cause temporary suppression of delayed-type hypersensitivity skin reactions or alter certain lymphocyte function tests in vitro. However, the short-lived immunosuppression caused by certain vaccines does not result in an increased risk of infections with other pathogens soon after vaccination. Vaccinated children are not at greater risk of subsequent infections with other pathogens than unvaccinated children. On the contrary, in Germany, a study of 496 vaccinated and unvaccinated children found that children who received immunizations against diphtheria, pertussis, tetanus, Hib, and polio within the first 3 months of life had fewer infections with vaccine-related and -unrelated pathogens than the nonvaccinated group."
(references:
Black SB, Cherry JD, Shinefield HR, et al. Apparent decreased risk of invasive bacterial disease after heterologous childhood immunization. Am J Dis Child.1991;145:746–749
Davidson M, Letson W, Ward JI, et al. DTP immunization and susceptibility to infectious diseases. Is there a relationship?Am J Dis Child.1991;145:750–754
Storsaeter J, Olin P, Renemar B, et al. Mortality and morbidity from invasive bacterial infections during a clinical trial of acellular pertussis vaccines in Sweden. Pediatr Infect Dis J.1988;7:637–645
Otto S, Mahner B, Kadow I, et al. General non-specific morbidity is reduced after vaccination within the third month of life—the Greifswald study. J Infect.2000;41:172–175)
Interestingly, it also points out that in the past, a child receiving the smallpox vaccine got over 200 proteins, while the current (as of 2002) schedule includes fewer than 130 proteins in total. That's partly due to smallpox no longer being needed, and partly due to improvements in protein chemistry resulting in vaccines containing fewer antigens.
These are all dated around 2002-2006. There's loads of current studies on individual vaccinations, but I couldn't find any recent ones about multiple vaccinations, maybe because it's hard to persuade funders to pay for something that's already been shown multiple times?
Happy Christmas to you as well.
One major difference I'm aware of is that Denmark doesn't vaccinate for Hep B, but they do screen all pregnant women for Hep B (which we do not do in the US). I think the logic here is that more kids go to the pediatrician, than pregnant women go to the OB (+follow up with required bloodwork).
In the US, we do *try* to screen all pregnant mothers in the US for Hep B. In practice, we manage to screen 82%-88% [1].
Most of the remainder is people who couldn't get into prenatal care promptly (maternal care deserts are a huge problem, especially rurally, and especially in states that restrict midwifery practices), or people who moved and switched providers and didn't transfer records. Denmark probably does not share those problems.
Trump's ACIP recently voted to change the Hep B recommendation from "everyone gets this at birth" to "everyone gets this at birth, unless the mother has tested negative, in which case she can either request it at birth or wait until 2 months."
I have heard the ACIP meetings have been chaotic and irrational, and have nearly missed making some much worse decisions. But as far as what they've actually passed so far, like the Hep B revision, I don't have strong opinions about it. It's not actually a big change from what we used to have. It's a little more complex, but because the exceptions are limited to cases where we do have data, I don't think it's a big problem. I don't appreciate politicized commenters elsewhere calling them a "death cult" over things like that, but I agree they certainly seem incompetent and at risk of passing actually detrimental recommendations in the future.
[1] https://yourlocalepidemiologist.substack.com/p/we-dodged-a-vaccine-disaster-for
Hey everyone, I'm a bit bummed about having missed the Oxford meetup back in October :/. Is the group still active/exists at all?
I wrote a short story that takes place at Oxford University and wanted some feedback from people who know it and how it works (especially since I think the story might be more interesting to people who have some attachment to the place). Anyways, any feedback is welcome!
https://open.substack.com/pub/emanueledipietro/p/automatic-phonetic-transcription?utm_campaignpost-expanded-share&utm_mediumweb
I'm a linguist (albeit not a phonetician), and I'm a bit confused about the experimental setup. It seems like the way it works is:
1. Two subjects have a conversation in a room and record themselves.
2. An automatic parser takes the audio file and outputs a transcript in some kind of phonemic or phonetic system.
3. The audio file is discarded, or at least locked away where NB can't access it.
4. NB transcribes the output of the parser into English orthography.
Step 3 doesn't make sense for a speech production phonetics experiment. If the researchers are interested in how the subjects pronounce /s/ (or whatever), they care about fine-grained acoustic details that would be entirely lost in even a very detailed phonetic transcription. They would want to keep the audio file so they could make a spectrogram and measure whatever they need to measure.
I guess I hadn't really fleshed out the experiment setup, but if I had to answer in order for it to make sense, I guess I would make the annotation phase preliminary to the human analysis/review part of the experiment, where the grad student (presumably) that is tasked with going over the material flags the specific part for later manual inspection.
It's probably not 100% satisfying to a phoneticist that actually carries out these sorts of studies, sorry :/
The story just gets going and then it stops? We are meant to guess that A and B are now hooking up, and maybe going to be dating. The Italian story then is a red herring? Because I don't think really serious Buddhists would be hanging out in a bar, or at least not drinking alcohol. Not if they've filed their teeth down to commit to vegetarianism.
And we don't even know that they *are* Buddhists, that's just a conjecture. It's a nice build-up to the conventional (and nothing wrong with that) Christmas ghost story set in Oxbridge, but we end up with "the transcription machine works but doesn't yet recognise kissing/groping/fucking noises" and that's not where I thought it was going.
Yeah, I guess I didn't really have a conclusion in mind when I started it. The general concept that I had in mind was: the physical makeup of objects (and people) has an impact on how they work/what their purpose is.
I kind of had the idea in mind that it would thematically tie in the Italian story and the flirting/love story (although now that I'm typing this out, it would be better to make it specific to A and B's connection rather than generically for male-female).
I guess in the back of my mind this is also tied into phonetics i.e. the physical manifestation of language in its most concrete form.
Anyways, I appreciate the feedback, and especially from you, Deiseach! I've gotten into trying to write more structured fiction, but I always struggle with keeping everything thematically consistent and also narratively satisfying. Who would have thought that writing would be this hard!
The stop was very abrupt, so it was hard to know what the point of the story was. You have a decent little traditional ghost/horror story going on there (maybe the flat-teeth people start showing up around Oxford? are they following A?) but the bit about the antagonists-to-lovers with A and B and then the experiment muddled it a little.
The bit about "the physical manifestation of language" sounds like a good idea to explore (does telling the story evoke the weirdness of the people? does it begin to cause physical change in A and B? what of the person transcribing the tapes?) so if you ever do work on the rest of it, let us know!
Oxford Phonetics - Very Lovecraftian :)
I'd bring the entry into the bar slightly sooner - I have a fondness for descriptions of Oxford but was wondering when we'd get to the bite. Or give us some other hook earlier on. "I didn’t even get to the point, or even the spooky part" - you need to :p
The recording stopping just after we get into the pub doesn't really work as a cliffhanger, because we don't yet know why we care about these weirdly-smiling "men and women of various ages". I think you need to get a hook in before any interruption. By the second interruption, I'm engaged.
"had I been sober, I would have immediately looked at the floor" - I liked this very much :)
Minor note: Oxford has its own names for many things. Rather than Autumn, Spring, and Summer we have Michaelmas, Hilary, and Trinity. The Thames gets called the Isis as it flows through the city (Tamesis is the ancient Latin name for the river).
