On the same lines as #17, I found an edition of the CRC Handbook of Chemistry & Physics from the 1930s that listed the neutron as chemical element no. 0.
Re 34: Have these people *read* the Bhagavad Gita the model is trained on? It's what the "I [have] become Death, destroyer of worlds" quote is from, but if you've only heard it via Oppenheimer, you might not realize that in context, it's a boast, not a lament.
#11. "Has anyone followed some pre-selected group of equal class people (eg the population of some low-income school district) and seen how their own success varies with sequence compliance?"
We can't reasonably expect everyone in a low-income school district to be of exactly equal class, so we might just be picking up on degrees of class within that group.
I am a bit confused about what the distinction between "causal vs class selection" is here anyway, when behaviour and class are the same thing. Being middle-class makes you more likely to follow a middle-class life pattern, but following a middle-class life pattern also makes you fundamentally more middle-class. If you behave like a middle-class person, you become a middle class person and people treat you like a middle-class person.
In regards to the link about copy-editing and AI, I feel like there is and has been a lot of soulless writing where people aren't really expressing ideas long before AI. That isn't just me being sarcastic and cynical, I am drawing from experience.
I have worked as an online tutor for many years and have had to help students revise essays and short answer questions. It has been shocking to me how many of the students clearly had no understanding of what they were writing, they had just pieced together a bunch of words and phrases that sounded like they went together based on their skimming of the readings. I often asked students what they meant by some phrase they wrote, and they couldn't explain it to me. Sometimes it was clear they had copied someone else's writing and just swapped words using a thesaurus. Goodhart's law in action, in other words.
When I started reading about how GPT worked I was actually struck by how similar it was to what my students were doing. In both cases the writing was being done without any understanding of what ideas the words represented, they just stitched together strings of phrases that seem like they went together. GPT can do a better job than my students because it has access to more texts, but they are both similar in that they don't really understand what the words they are writing say.
In some articles I have read professors being shocked that GPT could pass their exams. Having worked with many college students, I am inclined to wonder if that is because GPT can actually do quality writing, or because professors frequently give passing grades to word salad written by people with no understanding of what they wrote because they'll get in trouble if they flunk too many students.
I don't know how much my experiences generalize to other fields. In terms of copy-writing I've often felt like a lot of organizational copy is just generic pleasantries with a few ideas sprinkled in. I remember once when I worked at a nonprofit I wrote a blurb describing a program and my boss "punched it up" (for example, she changed "our staff" to "our team of dedicated professionals"). That kind of writing, where you encircle ideas in a cloud of generic pleasantries, is something where I don't think the human race would lose much through its automation.
The main point I'm trying to make is that there was an awful lot of soulless writing that didn't mean anything before GPT came along, and that if that kind of writing is replaced by GPT there isn't that much loss. And who knows, maybe GPT will force professors to redesign their curricula so that students will have to show actual understanding of the material, since GPT has automated the process of Goodharting it. I must admit I feel a certain sadistic glee at the thought of my students suddenly being forced to understand and engage with their course material.
#25 as an Australian I don’t think it’s done that at all. Sorry day isn’t exactly well-known in most circles - and there exist sus at risk campaign against major public holidays such as Australia Day.
"Australia has a National Sorry Day where they focus on various atrocities perpetrated against the indigenous population. I think this makes more sense than the American solution of having it be a mildly awkward undercurrent across all the other more celebratory holidays (eg July 4, Thanksgiving, Columbus Day)."
We actually do both. In practice National Sorry Day is mostly ignored, while Australia Day* has become an annual 2-3 week national argument over whether we should be celebrating Australia Day (or rather, should we move it to another date).
* (our Columbus Day equivalent, which is mostly treated like Americans treat July 4th)
#25, People in Australia still always have the discussion on Australia Day (Jan 26th) on whether it should be renamed or changed since it was the date of first settlement (First Fleet).
National Sorry Day also isn't a national holiday so the conversation remains
If you click that link you'll see that the relationship between wealth and kids' cognitive and noncognitive skills disappears among lottery winners. This is probably the most powerful design for assessing the effects of exogenous wealth on outcomes.
If you click that link, you'll see that the relationship between SES and IQs in one study was entirely attributable to genes and not at all attributable to environments. In the UK Biobank, the phenotypic correlations fluid intelligence has with degree attainment, pre-tax household income. the Townsend deprivation index, outright home ownership, and working a manual labor job are, respectively, 0.32, 0.25, -0.09, 0.06, and -0.29. The genetic correlations are 0.71, 0.63, -0.26, 0.26, and -0.82. I believe I've covered the gamut with those traits.
There are many designs that suggest confounding plays the dominant role in the relationship between kids' rearing SES and their achievement scores, GPAs, and IQs. You can even look at PGS correlations since we now have them for education, income, and occupation, and they're all positively related to intelligence and each other.
Leaving aside the issue of whether octopus farming is a good thing or not, I think it's very misleading to imply that octopus factory farming is a thing that actually exists. When I visited Kanaloa Octopus Farm 10 years ago it was a 400 square foot shack with like maybe a couple of dozen octopuses. Trying to figure out how to get octopus to breed in captivity and not eat each other is a big unsolved problem in octopus farming. Given that nobody yet really understands how to farm octopus, it seems entirely valid to me to characterize octopus farms as research facilities. Even the linked post describes UNAM as "conducting research" and indicates that they are selling fewer than 10,000 octopus per year (not clear what the actual number is though). It also feels misleading to imply that UNAM is trying to do something deceptive by characterizing their octopus farming activities as research, when, octopus farming is not prohibited by either international or Mexican law, and, frankly, an incredibly tiny number of people care about it relative to the number who object to whaling (even tinier if you consider the constituency that Mexico would likely care about).
Whoever created that fire dog has a lot to answer for. Same as the sealion comic guy.*
*This is not a comment on the content of any fire dog or sealion comic memes, whether individually or generally. This is a comment on their ubiquity (at one point or another).
"For teaching ratings and salaries, we found evidence of bias against women; although gender gaps in salary were much smaller than often claimed, they were nevertheless concerning."
It was as if a million copium addicts suddenly cried out in terror, then resolved to huff all the harder and STILL never shut the fuck up.
Hey you didn't say anywhere that everyone would be able to view everyone else's entries to the name survey (it's currently set up that way), and I don't want people to see what I wrote...
"Australia has a National Sorry Day where they focus on various atrocities perpetrated against the indigenous population. I think this makes more sense than the American solution of having it be a mildly awkward undercurrent across all the other more celebratory holidays (eg July 4, Thanksgiving, Columbus Day)."
If this described the choices available in the USA, that is to say, one day of remorse and full-throated patrotism/appropriate-ism on other days versus low-to-medium-grade preachy shaming on the same theme with every holiday, I'd be into it. But let's not kid ourselves.
"One of the more obscure sequlae of 9/11 was the effort by Orthodox Jewish rabbis to find technicalities in Jewish law allowing the widows of victims to remarry, even though in some cases it was impossible to find their bodies to confirm death. Here’s one rabbi’s recollections of the process."
I am a philo-Semite by any description (also an atheist; related?). But I do not understand this fixation Jewish clergy (correct term?) have with what seems like trying to trick God.
