domain:philippelemoine.com
Pasha
Defend Kebab
7hr ago
My benchmark is an actually good language learning tutor LLM. The task seems pretty much perfect for the current crop of LLMs, and there is infinite VC funding for literally anything that involves LLMs. As long as something so obvious isn't mastered yet, LLMs are hardly more than a gimmick imo.
SkoomaDentist
7hr ago
Meh. That just sounds like a collection of shitty djent cliches.
chooky
7hr ago
But children of the 90s are like 40 now and would have also grown up entirely under the post-1970s paradigm, while the rise of incel culture (and various other apparent symptoms of dysfunctional romance) seems like a phenomenom of the past 10-15 years. I am having a hard time ascribing this to the 1970s rather than technology shifts (Tinder, etc), high pace of housing inflation (which reduces incentives for household formation and makes it much harder to not rely on also-expensive daycare, aka the two-income trap), or the transition of church and religion out of mainstream (which I would argue began to rapidly occur during Bush 2 and was basically complete mid-Obama).
rae
7hr ago
I’m really struggling to understand how having a hard time finding a girlfriend can make someone want to dress up as an anthropomorphic animal and have sex with other men in similar costumes. I can understand having those urges, getting into a relationship and it being too embarrassing to share so you just suppress it (although I’m sure many still explore them in porn), or being single because you have non-standard sexual interests and can’t find someone that fulfils you.
I do get that the internet/porn amplifies underlying fetishes and makes you seek more extreme stimuli, but I don’t think it can make a straight man gay or vice-versa, or a vanilla person interested in furry fandom.
Mantergeistmann
8hr ago
That's... remarkably similar to what I was taught; same time period. And I moved around the country and went to both public and private schools, so it wasn't just localized.
2rafa
8hr ago
It’s flawed for sure, but there is substantial anecdotal evidence that the percentage of men willing to engage in homosexual activity, especially in substantially or entirely male communities (men at war, men in prison, all-male boarding schools, male-only religious institutions) is probably higher than the 3-5% estimates of gay men.
2rafa
8hr ago
Many male porn addicts seem to be in sexually active relationships though. Besides, I don’t think it’s clear that men with trans ‘girlfriends’ couldn’t find female partners, that seems spurious.
FistfullOfCrows
8hr ago
Tell me you haven't listened to The Alex Jones Prison Planet without telling me etc...
2rafa
8hr ago
My understanding of the data we have on sex and partner count is that you had the sexual revolution in the 60s, which took until the ~late 1970s/early 1980s to filter down into mainstream society. From that point (ie the youth of Gen X) everyone has been having pretty similar amounts of sex. Millennials weren’t much more promiscuous than GenXers, and Zoomers are as or less promiscuous than millennials.
The emergence of apps, online dating, social media, none of these seem to have substantially affected population-level promiscuity, only shifted it. The (heterosexual) people hooking up with dozens of people on the apps are the kind of people who would hang around dive bars and clubs until closing time to pick up the best option left had they been born twenty years earlier.
I think it may be different for gay men, although large parts of that are surely increased social acceptability and the fact that HIV is no longer a death sentence, but even then, my guess is many people racking up 4-digit grindr body counts would have been anonymous bathhouse regulars back in 1977 too.
kky
8hr ago
One of my little hobbies is examining historical evidence and trying to make a reasonable guess at the truth of things. I thought an example of such might be fun enough for people here, so here you are.
Bret Devereaux writes of Roman weavers:
That said, while the production of clothing was an essential task, it was not a well-remunerated task. Regular weavers – not specialized in rare or fine fabrics – are some of the least well paid individuals in Diocletian’s Price Edict, paid just 12-16 denarii per day (20-40 for those working high quality linen, 25-40 for those working on silk), compared to 25 denarii per day for an unskilled farm laborers, mule drivers, shepherds and 50 or more for skilled artisans working wood, stone or metal (Carpenters: 50; mosaic workers, 60, wall painters (fresco, one assumes): 75, shipwrights, 50-60, blacksmith or baker, 50, etc.).
First, the reason this passage drew my attention: that an at least partially skilled laborer could draw half the wage of a farmhand does not pass the smell test. Why would such individuals not simply up and leave? The farms await with their great bounties. Such a discrepancy demands explanation. Bret attempts one, but his is that capital ownership was much more important in the labor-rich premodern environment, and that therefore the earnings of weavers could be driven down. This explains nothing of the discrepancy between weavers and presumably equally disenfranchised hired hands. So what could account for this?
First, background on Diocletian’s edict. Normally we think of price controls as a minimum cost or wage for this and that, typically as a socialist dictator’s ploy to stay popular. Diocletian was setting a maximum to try and halt inflation. So in each of these, we should consider: given an environment where labor is in a relative position of strength, Diocletian forbids the worker demanding a wage beyond a certain amount. If labor is not strong, and currency is not too greatly debased, then we should expect actual prices to stay lower than these marks, or else for wages to float beyond them on the gray market.