Somehow, I found the... commitment device... very funny :)
I thought the ending had a pleasantly meta call-back, but it left me a bit underwhelmed tbh - I wanted a bit more setup->resolution, more payoff, in the main story. The writing successfully spins an atmospheric Oxford, but you could improve the pacing of the story's beats, and increase the stakes.
Overall fun! Thanks for sharing, would enjoy talking in person if you come by the pub some time.
Thank you for the feedback! I actually don't remember why I chose not to call the river Isis, but I definitely knew this. I'll see you at the pub then!
Hello! There's a small group that meets every third Wednesday, except when we don't :p https://tinyurl.com/oxrat-signup to get email announcements, and we tend to announce on https://www.lesswrong.com/groups/wQA8BE5e8mETeWb8A .
I have a pretty intense dislike for Deciding To Win, on several levels:
- It struck me as mostly poll-chasing and abdication of leadership
- Basically no discussion of which things are actually good, independent of polling (again, make an argument for good things, don't just assume the polling results are inevitable)
- I didn't feel like they even really followed their own polls that well (e.g. they conspicuously omitted "raise the top tax bracket" from the main report even though it polled quite well)
- They said Medicare for All was unpopular, which is only really true if you formulate the question as "Medicare For All, and we ban all private insurance", which is not a necessary formulation
[EDIT:] Also, it's not really accurate to describe people like Davids Axelrod and Plouffe as "rogue Democrats". They're longtime party insiders.
I mean you answered your own question. They left out popular, growth-killing taxes, because they are bad policy.
I really don't think that position survives a read of all of the stuff they left in.
Isn't that the entire point of democracy? They're supposed to be representatives of the people, not pursuing their own agenda. Poll-chasing to get the majority vote is simply the system working as intended. The only purpose politicians have is to express the will of the majority.
First, I strongly disagree that the point of elected representatives is to just vote whatever polls say. The point of elected representatives is to do what's actually useful at achieving what people want. But people can support counterproductive policies that will actually make them worse off. Having a class of professionals that (in theory) studies these things and tries to work out good solutions is valuable.
Second, if you do believe this, why would you specifically want one party to win? These people obviously want Democrats to win; they're just failing to articulate much of a reason for it, since they obviously don't think that Democrats are laser-focused on doing exactly what polls say.
> Having a class of professionals that (in theory) studies these things and tries to work out good solutions is valuable.
And what if people don't want those professionals? You are supplanting the people's will with your own agenda. If the people support policies that make them worse off, that is ultimately their problem, and fixing that is outside the purview of democratic politicians.
> Second, if you do believe this, why would you specifically want one party to win?
Because they represent the majority? Isn't that the entire point of democracy? Eliminating conflict by ensuring that the majority is satisfied?
On the first part I just fundamentally disagree that the best thing we can do is elect leaders who mindlessly poll and vote. We could switch to direct democracy for everything if we wanted to do that. Which I don't.
> Because they represent the majority? Isn't that the entire point of democracy? Eliminating conflict by ensuring that the majority is satisfied?
1. Only satisfying the majority does not eliminate conflict.
2. Also, no, I think many political philosophers would say that the point of democracy is to try to synthesize people's desires. Certainly the majority gets the most influence but that doesn't mean any 45% opinion is essentially meaningless and you should govern as though it doesn't exist.
3. I don't think the authors of Deciding To Win actually think that it's just inevitably true that Democrats represent a consistent majority of people. In fact, in general, most of them are people whose entire career is about helping Democrats win (or try to), regardless of whether that's true at any given moment or in any given place.
> Only satisfying the majority does not eliminate conflict.
It does eliminate conflict when dealing with rational actors. Granted, some people are very, very irrational, but... they do not pose any meaningful threat to the collective as long as they are heavily outnumbered and outgunned. Others will understand the position they are in and act accordingly.
> Certainly the majority gets the most influence but that doesn't mean any 45% opinion is essentially meaningless and you should govern as though it doesn't exist.
I mean sure, you shouldn't pretend they don't exist, but that doesn't mean you need to meaningfully accede to their demands. That's what military and law enforcement is for.
> I don't think the authors of Deciding To Win actually think that it's just inevitably true that Democrats represent a consistent majority of people.
Of course, and isn't that why they are promoting completely changing the DNC agenda to actually represent the majority? What they are advocating for is completely unrecognizable from the existing Democratic party. Going third party is not viable given the circumstances, and the Republicans are, well, not particularly interested in what liberals want. Which leaves taking over the Democratic party the only way of them maintaining the status quo, especially given that liberals have also failed to take any measures to protect their majority.
>> Only satisfying the majority does not eliminate conflict.
> It does eliminate conflict when dealing with rational actors.
I'm not sure what you base this on; this would imply any resistance movement ever was irrational no matter the stakes. Or that the entire civil rights movement was irrational.
> What they are advocating for is completely unrecognizable from the existing Democratic party.
Again I do not follow why you think this. They're literally taking a bunch of preexisting Democratic policy proposals and polling on them and then saying to focus on the ones that poll the best. As far as I can tell they do not suggest e.g. suggest adopting any Republican policy positions.
I get a little uncomfortable when people discuss strategies of the form "say you are going to do [the popular thing] publically, that way you'll win, and once you win you can do [the thing we actually want to do]." I get this vibe from a lot of Matt Yglessias' political shop-talk. Makes me distrust the political class.
Nate Silver had a really good article on the new factions within the Democratic Party (1) that I would highly recommend. He highlights those factions as:
(1) "Capital L Left", commies, think Mamdani & AOC
(2) "Abundance Libs", a la Ezra Klein & (frankly) Nate Silver.
(3) "Resistance Libs", of "No Kings" fame, basically SJWs who hit middle age
Deciding to Win seems like a pretty bog-standard "Abundance Libs" group but that's also a Dem internal debate going back to Bill Clinton's "centrism" in the 90s. Personally, I don't think the Democratic Party has been terribly good for Abundance Libs and I'm hopeful that Vance/DeSantis can round off some of the edges of Trumpism and do some awesome stuff. Not hoping that Abundance Libs will become Republicans but...man, an actual sensible block of centrist voters who can work with either party seems a much better future than remaining loyal Democratic voters to a party that seems determined to lose minority voters, alienate male voters, and abandon growth policies in favor of various ideological commitments. Remember when Kamala Harris ran on a capital gains tax that would have ruined the market and destroyed the entire startup ecosystem? (2)
C'mon Abundance Libs, come over to the Republican party. Our madness will subside in 3 years when one man leaves office; the Democratic party's madness will subside when tens of millions of committed ideologues all abandon their various communist/idpol ideas. Also, we have cookies!
(1) https://www.natesilver.net/p/what-is-heather-cox-richardsonism
(2) https://www.forbes.com/sites/robertwood/2024/09/09/kamala-harris-28-capital-gains-tax-plan-is-a-surprise-heres-why/
"C'mon Abundance Libs, come over to the Republican party. Our madness will subside in 3 years when one man leaves office"
Yes, Abundance Libs, you want to vote for Mitt Romney. You need to vote for Mitt Romney! Perhaps to lose once again, but who cares?
Well, Abundance Libs didn't lose with Mitt Romney, they won with Obama. Conservatives lost with Mitt Romney.
I do agree that this isn't just a "Republicans need to change" thing. If Republicans run Mitt Romney 2.0, a bunch of Abundance Libs would actually have to, ya know, vote for him so he doesn't lose.