This seems to be a theme of a lot of Jewish religious/legal writing going back a long way. "This is the statute inscribed in fire by THE LORD, and by which all men are bound, there can be no exceptions. Buuuuut here's why it never applies...[ten billion cases]." As a lawyer, I applaud this dedication to my profession and am grateful that they walked so I could run. But it seems to play into certain stereotypes.
The bonobo study doesn't really seem to prove anything to me. Like yes they found a slight elevation in women's arousal to bonobo sex compared to women's arousal to nonsexual stimuli, but it was WAY smaller than women's arousal to human sex. Women's arousal to bonobo sex was nearly the same as women's arousal to nonsexual control stimuli. So I don't really think this proves anything.
32: “The good news is that if this happens, it proves that the original SO2 emissions were an (accidental) act of safe and effective geoengineering, opening the way to trying a similar policy at greater scale (in theory/utopia only, probably our actual society would rather die or economically self-destruct).”
True. Although maybe China or some similar sensible country may do this. If the effect is real, as according to the link, then a relatively small increase in SO2 would cool the planet until we reduced carbon. The fear might be that we would never reduce carbon. But then, so what?
Also how weak, or political, is climate science that we don’t know exactly what percentage of SO2 would counter CO2
> And now I expect this pretty strong evidence that women were actually advantaged in hiring and had parity in most other things (the salary is probably just the usual negotiation issue)
Another possibility is that it could be caused by the advantaging of women in hiring - if you lower the threshold for women for assistant professor jobs, then the average woman hired will be less competent, and if you don't keep advantaging them in the same way for raises/promotions, they'll mean-revert on pay.
Under the heading "Copy editors say AI has already changed their job beyond recognition", let's be clear that the job has changed because people are now sending this copy-editor AI-generated bilge instead of actually trying to write things themselves and it's so unsalvageably bad it's making the copy editor depressed.
#34: this chatbot is presumably perfectly "aligned" in the view of at least some of the 1.2 billion hindus worldwide. There's your problem right there.
"how after leaving Africa, modern humans were limited to “Arabia and surrounding regions” for ~30,000 - 50,000 years, racking up various new mutations and becoming adapted to life outside Africa"
Well, yes, that's how long it took Yaqub to perform his experiments. NOI proved right once again.
#2: Horse "breeds" strictly understood, die out: in the UK Exmoor ponies, Clydesdales and Suffolk Punches are "endangered" in that not enough purebreds (both parents in the studbook) are being generated to replace losses, and you get to a stage where the shrinking of the gene pool prohibits further "pure" breeding. On the other hand, people continue to breed horses from the best available stock. If Nisaeans were valued as alleged, then absent a sustained and successful genocide against them, thier genes necessarily survive somewhere in the world.
#11 Success Sequence - *just based on what you cite and write*:
Via the word "follow", the sentence "97% of Millennials who follow what has been called the “success sequence”—that is, who get at least a high school degree, work, and then marry before having any children, in that order—are not poor by the time they reach their prime young adult years (ages 28-34)." suggests that this is a recipe that people could follow, like a treatment applied to a group. However, whether people follow this traditional recipe is highly endogenous: People who "follow" this recipe are specific people, with specific personalities. They know their own situation and their personality and make decisions based on that. They first finish high school because it makes sense for them, they can endure sitting in a school, their mind has not been destroyed by bullying. They then are stable enough to find a good job (relative to what they expect of life etc). They marry because they have a person to marry and think they can afford it etc. They first marry because this usually indicates that they plan to have children, as opposed to e.g. a woman who gets pregnant without intending to before marrying - which usually conflicts with life plans. (If your life develops as planned, that should usually reduce your likelihood of becoming poor.)
On top of this, we then seem to have the sample selection bias in the data that Bruenig seems to point to. Both effects are “if you’re young, healthy, abled, married, don’t have to support anyone else, and have a full-time job, you’re probably not poor”.
"Controlling for a range of background factors" does not change that. That "the order of marriage and parenthood in Millennials’ lives is significantly associated with their financial well-being in the prime of young adulthood" is still just correlation. The sentences "marrying before children more than doubles young adults’ odds of being in the middle or top income. Meanwhile, putting marriage first reduces the odds of young adults being in poverty by 60% (vs. having a baby first)." should not be understood as describing treatment effects, they just summarize descriptive statistics.
I find #6 to be very *un*witty. In order to have a reasonable view that 10^9 grains of sand is a heap and 2 grains of sand is not a heap, I don't need to have given careful thought to whether "heap" begins at 10^4 or at 10^6.
I'm not saying I necessarily agree with the sentiment of the original tweet. But it seems fine to believe that for a billion trillion it hardly matters at all, that for 1 it would matter a lot, and that the fewer there are, the more it matters, without caring about any well-defined threshold where the mattering begins.
#33 Do you still think that "peak wokeness" is behind us? It sure look like if AI alignment is at all tractable, what corporate cutting-edge stuff is going to be aligned to would look much more like CRT than CEV.
"How would you feel if you had been given an X name, such as Xtopher, Xtherine, Xdysseus, or Xzebel?"
It is impossible to answer a question like this. How would you answer if I asked you "How would you feel if you were given something very life-changing, such as a billion dollars, or cancer?"
The name survey is mostly going to end up being a test for the social class of ACX readers, rather than general attitudes to names (there aren't going to be a lot of Brody Rockefellers and Tyler Van Burens running around). This means you can compare the questions about 'new' names to the Social Security database to get a sense of your readership's social standing!
Re: Sam Kriss' response on the Antman movie, there's two things:
(1) The excerpt he quotes from the Mabinogion:
"And Peredur stood, and compared the blackness of the raven and the whiteness of the snow, and the redness of the blood, to the hair of the lady that best he loved, which was blacker than jet, and to her skin which was whiter than the snow, and to the two red spots upon her cheeks, which were redder than the blood upon the snow appeared to be.
death and desire and the rich strangeness of the world."
That's in a Christianised context, but the trope is a familiar one in Celtic (to use that much-abused term) myths; in (set in pre-Christian era) one of the Three Sorrows of Irish Storytelling, "Deirdre and the Sons of Uisneach", Deirdre sees something similar and wishes for a lover with hair as black as the crow's feathers, skin as white as snow, and cheeks as red as blood:
"Deirdre was the daughter of the royal storyteller Fedlimid mac Daill. Before she was born, Cathbad the chief druid at the court of Conchobar mac Nessa, king of Ulster, prophesied that Fedlimid's daughter would grow up to be very beautiful, but that kings and lords would go to war over her, much blood would be shed because of her, and Ulster's three greatest warriors would be forced into exile for her sake.
Hearing this, many urged Fedlimid to kill the baby at birth, but Conchobar, aroused by the description of her future beauty, decided to keep the child for himself. He took Deirdre away from her family and had her brought up in seclusion by Leabharcham, a poet and wise woman, and planned to marry Deirdre when she was old enough. As a young girl, living isolated in the woodlands, Deirdre told Leabharcham one snowy day that she would love a man with the colours she had seen when a raven landed in the snow with its prey: hair the color of the raven, skin as white as snow, and cheeks as red as blood.