In the edict itself, however, low-grade weavers were not paid by the day, but rather by the pound. This is likely what is generating Bret’s estimate range here, as it’s hard to know exactly how much a weaver can weave. But there’s another confounder here, which is that the edict does not specify the quality of the weaver, but rather the quality of the wool, which is coarse. Wool’s weight per yard is not fixed, but varies on the thickness of the thread - so a weaver using coarse thread is simply going to be producing more pounds per yard and per unit of labor than one using fine thread. Flax is finer than wool, and presumably is going to be priced higher per pound.
And since our estimates on historical productivity are at best sketchy, we really can’t rely on our figures here. Modern estimates are typically given by historical reenactors. Not to put too fine a point on it, these are amateurs and historians who are bookish and unlikely to be either driven or particularly skilled with their hands. Premoderns, on the other hand, were going to eat or starve based on their productivity, and starvation was not so very far away. They would be working hard (perhaps 12h/d instead of 8 max), and with no end to practice, and likely with more dexterity than book learning. We should expect their productivity to be substantially greater than our contemporaries. And, given that we know Diocletian was trying to set reasonable price caps to halt inflation, we can assume that he was working off of estimates to keep the overall income of these similar workers in line. The farmhand shares wages with a water carrier and a mule driver. The equivalent for the wool weaver would be the day-wage linen weaver, who made 20d. That should actually make us strongly suspect that a weaver of coarse wool fabric was making a pound and a third, or a yard and change, of cloth a day, rather than making much less money. (Also interesting: women make much less a day, down to 12d. Was this because women worked slower, because they were expected to work part-time alongside childcare, or because they were understood to be exceptionally vulnerable without a working man and therefore easy to exploit? The wage gap persists.)
Lastly, a couple considerations on the nature of the work. Farmhands are presumably not sharecroppers, but rather hired help during the backbreaking and urgent plowing and harvest seasons. They would not be needed the rest of the year. In contrast, given that thread does not spoil like food, a weaving workshop can operate year-round and would likely prefer to so distribute the work in order to fully utilize workspace and looms. So our farmhand is hired for a few weeks of brutal but reasonably paid work, while our weaver is steadily employed throughout the year. Even though the farmhand likely picks up additional work to cover the gaps, the needed rest after these periods means he is all but certain to average below his sticker price. How much lower is a hard estimate, but 20% of the time out of work is sufficient to bring his wage down to the linen-weaver. And we have to assume that Diocletian was completely aware of this fact.
So, in summary: looking at the actual numbers and the actual purpose of the edict, alongside some reasonable assumptions about the comparable nature of the work and our own limited ability to produce, yields a plausible interpretation of the evidence where the astonishing anomaly of Bret’s assertion that weavers were paid like women (on its own an anomaly) vanishes. And this is a technique, for what it’s worth, that Bret has used himself for things like military equipment weight, so I think he most likely is just less familiar with this field and took someone else’s uncited estimate as gospel instead of examining the strange details like he would for military matters.
If something doesn’t make sense on the roughest estimates, that’s almost always because one premise or another is false or misunderstood. This mutation of syllogistic reasoning holds quite broadly.
hydroacetylene
8hr ago
There were two sexual revolutions and they both had major social instability coming in about forty years later. Can I explain it entirely? Not really. But the 2010's social chaos occurred roughly the same timeframe after the sixties/seventies sexual revolution as the sixties/seventies social chaos occurred after the 20's/30's sexual revolution. Perhaps we'll see in the 2050's some chaotic results of something LGBTQ related, history rhymes.
As for the forty years timeframe, I would suggest that it's when children who grew up entirely under the new paradigm are reaching the age to start making bad decisions. Social change is slow even if at the surface level it looks like lightning.
hydroacetylene
8hr ago
And that most people ate them proportionally less often, you had a balogna sandwich if you couldn't cook a meal for whatever reason. That's not to say their home cooking was very good- it often wasn't- but there was a lot less of the other kind of any description. 50's purchasing power was just too low, and technology too backwards, to sustain any alternative to cooking.
Composing and recording music is already dirt cheap.
There's at least, like, one semester of instruction required to learn a DAW. I can teach myself C++ and how to sys admin and use a DJ mixer. But a DAW just feels too hard without hand holding, same as Photoshop for me.
Being able to prompt Suno and get something fun to listen to (albeit flawed) feels liberating.
Is this going to be akin to that study about how good programmers working on mature projects did not save time using AI tools even though they thought they did?
Bingbong
9hr ago
I've had a few friends kinda dabble in furry/gay/whatever culture whilst effectively 'incels' during University which generally went away as soon as they managed to somehow find a girlfriend. I also think that the whole transfetishist movement is a tad bit different to homosexual inclinations on part of a decent chunk of society, or atleast there's a feedback loop that's emphasizing it at present.
rae
9hr ago
I think the opposite explanation is far more likely, until recently the pressure to be in a hetero relation was extremely strong and any “divergent” behaviour was kept tightly under wraps. The Kinsey reports from the 1940s found 37% of males had at least one homosexual experience, 11.6% were about equally bisexual and 10% were more-or-less exclusively homosexual.