But it does remain to be seen whether:
#1 Democrats will moderate or keep running radicals (ie, the same kind of candidates who've lost to Trump 2/3 times)
# 2 Whether Republicans will moderate or double down on Trumpism (probably determined by the Vance/DeSantis primary)
#3 IF Democrats don't moderate AND IF Republicans do moderate, THEN will any "Abundance Dems" actually vote Republican?
Which feels stuck on the following three facts:
#1 Everyone agrees that less partisan voting and more centrists who are willing to switch sides would probably lead to better politics.
#2 Abundance Dems are probably one the most likely groups, both ideologically and practically, to adopt a centrist/switching voting pattern.
BUT
#3 Every Abundance Dem I've ever met in real life views voting for Republicans with essentially the social and moral horror of having a swastika tattooed on their face.
The problem with centrist voting is you eventually, you know, have to vote for the other party. And I've jumped parties, I'm an Obama->Trump voter, and I'd love to be able to pivot back to the Democratic party or at least have them as a serious option but...I'm far, far more optimistic about DeSantis or Vance reforming the Republican party than the Abundance Dems reforming the Democrats.
First Jeff Maurer piece that made me honestly laugh, and by coincidence it's on this very topic:
https://www.imightbewrong.org/p/what-do-resistance-libs-want
"There are parts of Nate Silver’s article about “Heather Cox Richardsonism” that I agree with and parts that I don’t, but the part that I think is emphatically true is that three groups are fighting for control of the Democratic Party. Silver calls these groups “Abundance Libs”, “Resistance Libs”, and “the Capital-L Left”, but I think it’s more fun to call them Techno Weenies, The Khmer Blue, and Trust Fundinistas. For the record, I’m a Techno Weenie, so be aware that this column is written by someone who is trying to foment a political sea-change based on support for single-stair buildings."
>Our madness will subside in 3 years when one man leaves office
I honestly think this remains to be seen. There is a major internal fight going on in the Republican party between the faction that is indeed likely to return to something resembling normalcy and the "groypers" for lack of a better term to describe the more insane fringe.
I personally wouldn't venture a guess on which group will remain ascendant, but I will look askance at anyone who claims to be very certain that _they_ know. (note that I could make a relatively similar comment in the other direction, so this is not meant to be an attack/defense of either political party)
I won't claim to be "in the know", but when long running, high status, conservative think tanks are disintegrating before our eyes, and major commentators are having public fights at large conservative conferences, I think we can, at a bare minimum, say that there is a fight going on for the future of the conservative movement.
And I'm really not sure what your point about Biden is. The fact that they were less obvious about it doesn't seem to have any bearing on whether it's happening now or now.
I honestly don't know what else someone would call what is currently happening unless one is so cynical as to think that it is all Keyfabe and theater, a circus for the ignorant masses. And I am not yet that cynical.
This entire conversation is about what will happen post-trump. Whether trump gives a shit about this people is entirely beside the point. You might be right that none of these groups ends up winning the power struggle and becoming relevant, but to argue that there _isn't_ a power struggle over what the post-trump party will look like seems obviously incorrect to me.
Not addressing her much of your comment, but, I think "Deciding To Win is taking an abundance position" is only true insofar as a lot of abundance positions are pretty popular. If they stop being popular, presumably the group will turn against them.
Note that it only discusses what policies to campaign on for winning elections, and not what you implement while in office. It seems good strategy to hold off on the leadership and "which things are actually good" until AFTER you get the suckers to vote for you.
True, although it would therefore be equally useful for the Republican party to use this exact strategy (including these specific positions).
I think that's just an artifact of there being no single party leader. They want the 2028 nominee to champion these positions, but there are other factions with different ideas. Bernie has been the avatar for left populism so his endorsement is all that really matters there. Obama's influence has waned because it's been a long time since he was in the White House. Biden never really had influence; he was a coalition manager.
My read of this is that they only really express any deep interest in the 2028 nominee *winning*, but have no deep attachment to any of these positions beyond that.
I am astonished how often I find myself in agreement with Shoe0nHead, but I have to agree with her once more here - wow, what a title. "Deciding to Win", huh? So all along they had decided to not-win, but now it dawns on them that they should have been deciding *to* win? Such an astounding insight! 😁
https://www.youtube.com/watch?v=gli4Ox4i54I
It's surprising how often people in politics DON'T prioritize winning.
People in politics generally DO prioritize winning, but the winning they prioritize may mean winning intra-coalitional struggles, which in turn may require the adoption of strategies that reduce the possibilities for winning in the general field of politics.
There are things people care about other than winning politically that can get in the way. For example: unpopular but strongly held moral convictions, personal reputation/social consequences, in-group loyalties, general hubris.
While I like what little I've seen of her, I disagree with this, and think the title is informative after all. It's reasonable to say that Party HAS lost focus on winning, and instead doing things out of habit, or moral conviction, or as a jobs program for its allies, and needs to now stop all that and actually try winning elections instead of occasionally doing so accidentally.
I feel strongly that American democracy works best when a strong candidate who believes in centrist positions manage to use their sincere beliefs and the fact that Joe Public agrees with them to secure their party's nomination, and we then choose between two such candidates.
I feel strongly that it does *not* work so well when party strategists decide that their terminal goal is "winning" and that they therefore must convince other party strategists to conceal their true beliefs in the hopes of sneaking into office where they can then take the mask off and do what they've been wanting to do all along.
The first case you describe is almost exactly the 1960 Kennedy–Nixon election.
The second is the election of Franklin Roosevelt.
You might disagree, but I believe most people today would consider the latter a better example of American democracy working.
I do.
I read the Washington Post and have been surprised by the number of stories on microplastics recently. Can anyone tell me if these article authors are right, and we should all be terrified of microplastics, or is this just a big cottage wellness industry? I can't tell if the concern for microplastics is based on legitimate scientific research and expertise or conducted by a bunch of people really motivated to find negative effects of microplastics. Thanks for any pointers.
my meetup group did a shallow lit review that you can read here: https://www.lesswrong.com/posts/GjbXGybzszw8eN3oB/microplastics-much-less-than-you-wanted-to-know
research on this topic really only started a few years ago, it'll be some time yet before we get anything definitive.
If microplastics have negative effects, they presumably aren't large, since most population health metrics have been significantly improving over the decades. E.g. people have microplastics in their brains, now, but the incidence of dementia is lower at every age than it was in the past: https://x.com/cremieuxrecueil/status/1907919796295053632.
The main exception is the rise in obesity, but for various reasons, that pretty clearly isn't caused by microplastics. Most of the microplastics studies I've seen touted haven't been causally informative, at all (e.g. https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(25)00174-4/fulltext ), yet their conclusions have been misrepresented as such in media (e.g. https://www.cnn.com/2025/04/29/health/phthalates-heart-disease-wellness).
Of course, study authors deserve a share of the blame for often misrepresenting their findings in titles and abstracts, and further misrepresenting them to media. E.g. the aforementioned study admits that "No inferential statistical comparisons were conducted," such that their observations were entirely observational and unequipped to make any sort of inferences of the causal effects of microplastics, but the entire study consists of just that - observing that countries with worse health outcomes tend to have higher levels of exposures to plastics, baselessly assuming that the former is entirely attributable to the latter, and deriving some massively negative estimate of the health of plastics, without any sort of sanity check, whatsoever.