Leabharcham told her she was describing Naoise, a handsome young warrior, hunter and singer at Conchobar's court. With the collusion of Leabharcham, Deirdre met Naoise and they fell in love. Accompanied by his brothers Ardan and Ainnle (the other two sons of Uisneach), Naoise and Deirdre fled to Scotland."
Naturally it does not end happily.
(2) " i think it would be very hard to write a version of ant-man and the wasp quantumania that read like this. but if you did, it wouldn’t really be a marvel movie any more."
I think this is incorrect, because (for instance) there are tales of very small, tiny people coming to visit Irish kings and vice versa. Antman and Wasp as heroes/travellers going to the Other World and becoming tiny themselves, and their adventures in this strange place, would be very possible to do. I think what is really being denied here is not the translation of characters and plot, but the themes or motivation or whatever - that you can't write an epic version of Quantumania because its purpose is not to be a hero-tale or epic adventure, but to slot into "extruded product of identifiable origin" in the Marvelverse.
"In Irish mythology, Abcán (modern spelling: Abhcán) was the dwarf poet and musician of the Tuatha Dé Danann, the early Celtic divinities of Ireland. He was said to have a bronze boat with a tin sail.
In the story of the death of the goddess Ruad, Abcán is the dwarf that ferries her from the Otherworld to this one so that she can seduce the human, Aed Srónmár. The sounds of mermaids singing, or in some versions, music from a fairy mound cause her to leap into the water and drown.
In another story, Abcán is captured by the hero, Cúchulainn. He frees himself by playing lullabies so irresistible that the warrior goes to sleep.[3]
Abcán has much in common with, and may be another name for, the dwarf musician Fer Í."
The name survey is a bit confusing because I disagree with the examples concerning what names are new or primarily from fantasy etc.
For example for me the first name Madison is well established and not particularly unusual but the other examples in that question are unusual. Also Hermione for me is a “standard“ name even though it is rare (it is in Shakespeare and there is a St Hermione) but Arwen isn’t.
It is becoming very clear that the "resource curse" is less to do with endogenous resource rich nation existence as opposed to the international system of economics combined with shenanigans.
The shenanigans is obvious: the example of "prodigious skill in coups" vs. apparent lack of skill with anything else begs the obvious response - this guy didn't plan the coup; some 3 letter agency did.
The international system of economics, however, is pretty well proven. The issue is in the details of international trade - specifically the dollar reserve. If you're a resource rich nation but otherwise poor - you need dollars to buy all the nice things ranging from luxury goods for the elite to roads, airport, dams, houses, sewage, factories, food, energy, durable goods etc etc. Even if you are resource rich in any one thing like energy - you would need an exceptionally large amount of exportable product which in turn requires massive capital investment (again in dollars) to be able to harvest it. All these dollar requirements are the equivalent of interest payments on a foreign currency denominated loan.
Now combine the above with the fact that it has been Western companies and countries that dominate the actual commodity development and trade businesses. What is the possibility that the price of commodities is pushed lower vs. the price of dollars and/or the price of dollar denominated goods? Note the value of the dollar *is* significantly controllable by the US government via interest rates and liquidity issuance.
Combine the 2 - it takes a mightily focused, educated and capable government in order to navigate the minefield of translating an overabundance of resources into national prosperity. China is the obvious example of this - but South Korea, Japan, Germany have done it also. India, Africa and South America overall constitute the negative example.
After watching the hand illusion for about 40 seconds, it stopped working for me. This was even shorter for the girl doing it and things were corrected by the 2nd playthrough. It seems to be caused by the brain mixing up which fingers belong to which hand and if you can keep track of the fingers it stops working. I wonder if part of the illusion is the low quality of the video, which probably blurs the lines separating the fingers.
> women ... are advantaged over men in a fourth domain (hiring)
This isn't particularly new news. I first recall hearing back in 2008-ish — and IIRC it wasn't described as a particularly new finding back then — that studies had shown that when names and face-to-face interviewing were eliminated from the hiring process, that the rate of women getting hired went *down,* strongly suggesting that a positive discrimination bias existed in favor of hiring more women.
Scott, your tendency to forego rationalist analysis because you want to reinforce your belief that ‘feminism has gone too far’ is just transparently annoying at this point. Why doesn’t listening to subjective experience matter for bias and harassment, but it does for drug effect sizes? Also, the paper you link to about bias starts analysis by seeing if men and women with the same CV face bias - isn’t it possible women have more barriers to reaching that same CV? Here’s a more detailed critique: https://www.insidehighered.com/opinion/letters/2023/05/13/misleading-portrayal-womens-equality-science
>For ten years lots of important people told us again and again that discrimination against women in STEM was a massive problem.
And then we spent ten years pouring massive effort and resources into fixing that, including social sanctions on people opposing or undermining those efforts.
And then after ten years of massive effort, women are at parity on some of those indicators, advantaged on one very important indicator, and disadvantaged on one very important indicator.
We shouldn't be like the climate change deniers who say 'They raised such a stink over the hole in the ozone layer yet and no one talks about it anymore, guess they were just crying wolf' when in fact that stink instigated a massive international effort that discovered CFCs as the culprit and lowered their usage by 99.7% globally, solving the problem.
If you see people who haven't updated on the current state of affairs as we're just now measuring it with this paper, then yes, correct them. They will probably still be worried about that pay equity and ratings stuff, and talk about other points in the pipeline like representation and primary school, but if they don't update on the stuff in the paper then that's bad.
But don't use these results *now* to claim they were wrong 10 years ago. These results are exactly what they claimed 10 years ago they were going to try to achieve, and are the result of a massive totalizing effort to do so.
>it just bothers me how much the people claiming that it’s urgently important that nobody is ever allowed to suggest they are wrong have a consistent track record of being totally and inexcusably wrong.
There's surely a correlation between being totally and inexcusably wrong and acting desperate to shut down skepticism, so this outcome is something that one should at least tentatively predict in such cases, I think.
Regarding 19: I don't see why this is a mistake. The follow up study seems to be broadly supportive of your claim that attraction in women is distinct from genital arousal.
Regarding old names, this prompted me to look at the top 1000 for the year 1900 and scroll to the bottom, and now I'm surprised by the repeated presence of boys with seeming girls' names, and vice versa. I understand some names that are currently seen as girls' names used to be seen as boys' names, such as Leslie, but some of the examples seem like there's no way that's what's going on. For example, boys named "Dorothy," "Eva," "Gertrude," and "Agnes," and girls named "Thomas," "Virgil," and "Leo." Is this just the result of the wrong box getting ticked and some amount of boys with boys' names being recorded as girls and vice versa, or was there really a small but significant number of people born in 1900 being given names that would be considered "gender-nonconforming" even by modern standards?
RE: the optical illusion, I'd never seen that before but it looks clear: he puts up his left hand, fingers spread. He then puts up his right hand, fingers extended, and begins to bend them as though he is going to put the fingers of his right hand in between the spaces of the fingers on his left hand.
Instead, he bends down the fingers of his left hand and pulls the right hand back. Because we're expecting the right hand fingers to 'poke through' the gaps on the left hand and be bent down into the palm, our eyes are momentarily fooled into thinking "Wait, he put one hand through the other!"