I’ve known plenty of gay and bisexual men and none fit the profile of “watched too much porn, couldn’t get a girlfriend”. Gay men just have completely different innate sexual appetites, and lots of bi men are closeted and cheating on their girlfriends and wives. The closest thing to what you’re describing would be bisexual men choosing to hook up with men because it takes less effort to organise than ordering from DoorDash, or men dating trans women because they’re more “chill”, but I can’t believe being an incel for long enough will make you want to shove a penis into your mouth with as much enthusiasm as these guys had.
orthoxerox
If you can read this, you're using a custom theme
10hr ago
The Gestrals, I guess. And Esquie.
ArjinFerman
Tinfoil Gigachad
10hr ago
3. They're deliberately distracting people from the issues at home
Didn't cross your mind? It especially makes sense when it's a public broadcaster.
RandomRanger
Just build nuclear plants!
10hr ago
Interesting story. I feel as though it's too well-written to be realistic. All these people are making smart moves, only to be checkmated by a smarter being.
Apparently Microsoft only just stopped outsourcing Pentagon IT to China: https://www.cnbc.com/2025/07/18/microsoft-china-digital-escorts-pentagon.html
The US military is not going to be making smart moves regarding human-in-the-loop AGI/ASI control loss situations. They don't make too many smart moves elsewhere in their area of expertise, developing weapons and winning wars. The Iraq-Afghanistan bunglers are still in power, there was no military purge or anything. They don't know how to win wars and they're not great at procuring weapons either - Constellation, LCS, Zumwalt... Why would they do a good job with AI, a seductive and dangerous, unknown and unprecedented technology?
It's like how back in the old days, people were concerned with how a powerful boxed AI might escape the box to interfere with the outside world. There's no boxing, there are huge companies working around the clock to give powerful AIs more access to the outside world! And in China they don't care about AI safety at all, they just laugh and move on.
Law of undignified failure: https://www.yahoo.com/news/watch-ed-sec-calls-artificial-184225319.html
President Donald Trump’s education secretary repeatedly referred to artificial intelligence as “A1”—like the popular steak sauce—instead of “AI” during a panel on artificial intelligence in education.
“I wish I could remember the source, but there’s a school system that’s going to start making sure that first graders or even pre-Ks have A1 teaching every year starting that far down in the grades,” Education Secretary Linda McMahon, 76, said at a summit put on by Silicon Valley investors in California on Tuesday.
Making the situation even more bizarre, McMahon pronounced “AI” correctly earlier during the same panel discussion. The Education Department didn’t respond to the Daily Beast’s request for comment on the secretary’s apparent mistake.
Ioper
11hr ago
I've been playing the game for quite a bit in early access and while I enjoy it well enough this does feel like "safe horny" the game, but extended to every single character and interaction. Everything is so HR approved and "safe" that I find myself just skipping through all the dialogue. Nothing is said and while it's mildly witty at times it's never funny.
I think a major issue is that they've replaced many of the major roles with women for the sequel and they don't think it's ok to make jokes at the expense of women so what we get is a whole load of anodyne nothing.
striker
12hr ago
Small observation, struck me as interesting. I’m in Ireland. In the car this morning, radio on. The top of the hour news came on (public radio, state-funded). Top three stories:
- Gaza situation, MSF and WHO statements
- Iran nuclear situation, new sanctions
- Denmark drones situation, new sightings over military bases
There’s plenty happening in Ireland (presidential election campaign, housing crisis, Budget 2026 to be announced next few weeks)
Two competing thoughts:
- We (and maybe many EU countries?) have a very broad perspective on the world, and this is generally a good thing for Ireland’s place in a globalised/interconnected world.
- Our state broadcaster is sourcing its news agenda from international wire services, probably with many political biases intact and unquestioned, and doesn’t fund enough of its own journalism and newsroom anymore, and this is generally a very bad thing for everyone in Ireland over any time horizon.
phailyoor
12hr ago
this looks and feels like a school shooting to me: no apparent motive beyond causing mass hysteria, which judging by this thread, was very much achieved.
Do you know what would cause even more mass hysteria? Probably gunning down normal people or alternatively important people who are popular. But instead he chose to try to gun down Satan's goons, the empire's stormtroopers, the left's greatest boogeyman and people nearly every leftist hates and despises.
Somehow I am not convinced that the guy just wanted to cause hysteria and had no political motive.
Southkraut
The rain fell gentlier.
13hr ago
Personal Assistant.
"AI, please do my taxes, handle the car damage insurance case in a maximally profitable way, and tell me which area of work to focus on today. Also, which one of my friends have I neglected to catch up with the longest? Remind me to call them this evening, but not too late."
ActuallyATleilaxuGhola
Axolotl Tank Class of '24
13hr ago
"Alice's Adventures Through the Windshield Glass"
Maybe AI will do what Ron Paul failed to do and make anime real.
More than 3-5% is is very far from 37%
More options
Context Copy link