E.g. their study is about cardiovascular risk, and doesn't consider that the age adjusted prevalence of heart failure in the US hasn't changed for decades (https://x.com/cremieuxrecueil/status/1993936408416055711). Is that consistent with a huge share of global cardiovascular risk being attributable to plastic exposure? Presumably not.
Emily Oster's take can be found here: https://parentdata.org/how-much-should-we-worry-about-microplastics/:
"Direct evidence of the health impacts of microplastics is extremely limited...The conclusion of this paper provides a nice summary of what we know and do not know about these issues. The authors argue that while there is clear potential for health impacts, the evidence we have from non-human models is difficult to extrapolate to people. And they argue that we need more research."
I think it's worthwhile to note that the existing literature lacks causal evidence of harm, but that it would have been valuable for her to run some sanity checks to get a better idea a priori of how plausible claims of various large effects are, like I did here and my comment below, regarding fertility.
Fertility is also a concern.
That’s because of instant yeast, not microplastics.
Oh, wait, that’s 1925 news.
Every new thing since instant yeast has had its turn at causing infertility; tight jeans, cell phones, carbs… AI and microplastics next, I guess.
In 1905 it was women being too educated; developing the brain takes the bone-minerals needed for developing babies. Also educated women get married very late or not at all, and so only have a couple of children at the very most:
https://archive.org/details/popularsciencemo66newy/page/466/mode/2up
"The duties of motherhood are direct rivals of brain work, for they both require for their performance an exclusive and plentiful supply of phosphates. These are obtained from the food in greater or less quantity, but rarely, if ever, in sufficient quantity to supply an active and highly educated intellect, and, at the same time, the wants of the growing child. The latter before birth must extract from its mother's blood all the chemical salts necessary for the formation of its bony skeleton and for other tissues; and in this rivalry between the offspring and the intellect how often has not the family physician seen the brain lose in the struggle.
...The highly educated woman seems to know that she will make a poor mother, for she marries rarely and late and, when she does, the number of children its very small."
Education, instant yeast, tight jeans, microplastics - is there anything that is not responsible for the impact on fertility?
>"…and in this rivalry between the offspring and the intellect how often has not the family physician seen the brain lose in the struggle."
"Mommy brain" at least seems to be a real phenomenon; my cousin (who designs stealth composites, so crazy smart normally) left for a vacation with her other car unlocked AND RUNNING in the driveway while pregnant with one of her kids.
Yeah, but that's due to hormones, not your cousin being over-educated because her parents let her go to university.
> In 1905 it was women being too educated; developing the brain takes the bone-minerals needed for developing babies. Also educated women get married very late or not at all, and so only have a couple of children at the very most
Women being too educated is still a major cause of infertility today.
But not because brain and baby are locked in a death-struggle for phosphates, Michael.
Again, sanity checks don't seem to suggest that microplastics have a large effect. In spite of large changes in environmental exposure, we haven't seen large changes in fertility markers. E.g. in males, per: https://www.sciencedirect.com/science/article/pii/S0098299725000627#sec4: "The authors found a substantial stability of total sperm number during the last decade in healthy subjects ... Auger et al. (2022) suggested that the literature does not support the conclusion that human sperm quality is declining globally or in the western world ... In recent years, there has been growing concern about endocrine disrupting chemicals (EDCs) in the environment, in food and in consumer products... For adults ... the association is limited. The data are conflicting, with some papers suggesting a negative correlation with at least one of the semen parameters, or with no parameter at all (Pallotti et al., 2020)."
The infertility rate for females has also been fairly constant for decades: https://www.hopkinsmedicine.org/news/newsroom/news-releases/2022/06/us-infertility-rate-plateaus.
there has been a significant increase in some types of cancer among younger people in the last 20 years.
Per: https://www.cremieux.xyz/p/early-onset-cancer-fast-facts, the apparent increase in various kinds of early onset cancer is mostly a result of increased diagnosis - not actual increased disease prevalence. The main exception is early-onset colorectal cancer, which has seen a moderate increase in prevalence and mortality. However, I have no reason to think that microplastics, in particular, are responsible for that. Notably, while the rate of colorectal cancer in younger people has risen slightly, it's been more than offset by a decrease in the rate for older people (https://www.cancer.org/cancer/types/colon-rectal-cancer/about/key-statistics.html).
Ok.
Hmm. Good to know. Another thing to take away from my we are doomed list. A few hundred to go.
So this testosterone drop has been totally misreported? Or had it just stabilised recently.
Apparently mostly a result of changed measurement methods: https://xcancel.com/cremieuxrecueil/status/1980487130209235003#m.
I believe most of the research is being conducted by Joe Rogan
For anyone wondering where the actual AI value saves to companies are at: it looks like they’re going to start hitting next year.
I’ve now seen two vendor demos that basically have similar infrastructure to what I worked out (they didn’t do structured language the way I did, but they mostly did state machines the way I did) and I believe another major company is working on similar. We’re in a period of mass industry learning which is why this has been slow to be realized. I know this from working on this stuff firsthand.
Companies are going to need to rethink how they are doing project teams for best practices. For instance, all the coders I know have bad instincts for how to build with LLM’s. I think this group (meaning ACX readers) might be biased by knowing lots of folks who are exceptionally forward looking but unless I personally get in the code and change prompts stuff doesn’t work when I work with developers who don’t breathe this stuff. It does seem that prompt engineering and sme knowledge is a skill that still matters for tasks that don’t have a ton of examples on the internet.
In summary, some value next year for early adopters. State machines are the way to go. Probably won’t see mass adoption for three to five years. But the ice is starting to thaw.
There was very little specific information there. In paragraph two we are informed that some vendors are selling something that you yourself worked out, the vendors rename unnamed and the thing unexplained.
Then there’s a blanket statement that coders make bad prompters (better to hire taxi drivers) without explanation, and then you yourself need to be in there writing the prompts or it’s just not going to work. Hope you can scale yourself up, or this is really a bubble.
Sierra and PolyAI were the two vendors. Had actually thought through practical business cases and audit readiness.
No idea how this works on the timeline of the markets as far as bubbles go.
You can think I’m full of crap and that’s fine, but it’s not like I can leave a comment that does every single aspect of the workarounds. My guess is if I did a deep dive into my specific use cases you’d say the way I’m handling it doesn’t even count as an agent. Even in the most limited case, do we expect the best ever language classifier to not have productivity boosts?
Sharing from my own perspective as someone trying to make them work at scale. Not a clue on the markets. My guess on timelines is what I’ve shared above. There’s a lot of learning and upskilling that needs to happen.
It seems like specialized SAAS is most at risk. Are there any companies offering a sort of "headless SAAS" architecture that allows you to spin up bespoke AI applications on top of them? Just thinking out loud...
There are and they get buy in but honestly they don’t seem that long term valuable to me. The people who can do the whole conversational layer stack seem valuable to me. To me the best business to run is middleware that doesn’t do anything but gets so much shit built through it that no one can afford to get rid of it.
A general impression I've gotten: OpenAI has veered away from the "create God" corporate model, and has joined Meta and xAI in basically trying to design yet another ad-supported engagement-maximizing enshittifying tech platform. That just seems consistent with all their recent moves. Anthropic seems to be the only one of the AI companies that are actually seriously committed (for better or worse) to building AGI, with all that entails.