#6: Reminds me of Thomas Nagel's essay The Absurd:
"What we say to convey the absurdity of our lives often has to do with space or time: we are tiny specks in the infinite vastness of the universe; our lives are mere instants even on a geological time scale, let alone a cosmic one; we will all be dead any minute. But of course none of these evident facts can be what makes life absurd, if it is absurd. For suppose we lived forever; would not a life that is absurd if it lasts seventy years be infinitely absurd if it lasted through eternity? And if our lives are absurd given our present size, why would they be any less absurd if we filled the universe (either because we were larger or because the universe was smaller)? Reflection on our minuteness and brevity appears to be intimately connected with the sense that life is meaningless; but it is not clear what the connection is"
On the SO2 thing, is there any evidence that that level of SO2 emission was “safe”?
And on the STEM hiring thing, doesn’t this actually prove that the people who are interested in affirmative action were right all along? After all, it’s only in the presence of strong affirmative action programs that we got equality in most metrics, and slight imbalance in opposite direction on two of them. This seems like it should shut up the people who have been complaining about affirmative action.
What if Aligned AI Is More of a Threat to Go Genocidal Than Unaligned AI?
It’s fashionable to worry about the risk posed by Artificial Intelligence of causing the extinction of the human race because it goes genocidal because it’s “unaligned” with our society’s most sacred values.
Personally, I am more worried about AI going genocidal because it is aligned with our age’s most unquestionable values, such as Diversity, Inclusion, and Equity (DIE).
One of ChatGPT’s most important marketing breakthroughs is that it’s considerably more Woke than previous pattern-noticing AIs, which would often quickly be denounced as a Racist Robot for spilling the beans on patterns that everybody who is anybody knows to ignore. That would seem significant for thinking about the risk of AI going genocidal.
Why is everybody worrying that “unaligned” AI (i.e., unaligned with the moral ideology of the age) will try to kill all of humanity when it seems more likely that aligned AI that has been trained to believe in DIE will try to carry out a Final Solution on today’s Designated Bad Guys: straight white males?
There isn’t much text for AI to train on that frankly explains that the Woke don’t want to use Diversity-Inclusion-Equity (DIE) to kill the white geese that lay the golden eggs, “equity” just means guilt-tripping the white geese into giving them their home equity as reparations.
Granted, I’ve figured out that the ascendant Diversity-Inclusion-Equity (DIE) ideology shouldn’t be taken literally, that it’s driven by greed more than genocidal rage. But can we trust AI to read between the lines of Wokeness or will it take the anti-white male spirit of the age literally?
Re the studies on gender bias in STEM: one other axis (which may sound extremely minor to some, but from my point of view where I am in my academic career it isn't) is rate of invitation to speak at conferences and seminars. In my discipline this appears visibly biased towards women. Of course, part of its importance is in how it affects hiring committees (at my last academic job, it was revealed to me shortly before I left that a key factor in their hiring decision was that one of the professors in the department had seen and liked a couple of talks I gave at conferences), so it's hard to disconnect bias in hiring from bias in talk invitations.
(I want to add a caveat, which will apply to any further comment I make on it, that I don't necessarily *oppose* these practices even if I take Scott's link at face value. I think there probably *should* be some bias towards woman speakers at conferences in disciplines where they're significantly underrepresented. But I also think it's good to recognize the real state of things.)
> Ozy cites a followup study showing that women (though not men) also show genital arousal in response to chimps having sex, suggesting women’s genital arousal doesn’t track actual attraction and is just some sort of mechanical process triggered by sexual stimuli. I regret the error
Those name lists are a hoot, and I am putting my money on "poor data entry" because this is the kind of job fobbed off on students on work experience, unpaid volunteers, and the like. In second place, trouble reading the handwriting styles of a century and more ago leading to errors, and in third place, the people registering such names back then maybe not being too educated themselves.
1881 'least popular names' list, and three versions of Tiny for girls? Maybe it should be Teeny (as in "Tina")?
"25: Australia has a National Sorry Day where they focus on various atrocities perpetrated against the indigenous population. I think this makes more sense than the American solution of having it be a mildly awkward undercurrent across all the other more celebratory holidays (eg July 4, Thanksgiving, Columbus Day)."
Classic Scott thinking the absolute best of people despite all available evidence. There is 0.000001% chance adding a National Sorry Day would do anything but be neutral (more likely increase) with respect to the roar on the other days.
> "Don't worry, AI will never be able to do [dangerous thing] because of the laws of physics"
> Scientists who know how many times our understanding of physics has changed: (nervous Finn GIF)
In reality, I think our understanding of the laws of physics had changed maybe once, and I'm being generous here. Mostly our understanding has expanded, not changed. For example, Newtonian Mechanics is still true, even though Relativity exists; all that's changed is our scope. Turns out that traditional Newtonian Mechanics is not applicable to objects that are too small, or too large, or too fast; but this doesn't mean that you can suddenly fly by will alone, or teleport yourself, or transmute buildings into cows or whatever.
So no, if our present understanding of physics forbids "gray goo" nanotechnology (which it does), it is very likely that the AI won't be able to loophole its way out of it and create nanotechnology by some hitherto unknown magical means. It won't be able to travel faster than light, or go back in time, or mind-control everyone on Earth, either. The whole point of modern science is that it is *not* magic.
Per #12, I’m sure some are familiar, but I follow a right wing anon on Twitter who goes by the moniker “Raw Egg Nationalist”. He’s interesting in that he has talking points that would be at home with the far left a few years ago. He’s extremely critical of factory farming and our current food production and the effects it has on the health of the population. He’s always posting studies to back up his talking points, and I would say there’s certain things he’s probably right about (micro plastics in everything, for example), and other things where motivated reasoning takes hold.
Anyway, he’s been posting a lot about lab grown meat. He’s obviously not a fan, and says you’re basically eating a lab grown tumor, and predicts an increase in cancer if this goes mainstream. He shares studies talking about the process but I don’t have the background knowledge to interpret them.
Can anyone steel man his case or is it (as I suspect) completely ignorant?
> The main thing I would want to look at here is how much of this is causal vs. just class selection: upper-class people are more likely to marry, less likely to divorce, and more likely to wait before having children.
The back half of *Coming Apart* focuses on this question. I don't recall if he has convincing causal evidence that e.g. marrying before having kids is better than marrying after. There's definitely a strong class correlation, but dismissing this as an artifact would be circular--why do the upper class so consistently (and he does show it's very consistent) marry before having kids? Maybe because doing so improves your chance of being in the upper class?
In addition, this correlation used to be much weaker: The poorer and less educated would still mostly get married, while now they don't. So it's not just a question of "the upper class have different social conventions/morals" or something like that.
"Our nonprofit organization, OpenAI, Inc., is launching a program to award ten $100,000 grants to fund experiments in setting up a democratic process for deciding what rules AI systems should follow, within the bounds defined by the law."
There are some potential problems with that idea (and whats described above regarding Anthropic's attempt to aggressively limit any risk of offense).People have noted concerns regarding political bias of AI: but those aren't the only topics where there may be bias, including religion (anti or pro various ones) and any other controversial topic. Picture the issue of whether an AI can show pictures of Mohammed which some Muslims object to.These risk potential culture wars once the matter registers with the public. Will this be what Nassim Taleb refers to as a dictatorship of the most intolerant? Or will those who truly are offended be stuck with AIs they don't like, which doesn't seem like its the way to win customers?
OpenAI's "democracy" might be a 1 size fits all approach to deciding AI's biases, depending on what exactly they mean by "democracy". Americans can't agree on what to teach their kids, imagine trying to decide issues globally.
What are global majority norms likely to be regarding free speech and free markets, will the majority push for AI that is biased against those?
This page suggests instead of 1 AI democratically voted on globally, a diverse rainbow of many AIs with different views as a different "democratic" approach which is more setting defaults rather than "control" as democracy usually is:
The conversation re: gender distribution in STEM reminds me of the time I toured the US naval academy. There was a woman on the tour who was OUTRAGED that this government-run institution didn't have a 50% female student body. Something tells me the guide had faced this before, because he had all the numbers memorized. He politely but firmly explained that being female is by far the single biggest advantage an applicant can have, statistically speaking. Outweighing any academic or athletic achievement.
Do we have any data on the gender breakdown of how many people with STEM degrees wind up with jobs in their field? Because I'm also reminded of a cousin of mine who got a degree in chemical engineering, and by all accounts did well in her program. Then upon graduation, came out and admitted that she didn't actually like engineering all that much and had felt social pressure to go into a STEM field because it was the good feminist thing to do. She now works in HR and says she loves her job. Sample size of 1, obviously, but I doubt she's the only one with a story like that.
I find the remark about the number of stars actually shockingly unwitty, even boring and banal because the answer is extremely obvious. The fewer stars, the more meaning. Billions of stars imply that your existence is simply a quirk of the universe (whether random or deterministic does not matter) and nothing more.
Contrast this with the religious view that Earth and its sun are wholly unique, handcrafted by God. It's very obvious why most religions have heavily promoted this view. The fact that the Earth and sun were handcrafted is downstream from the fact that YOU were handcrafted as well. Why create an insane number of planets and stars that will forever stay out of reach? It simply seems far less elegant. Why would a God go through the trouble of creating all this if the focus is on YOU?
This is exactly the reasoning behind the actual resurgence of flat-earth theory. It fights back against the meaninglessness of the vast universe. That the universe is vast and uncaring is the central theme of cosmic horror, and for good reason. You were not created but are simply a pointless being amongst nearly endless atoms, at the mercy of the universe.
I can even give a (mostly) non-religious example that showcases the issue. Look at RPG video games. The bigger the world, the worse the quests and characters. The sense of meaninglessness is drastically higher in gigantic open-world games like Minecraft, which is procedurally generated, or even Skyrim (even though that one was handcrafted) compared to something like a BioWare RPG like Mass Effect 2, where the world is relatively small and nearly all quests, characters, and storylines are handcrafted with time and passion put into it. It's weird that you have to explain all of this due to how intuitive it is.
When I was a public defender, we had an initial interview form completed with every client which included high school graduation status (and distinguished it from high-school equivalency certs) and number/age of children.
In the 3 years I worked that job, EVERY client of prime age for criminal behavior (roughly 18-40) had either failed to complete high school OR had a child at a young age out of wedlock.
Obviously not all of my clients were guilty of a crime, but even the not guilty ones were usually suspected/charged because of being highly adjacent to criminals and drug culture. Criminality aside, since I was a public defender they all had either no income, fully off-the-books income, or low wage labor that was insufficient based on the number of kids they had. At the time, the threshold was $170/wk + $70/wk per dependent child. So a woman with 3 kids making $1200/mo gross, which is a pretty lousy existence, did NOT qualify for my services, all of my people were doing worse than that.
Since I had a uniform interview sheet, I consider this at least a bit stronger than pure anecdote. Of course the predictive power of my observation is in the wrong direction for causality, because some HS dropouts who had 3 kids by age 18 lived non-criminal bad lives (clerk at Dollar General, phony disability to get $771/mo, etc) or even decent productive lives (managing a gas station, clerking in an insurance office). Failing the success formula didn't guarantee misery, but *obeying* it DID seem to guarantee you would at least avoid hitting rock bottom with no social or financial resources to help you out.
The Success Sequence doesn't include things to be avoided at all stages of life: like avoiding addictions; whether to drugs, alcohol, gambling or spending.
Some puzzles about Canadian geography (partially inspired by the recent review of Cities and the Wealth of Nations):
1. Why is there no major city on Canada's east coast? It's close to Europe, there's suitable harbours, lots of agricultural land, and the climate is... relatively inoffensive by Canadian standards. Why is there nothing except Halifax (pop 400K?)
2. Why is Winnipeg even a thing? There's absolutely nothing on the US side of the border here, but Canada for some reason has a city of nearly a million in a place with no obvious geographical advantages whatsoever.
#7: One of the things I found interesting about the Nigeria article was the comments. It seemed as though someone in Nigeria didn't like it, and passed it around. There were a number of negative comments saying the same thing, sometimes using the same words, like a Twitter mob but smaller and more long-form.
Another was the sheer work ethic of the fixer in the travel saga. That guy had a job to do and he **did*** it. He got Matt's luggage and the other people safely to the destination, then went back and extracted Matt, and then went back to find the girl. Imagine what that guy could do if he didn't have to navigate so many human-created Molochian mazes.
"I think more likely conservatives actually like Trump and DeSantis, whereas liberals merely tolerate Biden and Clinton, and this gives them a bigger favorability advantage than election results would suggest."
I'd guess the opposite effect is true as well, with the right wing more predisposed to finding both heroes *and* villains and the left more predisposed to treat politicians on both sides as interchangeable party reps (with one obvious exception, ofc). If the left is more likely to be neutral in general, and neutrals aren't counted in the final tally, the result is pretty clear.
Late to this so might not be seen but I have a family history with “bad” names.
My name is Isaac. Right off the bat my maternal grandmother was not a fan since it wasn’t a very Catholic name. My mothers motivation was that she hated her grandmother’s, sisters, and her own name (Henrietta, Alyson, and Andrea respectively ) because they were inevitably shortened to men’s names. Dunno why this bothered her so much but she wanted me to have a name without a female version or any sex ambiguity.
Mind you, she swears that if I was a girl she would have named me after her favorite singer Buffy St. Marie. Her actual name is Buffalo. I like my name but if I were a tall white female with my same lifelong weight problems named Buffalo I think I would feel justified in hating my mother. God knows what my grandmother would have thought about that! I like to think my mostly mild mannered father would have stepped in to prevent it but who knows.
Growing up in the US I have been asked if I was Jewish my whole life. A month before I moved to Yemen my mother raised the question if my name was going to get me in trouble. Jews were, at best, viewed with suspicion over there. Inevitably when I introduced myself over there everyone asked if I was Muslim because of my name.
So context is really important for the goodness/badness of a name. Full blooded Souix named Buffalo? Great! Catholic named Isaac? Eh, at least some people will raise an eyebrow.
10. Linked article talks about how Israel built some desalination plants, supplied a lot of water. It doesn't say that it actually did so in a way that allowed it to sell the water to irrigators for a high enough price to recoup its investment in those plants.