Does that seem accurate?
Anthropic is also the only one that explicitly disavows the term “AGI”! I think they’ve had a more cautious and open-minded approach to what sorts of powerful AI might be possible, so that if everyone is thinking that early visions of AGI might not actually be possible (or likely any time soon), Anthropic may be the only one with significant thoughts that something else powerful might be nearby.
Another thought might be just that Anthropic thinks this sort of social media slop factory is bad for humanity, and thus they don’t want to do it even if it’s the most profitable next step for everyone on the way to something bigger.
There's also Sutskever's SSI: https://www.youtube.com/watch?v=aR20FWCCjAs&t=2581s
I don't know if what they're saying matters, really. Both OpenAI and Anthropic want a better model than the other. For as long as this is true, they are working towards building AGI. The switch to ad-support could mean that OpenAI leadership have decided that creating God is unlikely. It could mean that the VCs have decided creating God is unlikely, or are just getting impatient, and want to see some revenue. It could just mean that they're trying to extend the create-God runway.
I suppose that makes it weak evidence that OpenAI leadership have longer timelines, and since they presumably have a lot of nonpublic information, their beliefs should be somewhat persuasive. But I think as things currently stand, OpenAI is making just as much progress towards AGI as they were before the switch. (I have no idea how much progress that is.)
"Better" can mean different things. I'd rather OpenAI focus on making small models that can easily run just on ads rather than investing in large ones that will be difficult to run but are more intelligent.
Everyone and their dog is writing that OpenAI are beyond fucked at this point... check out this guy: https://www.thealgorithmicbridge.com/p/you-have-no-idea-how-screwed-openai
Deepmind is explicitly attempting to create AGI. And while I think Altman has too much VC in him for his own good, I do think they're still aiming at AGI. "Lord, give me AGI, just not yet", basically.
If Deepmind was serious about AGI, and believed that it could be achieved with some fixed amount of R&D (to rough approximation) then the rational strategy would be to go all in on R&D, stop giving away free inference, and worry about revenue and market share after achieving AGI. It doesn't matter how many subscribers chatGPT or Gemini have once AGI arrives.
So the fact that both OpenAI and Google are chasing chatbot market share, and giving away tons of free inference makes me think that neither are truly serious about AGI and are hedging their bets.
Only Anthropic is behaving in a way that makes me think they really believe in AGI soon. Most of their efforts go towards R&D and revenue generating products like Claude Code and API credits.
FWIW, I think Deepmind is the only major lab without a really free offering - if you use Gemini for free, Google can use your data for training, whether in the app or API. No optout like Anthropic/OpenAI.
User data is mostly useless for developing AGI. Most people are using LLMs for information lookup, generic advice and writing. The things that LLMs are already pretty good at*.
The main gaps missing between current models and AGI are the ability to interact with multimodal environments in a continuous fashion, perform heavily multi-step tasks rather than one shot tasks, and continual learning - being able to seek out and integrate new information to solve problems and self improve.
User data doesn't help much with any of those.
*Source: https://cdn.openai.com/pdf/a253471f-8260-40c6-a2cc-aa93fe9f142e/economic-research-chatgpt-usage-paper.pdf
I would love someone making an effort-post on Denmark's vaccine schedule and why it differs from the US/what to think about it. Does it boil down to "Nordic countries can do things differently because Nordic"? I could see the communicable disease burden being lower in a Nordic country.
I often see this sentiment - "nordic countries does x well, but this obviously don't apply to other countries". Where does this come from, and why exactly can policies that work well in the nordics not be successfully copied other places?
Very small (things that work for countries of 5-10 million become harder at 330), culturally uniform, high trust, Norway has tons of resource wealth, etc
I agree with you that high trust is an important factor for nordic country successes, but if so, it seems important to understand how to archive high trust? Unless you think the successes are statistic noice, I don't see why something that works for 5-10M can't be scaled prima facie. Are nordic contries really that much more culturally uniform. Norway is special due to oil wealth, but that does not explain similar success in Denmark, Iceland or Finland.
It's easier to achieve high trust when you are small and ethnically homogenous.
If you don't understand why it's easier to do something with 5 million people than 340 I'm not sure what to tell you. Go try to manage 5 people and then 340 and report back.
Well, I'm not saying that it is not easier exactly, and there probably are some things that are ok for 5M people, but hard to do with 300M (and vice versa). What I react to is this blanket statement, when some good outcome is seen from a nordic country - well this will not work in the US. I would think there are many policies that can be done at a larger scale too. I wonder were this thought comes from? Did someone actually analyze, say nordic countries subsidized higher education policies and come to the conclusion that this could never work in a larger country?
Now your example of 5 to 340 people is not a good one. 5M people are still a lot of people, and requires basically the same level of management as 300M, but on a smaller scale. The management of 5 people can be done done more or less ad-hoc and by direct communication, while 340 people requires significant overhead and organization.
I'm not sure that that nordic countries are as ethnically homogenous as you think. They were ethnically homogenous 50 years ago, but today I'm not so sure that this is true any more.
Id have to pull the statistics, but they were ethnically homogenous 20 years ago, and to the extent that had changed, there seem to be meaningful anti-immigrant pushes in a number of them.
I won't have time to dig deep until after the holidays, but you've been a reasonable interlocutor, so certainly can pull numbers after if you are interested.
Here is an un-paywalled version of a StatNews article on the topic: https://web.archive.org/web/20251220011615/https://www.statnews.com/2025/12/19/denmark-vaccine-schedule-vs-us/
tl;dr seems to be some combo of "Danish kids are at much lower risk for some of these diseases" and "Denmark has made the (questionable) policy choice to deal with these diseases within their socialized healthcare system when they arise instead of trying to prevent them"
I am still biased towards interpreting the phenomenon as like a nimitta - but from that perspective I am particularly interested in how their sun vision turned "black almost entirely except for the edge, which became pink". This sounds a lot like recent experiences of my nimitta - it turns black and silhouettes itself when focused on in some cases. I do not understand myself but it still seems highly comparable...
I’ve been toying with a framework that might unify a few things people usually talk about separately: stress, health gradients, and socioeconomic inequality. This is very tentative, but it feels mechanistic enough to be worth stress-testing rather than dismissing.
The starting point is boring physiology. Living systems stay functional by combining buffering (passive absorption of disturbance), export (removing disturbance), and regulation (active control when the first two aren’t enough). When disturbances are chronic and buffering/export are inadequate, regulation stays elevated, and you get what medicine already calls allostatic load: elevated cortisol, inflammation, metabolic drift, cardiovascular strain. None of this requires psychology or subjective distress; it’s measurable.
The extension is to treat “stress” not as an emotion but as regulatory load: how much active control a system has to exert just to avoid degradation. From that angle, wealth mostly buys buffering and export capacity. Stable housing, temperature control, predictable schedules, slack, delegation, early medical intervention — all of these reduce how often regulatory systems have to fire. Poverty does the opposite: fewer buffers, more exposure to shocks, and more situations where the only way to keep things together is continuous vigilance.
This predicts that income–health gradients persist even when you control for individual behaviors, but are mediated by structural buffer variables like housing stability, schedule control, and shock absorbers such as savings. That seems broadly consistent with existing data, but I don’t know how cleanly it’s been tested.