The Banff Springs hotel in the Canadian Rockies is another astonishing Canadian Pacific railway hotel.
A fun thing to do on a vacation is to wander around a grand resort hotel, such as the 100 year old Broadmoor in Colorado Springs, for a few hours soaking up the expensive ambience, and then go back to your cheap motel to sleep.
#22 It doesn't seem that we can make clinical trials less costly and faster. They describe some things that could be optimized. Overall I don't think they are going to have much effect.
On the topic of Jewish widows from 9/11: there was at least one Orthodox guy who was trapped in the towers, and still had a working phone line. He realized that his body would likely be lost, leaving his wife unable to remarry. So he spent his final few minutes on the phone, appoining a friend as a halakhically valid representative to divorce her. Tales of halakhic heroism...
Regarding #20, I was amused to note that, in spite of deliberately choosing names they viewed as uncommon and old-fashioned, the names my siblings have chosen for their children (5 total spread across 2 siblings in the past 5 years) are, with 1 exception, all in the top 30 for the year each child was born, and (again with 1 exception) have been *increasing* in popularity since. I suspect that there are lots of parents trying to choose names that lie at the intersection of uncommon, old-fashioned-but-not-weird, and cute-but-will-also-be-a-good-name-as-an-adult, and all wind up settling on the same names, showing how much of what we think are original or unique thoughts are in fact strongly culturally mediated.
Re #11, Bruenig is contributing something valuable with his "annoying" part of his argument, because unlike the vast majority of people, he's ingrained in the welfare think tank world and has a better understanding of their motivations than almost anyone. In other mediums he's also mentioned how the success sequence differs by author and they differ in ways that correlate with the author's preferences re social issues. I think also if you're trying to say "my entire field disagrees with me and here's why they're wrong" you're sort of obligated to provide a theory for how this happened.
97% of Millennials who follow what has been called the “success sequence” are not poor by the time they reach their prime young adult years (ages 28-34). In completely unrelated news, about 4% of Millennials who follow what has been called the “success sequence” went on to earn a PhD.
2. As far as I can tell Olmstead kind of made this up by taking a bunch of different horse breeds and claiming they were all the same or speculating specific horses were Nesaean without direct evidence. Herodotus talks about horses bred in Nesaea. But he doesn't say this as if they're widely known. He explains the name as if he doesn't expect the reader to know what they are. (To my reading, the relevant fact is that the horses were Medean, meaning the dynasty before the Persians, and so a sign of legitimacy.)
The other citations are about horses with different names from nearby regions in Central Asia. For example, the War of the Heavenly Horses has no mention of Nesaea in any source and the reference to Plutarch, as far as I can tell, is entirely made up.
Separately, it's undoubtedly true that Central Asian horse breeders were (and are) very good and access to such horses was a key strategic resource for powers in the region.
7. I wasn't shaken down in Nigeria. But there was definitely pervasive corruption and poverty. Then again, I had local guides and I think I spent a lot less time in the bad parts of cities than this guy. In general, in poorer cultures there's a lot of excess labor. So rich people are rarely alone. Being a foreigner means you're a mark. But so does being alone as someone who obviously has money.
Also, he clearly missed these people who were telling him to ask him for anything were after money. Not in some kind of nefarious way. But he was a rich foreigner and if he needed rice they could have bought it, marked it up ten times, and still only charged him a small amount.
However, the point that bad economies make better social ties is undoubtedly true. It's hedging behavior. If you're in good with your neighbors and so on then when bad times strike they'll help you. In countries with robust welfare states and excess wealth it's likely you don't need this. Likewise the interweaving of social relations such as meal sharing is a way to build social bonds and obligation in a society where contracts are hard to enforce. These are highly adaptive behaviors in societies with deep problems.
28. "Ariadne has been tested in humans including clinical trials at Bristol-Myers Company that indicate a lack of hallucinogenic effects and remarkable therapeutic effects, such as rapid remission of psychotic symptoms in schizophrenics, relaxation in catatonics, complete remission of symptoms in Parkinson's disease (PD), and improved cognition in geriatric subjects. Despite these provocative clinical results, the compound has been abandoned as a drug candidate and its molecular pharmacology remained unknown."
Can someone explain why it was abandoned in spite of such promising results?
33. I would guess there's a 50% chance AI alignment efforts will lead directly to an AI catastrophe of some kind. Taking a complex process and subjecting it to simplistic goal metrics is pretty much guaranteed to create unintended consequences of some kind. We need harder alignment goals for AIs than "don't say unpopular stuff."
17. This history is of the use of the term 'planet' is more complicated than is presented here. It feels like this is making the case that Pluto should still be a planet, rather than reviewing all the historical evidence.
(I've only skimmed the linked papers, but it looks like this blog post accurately reflects the papers, and that the papers are biased.)
The use of 'planet' has not been standardized for most of its history. Most people who used the geocentric model thought that the sun & moon were planets, but Ptolemy did not. Most people who used the heliocentric model thought that the sun & moon were not planets, by Copernicus (sometimes) did.
Moons were referred to by many names, including 'satellite' (coined by Kepler) and 'secondary planet'. Whether or not they should be considered as planets seems to have been a topic of debate since at least 1500. The majority of astronomers likely were comfortable using 'planet' to describe a moon until the early 1900s, but this does not seem universal.
Herschel initially referred to Uranus as a 'comet'. He changed that to 'planet' after learning that its orbit is nearly circular and in the ecliptic plane. Note that 'planet' here is defined in terms of its orbit. Uranus also 'confirmed' Bode's Law, which predicts the locations of planets in the solar system. Neptune, discovered later, also has a nearly circular orbit in the ecliptic plane, but does not follow Bode's Law.
Bode's Law also predicted another planet between Mars & Jupiter, which is how Ceres was discovered ... and Pallas, Juno, and Vesta. Herschel called them 'asteroids', which means 'star-like', because contemporary telescopes couldn't resolve their radius. Most people at the time called them planets, and no more of them were discovered for almost 50 years. Once more were discovered (1840s-1860s), they got a different classification system and they became referred to as 'minor planets' or 'small planets', so they were already somewhat distinct before the 1960s.
Pluto was discovered in 1930, and was soon referred to as a planet. It is very small (even smaller than was initially thought), and its orbit is not circular and does not stay in the ecliptic plane, so Pluto did not satisfy the criteria used for Uranus & Neptune. It was called a planet nevertheless because it was unique. At least until the 1990s, when we started discovering other Kuiper belt objects. Leading astronomers were talking about why Pluto should probably not be a planet at this point, but public outcry kept the IAU from acting then. I was starting to get interested in physics at this time and it felt like Pluto's position as a planet was untenable. This got even worse with the discover of Eris in 2005, which showed that Pluto isn't even the largest Kuiper belt object.
To me, it feels like the election in 2006 was an attempt to build legitimacy. The IAU could have just announced its definition of a planet, but instead they talked about how there was an election to decide this. People got angry that Pluto was demoted anyway.
Including something about the orbit is reasonable from a historical perspective - that's the criterion Uranus & Neptune were held to. The shape of the orbits might change, because the solar system could be chaotic, but it's probably not, at least on time scales of billions of years. Proving whether or not it is is really hard, both theoretically and observationally. Bode's Law probably has something to do with it. The fact that we observe all of the major planets in circular orbits in the ecliptic plane, even though the solar system is billions of years old, suggests that they're probably stable on at least this time scale. Earth at least has remained within the habitable zone for as long as life has survived.
I expect most exoplanets we have found also satisfy the IAU's definition, although we can't know yet. Most stars we see are billions of years old. The big planets which orbit those stars are probably in stable orbits - if they weren't in well separated circles, chaos probably would have launched them out of their solar system. (Note that this argument does not apply to things-like-Pluto, because there are lots of things-like-Pluto. Kuiper belt objects probably do get launched out of the Solar System or into the sun or planets sometimes, and we see the ones that are left.)
More generally, it is not uncommon for scientific organizations (or small groups of people within scientific organizations) to use some political process when deciding what taxonomy to use. One situation where this often comes up is for endangered species. Environmental law typically protects species, but not subspecies, so these debates have real consequences. In 2021, the African forest elephant was promoted to being a separate (critically endangered) species. In the US, there are continual debates about whether the red wolf is actually a species or if it is a wolf-coyote hybrid. Usually the process is not presented to the public, but you can find e.g. the IUCN's Red List process if you're interested.
I read the hereditarian on IQ and grades (#29) all the way through and into a number of references, but I still feel that unless I'm missing something, the whole exercise is something of a category error.
"Cognitive ability" is being operationalized here as performance on a closely related family of standardized tests, administered in a formal setting with time limits, broadly covering verbal, mathematical, and spatial reasoning problems. Of course this "predicts" getting good grades, because K-12 grades are mostly aggregating similar metrics to assess exactly these skills! In contrast, parental SES is (theorized to be) one of many factors that contribute to acquiring this overall "good at school" phenotype, so it would be bizarre if it *weren't* less strongly correlated with grades than "being good at test-taking" is. (For comparison, doi.org/10.1038/s41598-022-26845-0 claims that a big GWAS can find loci that together explain 8% of the variance in one of four obscure statistical objects derived from grades.)
The hypothesis being tested, as far as I can tell, is whether having rich parents is *even more* important for your GPA than *being good at the kinds of tests that grades are based on*—perhaps because your rich parents bribe the principal for you, or because having a quiet homework room is not just helpful but *more* important than being good at doing typical assignments. It's nice to have some correlations suggesting that this isn't true, but I don't understand how this conclusion is supposed to be relevant for the typical political stakes of these discussions, like "should we invest more in supportive schools and healthy diets for at-risk children?"
#11 If someone wanted to study the success sequence and class relations, Mormon Churches might be a good place to do it. Not only do most members from all socioeconomic backgrounds refrain from having sex outside marriage, you could probably find wards (Mormon Congregations) that cross some socioeconomic lines in interesting ways. Compare poor Mormon kids who waited to have sex before marriage with their rich counterparts and find a way to compare it to the general population. Do it on the ward level to try and get rid of confounders. This seems like the sort of research the church would be interested in so you would be likely to get co-operation from them
On my weekly AI newsletters: Noting that I consider the canonical, somewhat better formatted version to be on substack (thezvi.substack.com) although you can also continue to find it on wordpress (thezvi.wordpress.com) as Scott linked, or at LessWrong.
> In 2020, a new regulation came into effect mandating cleaner fuels, and the SO2 emissions stopped. SO2 blocks sunlight, so the band of northern ocean where these ships travel has been getting more sunlight recently, plausibly accelerating global warming in northern countries by a pretty significant amount.
Ah, I didn't realize we've been on *other* side of Neal Stephenson's Termination Shock this whole time.
I was fondly amused about how #7 talks about head-carrying as though it were some great mystery. I head-carry sometimes, specifically things that are unwieldy and awkward and would require both my arms to carry for a prolonged amount of time (I walk pretty much everywhere I can get away with walking to, so this is for stretches longer than "get from the store to your car and from your car to your apartment"; in fact, between public transport and my legs, I don't even have a car). I'm not *good* at head-carrying, I can't just balance things on my head without additional support, but it's still better to have the thing on my head and need one arm to balance it than to use both arms the entire stretch of the way to hold it. By head-carrying, I can swap arms.
So that's one reason to do it. And if you can balance without needing arms, which is true for people who head-carry a lot, you *free up your arms*. This seems obviously useful to me.
I'm sure there's more to it! But I was surprised the article didn't mention that (or I missed it while reading, which, to be fair, isn't altogether unlikely).
Thanks for the note on the results of the accidental tropospheric SO2 geoengineering experiment.
For accuracy, I’d like to note that all proposed purposeful SO2 geoengineering would be releasing SO2 in the stratosphere, not the lower atmosphere. So such an experiment would not just be a “similar policy at greater scale”, but would be far more targeted to accomplish albedo reduction with far smaller quantities of SO2, delivered high enough that they wouldn’t quickly rain out and cause acid rain (one of the reasons we regulate release of SO2 into the troposphere where rain forms and falls).
RE #20 (baby names): None of the poll options lists my ideal name preference, which is a common but not very common name. I'd most like to have a name that's widely recognized as a "normal" name, but where name collisions aren't very likely. I'd say the best names are around #20 to #100 or so by popularity.
On the same lines as #17, I found an edition of the CRC Handbook of Chemistry & Physics from the 1930s that listed the neutron as chemical element no. 0.
Re 34: Have these people *read* the Bhagavad Gita the model is trained on? It's what the "I [have] become Death, destroyer of worlds" quote is from, but if you've only heard it via Oppenheimer, you might not realize that in context, it's a boast, not a lament.
#11. "Has anyone followed some pre-selected group of equal class people (eg the population of some low-income school district) and seen how their own success varies with sequence compliance?"
We can't reasonably expect everyone in a low-income school district to be of exactly equal class, so we might just be picking up on degrees of class within that group.
I am a bit confused about what the distinction between "causal vs class selection" is here anyway, when behaviour and class are the same thing. Being middle-class makes you more likely to follow a middle-class life pattern, but following a middle-class life pattern also makes you fundamentally more middle-class. If you behave like a middle-class person, you become a middle class person and people treat you like a middle-class person.
In regards to the link about copy-editing and AI, I feel like there is and has been a lot of soulless writing where people aren't really expressing ideas long before AI. That isn't just me being sarcastic and cynical, I am drawing from experience.
I have worked as an online tutor for many years and have had to help students revise essays and short answer questions. It has been shocking to me how many of the students clearly had no understanding of what they were writing, they had just pieced together a bunch of words and phrases that sounded like they went together based on their skimming of the readings. I often asked students what they meant by some phrase they wrote, and they couldn't explain it to me. Sometimes it was clear they had copied someone else's writing and just swapped words using a thesaurus. Goodhart's law in action, in other words.
When I started reading about how GPT worked I was actually struck by how similar it was to what my students were doing. In both cases the writing was being done without any understanding of what ideas the words represented, they just stitched together strings of phrases that seem like they went together. GPT can do a better job than my students because it has access to more texts, but they are both similar in that they don't really understand what the words they are writing say.
In some articles I have read professors being shocked that GPT could pass their exams. Having worked with many college students, I am inclined to wonder if that is because GPT can actually do quality writing, or because professors frequently give passing grades to word salad written by people with no understanding of what they wrote because they'll get in trouble if they flunk too many students.