What I like about this framing is that it doesn’t require bad choices, weak character, or malicious elites. You just need unequal buffer distribution and finite sinks. Systems that don’t assign where disturbance is absorbed don’t last; systems that do will concentrate load somewhere. That’s selection, not conspiracy.
It also makes some concrete predictions. External structure should reduce physiological stress markers more reliably than additional cognitive reframing alone. People reporting high “load” should show vigilance signatures like micro-checking behavior, HRV suppression, or flattened cortisol slopes regardless of mood. Matching people on diet and exercise shouldn’t erase SES biomarker gaps. And AI tools, if they help, should mostly reduce vigilance and context-tracking costs rather than total effort — and mainly in domains where the constraint is informational rather than physical.
The whole thing lives or dies on those predictions. If structural changes don’t measurably reduce load, if buffer variables don’t mediate health gradients, or if AI reliably increases vigilance even when people say it helps, then this collapses pretty fast. I might be missing something obvious, but as a unifying mechanism it seems at least falsifiable enough to be interesting. Curious if anyone knows literature that already kills this idea or, alternatively, supports it more strongly than I realize.
Interesting! But I'd expect that buffering should saturate at some level if that is the only significant mechanism for turning differences in socioeconomic status into different e.g. medical outcomes. Once one can control the temperature of one's home to +-3 degrees, tightening the control to +-0.3 degrees doesn't matter much.
Unfortunately, there also seems to be a (zero-sum) status component too. I'd vaguely recalled that there were differences even between e.g. "chief statistician" vs. "assistant chief statistician"
I checked with ChatGPT about this, and there does indeed seem to be an effect that doesn't wash out at high socioeconomic levels. They cited "Health inequalities among British civil servants: the Whitehall II study" https://pubmed.ncbi.nlm.nih.gov/1674771/
The ChatGPT conversation is in https://chatgpt.com/share/694993f7-6144-8006-bccc-1ca31fd1e8ef
Interesting, thanks — I wasn’t aware of the Whitehall II study. I’ll take a look.
Many Thanks!
Sounds interesting, and at least somewhere in a realm of plausibility to my lay ears.
Have you tried to build a model/ simulation to play around with?
Not yet. If you know of prior work along those lines, I’d love pointers.
There is a disturbing possibility I have considered derived from assuming the truth of panpsychism. If conscious states come about from some fundamental property of matter, then presumably there is some material property that corresponds to the valence of the experienced emotional state. What if this property is a field or property of matter that is conserved? That is to say, what if the material correlates of emotional valence work out such that universal emotional valence is zero-sum? For example, the electromagnetic field that is my mind might experience some excitation that corresponds with pleasure, but then this excitation produces an equal and opposite excitation in the surrounding fields (which are also conscious, if on a lower level), producing an exactly equal in magnitude amount of experienced displeasure. Wouldn't this be a catastrophe for any form of utilitarian morality if valence is mapped to physical properties in such a manner that it is universally zero-sum? You couldn't maximize universal positive valence, you could only make certain areas of space happier while making others unhappier in exact correspondence. The only utilitarian morality that could be derived is hedonic egoism, maximizing your own felt pleasure even if it is exactly cancelled out by elsewhere's displeasure, because if the universe is net zero valence, the only thing that matters is "getting yours" and assuring that the fields of valence are the most beneficial to your own selfish perspective, because any higher desires for maximization are thwarted by the universe inexorably summing to neutral value. What else could be said about desired moral projects if emotional valence turns out to be a conserved, zero-sum value? How would one morally cope with such a universe?
one would have no choice, which makes it easier. "morally cope" doesnt mean anything to me, so idk about that.
Why should they cancel out? Why no replicating joy fields?
But to answer in the spirit of your thought experiment: maximize joy in our galaxy, pump the displeasure out into black holes.
this is a common joke in the US Navy submarine service-- the "conservation of happiness," except instead of being a general property of the universe, it is a property of the closed system of a submarine once the hatches are sealed. This "Theory" is typically used to explain why its proponent is being a dick to you.
I don't see why or how that could be the case, except in the trivial sense that there is probably finite matter in the observable universe so everything is technically zero-sum. Are you suggesting there's some kind of "emotional valence" field similar to the electromagnetic field? If the only thing in the universe was a brain floating in the void, would its emotions have zero valence, or would the vacuum somehow be able to experience happiness and suffering to counteract it? What would happen if everything in the universe had the equivalent of pain asymbolia, where they're not capable of having sensations with negative valence?
Basically, my worry is that valence might map to a quantity or field that is conserved or universally zero-sum. It's not that it's a separate "valence field" per se, but rather that there is some recognized physical field that, under panpsychism, happens to be what determines the valence of conscious experience. Do I think this is likely? No. But I worry it is even theoretically possible. If there was only one brain in existence, then if this were true, it's valence would have to sum to zero. How I think this would play out is that there would be one "macro" consciousness that feels some amount of pleasure or pain, and this causes an equal and opposite reaction in the "micro" consciousnesses of the surrounding brain matter that biologically supports but does not consciously participate in the primary mind. So the electromagnetic field that is the mind feels one thing, and the micro-consciousnesses of the grey and white matter that supports that mind feels the opposite. Again, this is pure doomerish speculation, but that's how I'd imagine it would work.
A question to panpsychists is this “would you rather get shot in the leg, the brain or the nearest mountain?”
Your statement falsely assumes that panpsychists don't believe in the existence of separate conscious identities. Panpsychists like Bergson believed that the mind is a receptacle or selector, not a producer of consciousness. Modern panpsychists like Strawson seem to admit that the brain is a necessary substrate for complex consciousness — and that mind is a composite of micro-experiences — and if physicalism is true in a substantive sense, it entails panpsychism.
And it's not just philosophers who believe this, but physicists like Federico Faggin (designer of the Intel 4004 microprocessor) believe that consciousness is not emergent from neural activity but is fundamental to reality. Faggin called his view Quantum Information Panpsychism (QIP), according to which quantum interactions are minimally conscious.
I find it interesting that many physicalists and modern panpsychists agree that consciousness is an illusion. I am not a panpsychist because I find their arguments just as unconvincing as those of Dennett-style physicalists. However, even though I disagree with them, I don't believe you did them justice with your flippant (but amusing) snark. ;-)
For what it's worth, some schools of Mahayana Buddhist philosophy conceived that the basic fabric of the universe had a ground-state of awareness that was innately joyful. Their term for it was "Ālaya-vijñāna" in Sanskrit and "kun gzhi rnam shes pa" in Tibetan. It is to a most fundamental layer of consciousness that functions as the ground or basis from which the other individual consciousnesses and experiences arise — in Tibetan, “all” (kun) + “ground” (gzhi) + “awareness” (rnam shes) + "that which is" (pa). Their view (which I'm probably grossly oversimplifying) is that the mundane pain and suffering separate us from that ground-state awareness. And the impression that I've gotten from some of my Buddhist teachers was that ultimate Nirvana consists of merging our consciousness with that state, so we're humming in the background, formless, with that ground state.
Mahayana Buddhism is quite extensive and varied, so I wouldn't be surprised if some subvariety had the view you give... But what I remember from my Tib Buddhist years was that alayavijñana was the subtlest layer of consciousness that was still samsara, so that nirvana requires breaking beyond that. It is, however, far removed from ordinary pain, and quite blissful in itself, hence the danger of remaining stuck there and failing to take the final leap.
Edit: Hindus basically have the same idea, they call it anandamayakosha.
I believe this is a Yogachara view, and that the Bön have a similar view, but they call it something else.
If all emotional valence cancelling out is my nightmare scenario, the fundamental fabric of the universe being positive valence is my wet dream scenario, if you permit the vulgar extension of the metaphor. I should love to hope that existence itself "feels good" and that even without doing anything all existence has positive hedonic value. But if a net zero universe is pure pessimistic speculation, I worry that this is just optimistic speculation. The fact that people who are experts at manipulating their own emotional valence at the deepest level believe this seems like decent evidence to update in favor of it being true, but I'd like to see harder evidence than just "Buddhism claims this". Still, a comforting thought, nonetheless.
> If all emotional valence cancelling out is my nightmare scenario, the fundamental fabric of the universe being positive valence is my wet dream scenario
I think this is a pretty common experience. You don't need to be an advanced meditator or to have mystical experiences to get a strong intuition.... I'm sure nearly everyone has experienced a moment of flow when everything is just quietly blissful. These moments don't happen because there's something extraordinary out there; the world out there is the same as ever. So the problem can only be stuff that our mind is *adding* on top of our basic conscious experience of what's there. It's not just your dream, existing as a conscious being feels good in itself!
Well, I admit that none of this is falsifiable outside of Mahayana (Vajrayana) praxis. YMMV. :-)
And I never got to that level with ten plus years of Buddhist meditation. But I did attend a Sufi retreat, where after a day of visualizing dancing energy cones and reciting a specific dikr over and over, I felt like I had merged with the energy of the universe. It was profoundly joyful (and it led to my second of two telekinesis experiences). I probably should have continued with the Sufi practices, but not long after I moved to the West Coast and lost contact with that group.
Alright, you can't just casually mention that you experienced telekinesis without elaborating. What did you move with your mind, and what powers (yours or the universe's) did you attribute this to?
I've described these two incidents before on earlier open threads in the discussions of psychedelics and ritual magic. I suspect most of the rationalist tribe consider me to be either a loon or a bullshitter. Feel free to accuse me of the former, but not the latter, because, at least from my internal perspective, these events happened. And there were witnesses. So, I don't think I was either deluded or delusional.
The sufi psychokinetic incident happened at a meditation retreat, after I spent four or five hours sitting, visualizing lotus flowers becoming energy vortices as I recited a dikr (sort of like a mantra) over and over. The instructor had us imagine multiplying these vortices until they were a grid of dancing energy flowers before us. Then she’d bring us back to visualizing just one. I’m not very good at visualizing things in my imagination, but a few energy vortices weren’t difficult. Placing them in ranks was harder, but I was able to create the vague impression in my mind of a legion of dancing energy vortices. As for the dikr, I don’t know whether the word we were repeating had a meaning in Arabic or was just a combination of sounds to create a puff of exhaled breath and vowels whose laryngealization made my vocal cords buzz. The instructor guided our visualization all morning long, so my thoughts didn’t wander from the process.
We took a break sometime after noon. We were told to go outside and walk around the grounds, and not to speak to each other, while a simple lunch was prepared for us. The retreat was held at an old farmhouse in Eastern Connecticut. It was a sunny May day, the ancient apple orchard next to the farmhouse was in bloom, and the red-winged blackbirds were trilling their spring courtship songs. The instructor sat cross-legged on the farmhouse porch, and she pulled the hood of her khirqah robe over her head to meditate. I wandered out into the orchard to enjoy the sun and scent of the apple blossoms. I felt intensely joyful. It was as if I were filled with some sort of divine happiness. I felt a prickling all over my head as if my hair were charged with static electricity — but it wasn’t static electricity. And my whole body felt as if it was channeling that same energy.
I felt sooooo good! I stretched and raised my arms, and the breeze came up. I lowered them and the breeze stopped. The day was otherwise still, but I didn’t think much of it. Then I stretched again, and the breeze came up again. Hmmm. I lowered my arms, and it stopped. I laughed. I raised my arms again, trying to channel the energy within me up my arms, and the breeze turned into a wind. I lowered my arms, and it stopped. I was giggling like a fool. Not really believing it, I raised my arms again and visualized the energy within me coming out of my fingertips, and the wind came up again, stronger. I dropped my arms, dazed at what I had experienced. I looked over at the farmhouse. The instructor had pulled her hood back. She looked directly at me and smiled. But she wagged her finger at me and shook her head negatively, indicating that I should stop.
We performed a different set of visualizations and recited a different dikr after lunch. Nothing fantastical happened after the second session. As I left the retreat, the instructor took me aside and said, “These things happen, but they’re just a distraction. Don’t indulge in them.” That evening, when I closed my eyes, geometric phosphene whirlpools danced across my vision. It was as if I had dosed with psychedelics, but I hadn’t touched psychedelics for a couple of years at that point. I didn’t sleep much that night, but the next day I felt rested and energized — not with mystical energy, but that normal feel-good energy when mind and body are perfectly in tune.
Very interesting, thank you for sharing! I tend to be skeptical, but my mom has had convincing paranormal experiences that I mostly believe her about (I don't think the usual explanation of hallucination from being half-asleep applies to her experiences), so I'm willing to accept that maybe otherworldly phenomena like controlling the wind with your thoughts might, under extraordinary circumstances, be real.
That's an awful lot of assumptions about a phenomenon we know almost nothing about. But even then, consider this analogy: The atoms that constitute our planet are preserved in number and in kind to a very, very close approximation. And yet, if I go out into the forest, pick up a piece of wood, bring it home, and carve a spoon from it, I've become richer by one wooden spoon, without making anyone poorer by one wooden spoon, even though the number and kinds of the atoms in the universe hasn't changed.
Are you proposing it as a thought experiment or as a way reality might really work?
For the purpose of the thought experiment, I guess torture a bunch of lesser moral patients (e. g. shrimp) to make greater moral patients happy. A partialist framework (permitting one to allocate resources to family > friends > proximate strangers > distant strangers) will be okay with this.
I really hope it's just a thought experiment. I'm a hedonic utilitarian who thinks all that matters is maximizing universal pleasure, but I also tend towards panpsychism as an explanation for consciousness, and the possibility struck me that it's at least theoretically possible that the material property that corresponds to experiential valence is a conserved, zero-sum property. I guess you could say that this is my worst-case, nightmare scenario.
It’s an interesting worry, and I think motivates more specifically the idea that pure conscious hedonism probably isn’t what matters if everything is conscious. But even if this is all true, it ends up leading to a sort of nihilism, which is better than the negative utilitarian idea that the world might be much worse for existing than not.
I wouldn't worry about creating too much pleasure. Suffering is bottomless. Happiness is capped. You can be a multizillionaire with yachts and jets, and win the Nobel prize today, but you'll still only be about as happy as a middle schooler with an A+. The only way to make more pleasure is to distribute it to more agents, and even then you're capped at n*h(A+).
>The only way to make more pleasure is to distribute it to more agents
Well, I specifically prioritize creating more moral agents so that more happinesses can be felt than ever would among those that already exist. My ideal outcome is converting the lifeless matter of the universe into as many artificial minds as possible, and then making those minds as happy as possible. So making more people (or technically artificial consciousnesses) is the way I think utility should be maximized.
Email I received on Fatima, from someone who gave me their real name (but I won't post it here). Keep in mind that there were at least tens of thousands of readers who *didn't* report anything like this:
"Hey Scott, big fan of your substack. I am just writing to you because I had read your article on the Fatima miracle (which was very interesting, thanks) about the 'dancing sun' a few weeks ago and ever since then I am 100% sure I have witnessed the phenomenon (or something like it) twice, and I think your conclusion about it being some type of unexplained atmospheric/optical phenomenon seems likely from my perspective.
[I had read your article a few weeks before this, and I was curious if I could see a similar phenomenon, otherwise I wouldn't really have looked at the sun at all besides for maybe a split second (even if it wasn't bright and did not hurt to look at) (and the phenomenon started after only a couple seconds). I did try to mentally 'unsee' the phenomenon whilst looking at it and even looked away to the ground or to the side to see if it would go away but the 'trembling' and shaking continued regardless. However, that being said, I have to admit I did mentally have the Fatima event in mind, so in that case I am not a perfect sample. I am mostly very curious if others can view something similar in similar conditions as I am only one person at the end of the day.]
The first time I witnessed it was in the late afternoon about a week ago, a short while after it had rained (I am currently living in the UK) and the sun had come out. I looked up at the sky from my window (I was inside my house) and noticed that the sun did not seem too bright and could be looked at without immediate discomfort to your eyes. I remembered your article and so looked at the sun for a couple of seconds out of curiosity (yes, dumb I know) and after just a couple of seconds, the sun seemed to 'vibrate' and tremble or shake, and also began to 'spin' (I don't know how else to describe it) at the same time.Throughout this the sun appeared as a flat lustrous shiny silvery disc with a thin but clearly defined border that alternated light and dark. After around 10-15 seconds of staring this way, the sky seemed to become pink, though after a few seconds after looking away from the sun, it would turn back to a normal colour. This lasted for around 2 minutes or so before the sun dipped behind the clouds. Afterwards, I had very mild eye discomfort for some seconds but besides that my eyes were fine.
The second time was much more interesting, and happened yesterday. I was walking through a park, again around only 20-30 mins after it had rained, and I again noticed that the sun was seemed to be a dull silvery disc that was harmless to look at, so decided to try and observe the phenomena again. After barely 1-2 seconds the sun began to shake, vibrate and spin, and had the same border around it with the alternating light and dark shadows on it- the shadow and light would alternate from being at the 'top' of the sun to the 'bottom' of the sun giving it an even greater impression of spinning. After a few more seconds, the sun began to shake faster and faster and looked like it was taking very short and rapid 'jumps', and took on a pink halo. I decided to focus my eyes away from the sun whilst still keeping it in my vision (I thought if I did this the sun would become 'normal' again) but the sun actually became faster when I did this, and it actually began to zigzag in a short but straight line, as in it would move from point A in my vision to point B in my vision and back again continually in a straight line quickly. Whilst this happened the sun actually became black almost entirely except for the edge, which became pink and would emit pink streams of light. When I looked directly at the sun again it would stop doing this but would keep shaking and 'jiggling' from side to side like a ('dancing' could be used as a description though I would describe the sun more like the end of a macarena instrument when you move it in short and rapid shakes except if the macarena end seemed like it was doing a spinning motion at the same time). This went on for some minutes (significantly longer than the first time and the movements were more pronounced), and then I got somewhat bored and looked away for a while and by the time I decided to look back the sun was behind some clouds again and was glaring again. I felt no eye discomfort during or after the 'event' at all.
For some context, both of these 'events' happened during completely non-religious occasions as (obviously) described above. I am not a Christian nor do I come from a Christian background (my family background is actually Hindu but I am non religious myself). I didn't see the multicoloured lights phenomenon you described in the article nor did the sun ever seem to come down to Earth, but I definitely saw the spinning, trembling, shaking, short 'jumping' (though this only for a short while) and it 'jiggling' around from side to side even when I wasn't focussed on the sun at all.
Very sorry for the long email/story as you are likely a busy man but I just thought it would be interesting to share my experience, especially as it suggests to me personally that your conclusion in the article is likely correct. Both events happened to me some time (by which I mean around 10-30 mins) after it had been raining and it was still very wet/damp (and it seemed to have rained just before the Fatima event as well). As a disclaimer, I had read your article some time before witnessing this (so priming is not out of the question) but I did try to 'unsee' it whilst witnessing the phenomenon but the sun kept shaking/trembling despite this mental effort/awareness and regardless I would never have thought to look at the sun in a million years in any context without this article. "
Interesting email. I've often looked at the sun when it's cloudy, where it looks like a silver disc and doesn't damage your eyes to look at it, but I've never had any of these intriguing visual effects reported by others.
Fun fact: you can get a similar type of phenomenon with a campfire by looking at a bed of hot coals after a fire has burnt out. Similar type--different presentation. It looks like a little city where fire-energy is being transported around, more or less. I think it's possible it's the same biological phenomenon
I've gotten a similar effect looking at the full moon on very clear nights.
I tried it with sun through clouds in North Carolina and Alaska. I tried it with a cell phone light, and a candle, and a ceiling light. Nothing.
1. Would be nice if we could get approximated GPS coordinates so we could go back and look at weather logs.
2. Maybe this is some sort of optical illusion? If the clouds are moving or swirling, and your eyes have no other reference point, you may think the sun is wobbling. (And maybe it doesn't happen with the moon because it doesn't illuminate the clouds nearly as much?)
3. Stars at night can appear to wobble (and twinkle) due to a warm and turbulent atmosphere. You can see this easily through even a weak telescope any time of year. Maybe things are warm and turbulent enough during a daytime storm to be more visible. Combine that with a cloud cover so you can see something (the sun) all the way through the atmosphere (which you normally can't see anything during the day all the way through the atmosphere) and you get this.
I don't know if you already thought about this, but the jittering and zigzagging motion of the sun might have to do with microsaccades? Apparently the wikipedia page says that they occur during "prolonged visual fixation (of at least several seconds)".
It's weird that the atmospheric conditions were similar to Fatima in this case too, perhaps it really is a combination of biological and atmospheric phenomena; that would explain why it seems so rare, and why it's not studied: it doesn't fall neatly in a single scientific discipline "basket", so opticians/perception neurologists attribute the miracle to meteorology and meteorologists attribute the miracle to optical phenomena or perception.
Could be the combination of saccades and an extremely bright target.
If you relax and fixate on cracks in the ceiling (say) they will appear and disappear because your eyes over time see only things that change. But it might be expected that if you can do that with the sun (and it is probably not recommended) more dramatic effects could appear.
This makes sense to me. Deep meditators sometimes have their visual systems start to act funny. (In Zen, the koan "Make the mountain dance," points to this.) The world isn't actually as stable and solid at it appears — we "see" light waves which are constantly in flux. If something takes out the functionality of the visual system, we might see the world as jumping, jittery, jagged. Staring at even a clouded-over sun may overwhelm nerve pathways and permit the visibility of what was never stable in the first place.
Yeah, but come next week the Court will prob. be dealing Trump a major blow job.
> If we save vintage Americana, we save the American Dream. If we save the American Dream, we save America. If we save America, we save the world.
I think this needs more justification than you're offering. It seems false to me; I think to "save America" you'd want to think about inequality, cost of housing, AI, the national budget, climate change, and of course political polarization.
Do you use AI to help with composing this? When did you start, and how do you find it helps you?