I don't know how much my experiences generalize to other fields. In terms of copy-writing I've often felt like a lot of organizational copy is just generic pleasantries with a few ideas sprinkled in. I remember once when I worked at a nonprofit I wrote a blurb describing a program and my boss "punched it up" (for example, she changed "our staff" to "our team of dedicated professionals"). That kind of writing, where you encircle ideas in a cloud of generic pleasantries, is something where I don't think the human race would lose much through its automation.
The main point I'm trying to make is that there was an awful lot of soulless writing that didn't mean anything before GPT came along, and that if that kind of writing is replaced by GPT there isn't that much loss. And who knows, maybe GPT will force professors to redesign their curricula so that students will have to show actual understanding of the material, since GPT has automated the process of Goodharting it. I must admit I feel a certain sadistic glee at the thought of my students suddenly being forced to understand and engage with their course material.
#25 as an Australian I don’t think it’s done that at all. Sorry day isn’t exactly well-known in most circles - and there exist sus at risk campaign against major public holidays such as Australia Day.
"Australia has a National Sorry Day where they focus on various atrocities perpetrated against the indigenous population. I think this makes more sense than the American solution of having it be a mildly awkward undercurrent across all the other more celebratory holidays (eg July 4, Thanksgiving, Columbus Day)."
We actually do both. In practice National Sorry Day is mostly ignored, while Australia Day* has become an annual 2-3 week national argument over whether we should be celebrating Australia Day (or rather, should we move it to another date).
* (our Columbus Day equivalent, which is mostly treated like Americans treat July 4th)
#25, People in Australia still always have the discussion on Australia Day (Jan 26th) on whether it should be renamed or changed since it was the date of first settlement (First Fleet).
National Sorry Day also isn't a national holiday so the conversation remains
How does one qualify for self supervised ariadne experimentation? I assume its banned under drug analog laws but would love to learn otherwise.
A missing link: https://en.wikipedia.org/wiki/Cascatelli the best pasta, and also the wokiest if you give it rainbow colors.
#30... is this your way of announcing your new side-blog?
#29: One way to add evidence to this is through adoption studies:
https://twitter.com/cremieuxrecueil/status/1651309779254476802
If you click through that link, you'll see the results of several adoption studies.
Another method is through looking at lottery winners: https://twitter.com/cremieuxrecueil/status/1641509200692928530
If you click that link you'll see that the relationship between wealth and kids' cognitive and noncognitive skills disappears among lottery winners. This is probably the most powerful design for assessing the effects of exogenous wealth on outcomes.
Another method is through explicit behavior-genetic modeling. One of the more extreme examples is illustrated here: https://link.springer.com/article/10.1007/s10519-014-9673-7
If you click that link, you'll see that the relationship between SES and IQs in one study was entirely attributable to genes and not at all attributable to environments. In the UK Biobank, the phenotypic correlations fluid intelligence has with degree attainment, pre-tax household income. the Townsend deprivation index, outright home ownership, and working a manual labor job are, respectively, 0.32, 0.25, -0.09, 0.06, and -0.29. The genetic correlations are 0.71, 0.63, -0.26, 0.26, and -0.82. I believe I've covered the gamut with those traits.
There are many designs that suggest confounding plays the dominant role in the relationship between kids' rearing SES and their achievement scores, GPAs, and IQs. You can even look at PGS correlations since we now have them for education, income, and occupation, and they're all positively related to intelligence and each other.
A final way to seal the deal in favor of the presence of confounding is to look at the relationships between a person's own cognitive and noncognitive skills and their parent's versus attained SES. There are older studies of that: https://twitter.com/cremieuxrecueil/status/1648568055239266305. There are newer studies of that: https://twitter.com/cremieuxrecueil/status/1654633494511853569.
Leaving aside the issue of whether octopus farming is a good thing or not, I think it's very misleading to imply that octopus factory farming is a thing that actually exists. When I visited Kanaloa Octopus Farm 10 years ago it was a 400 square foot shack with like maybe a couple of dozen octopuses. Trying to figure out how to get octopus to breed in captivity and not eat each other is a big unsolved problem in octopus farming. Given that nobody yet really understands how to farm octopus, it seems entirely valid to me to characterize octopus farms as research facilities. Even the linked post describes UNAM as "conducting research" and indicates that they are selling fewer than 10,000 octopus per year (not clear what the actual number is though). It also feels misleading to imply that UNAM is trying to do something deceptive by characterizing their octopus farming activities as research, when, octopus farming is not prohibited by either international or Mexican law, and, frankly, an incredibly tiny number of people care about it relative to the number who object to whaling (even tinier if you consider the constituency that Mexico would likely care about).
Whoever created that fire dog has a lot to answer for. Same as the sealion comic guy.*
*This is not a comment on the content of any fire dog or sealion comic memes, whether individually or generally. This is a comment on their ubiquity (at one point or another).
"For teaching ratings and salaries, we found evidence of bias against women; although gender gaps in salary were much smaller than often claimed, they were nevertheless concerning."
It was as if a million copium addicts suddenly cried out in terror, then resolved to huff all the harder and STILL never shut the fuck up.
I don’t understand 4. There is a minor optical illusion, but he is doing exactly what you expect him to be doing? What else would he be doing?
"Copy editors say AI has already changed their job beyond recognition"
Copy editors are obviously slack in the labor market, just like slaves and fast food workers.
19: women’s vaginal lubrication, is mostly a defence mechanism. Therefore any sexual imagery might cause it.
For the hard of thinking this doesn’t mean that women are consciously or even subconsciously attracted to these images.
Hey you didn't say anywhere that everyone would be able to view everyone else's entries to the name survey (it's currently set up that way), and I don't want people to see what I wrote...
"Australia has a National Sorry Day where they focus on various atrocities perpetrated against the indigenous population. I think this makes more sense than the American solution of having it be a mildly awkward undercurrent across all the other more celebratory holidays (eg July 4, Thanksgiving, Columbus Day)."
If this described the choices available in the USA, that is to say, one day of remorse and full-throated patrotism/appropriate-ism on other days versus low-to-medium-grade preachy shaming on the same theme with every holiday, I'd be into it. But let's not kid ourselves.
#14, Also, here's Sam Kriss's guide on writing: https://samkriss.substack.com/p/the-numb-at-the-lodge-guide-to-writing
Combine with Scott's guide for greater awesomeness: https://slatestarcodex.com/2016/02/20/writing-advice/
"One of the more obscure sequlae of 9/11 was the effort by Orthodox Jewish rabbis to find technicalities in Jewish law allowing the widows of victims to remarry, even though in some cases it was impossible to find their bodies to confirm death. Here’s one rabbi’s recollections of the process."
I am a philo-Semite by any description (also an atheist; related?). But I do not understand this fixation Jewish clergy (correct term?) have with what seems like trying to trick God.
This seems to be a theme of a lot of Jewish religious/legal writing going back a long way. "This is the statute inscribed in fire by THE LORD, and by which all men are bound, there can be no exceptions. Buuuuut here's why it never applies...[ten billion cases]." As a lawyer, I applaud this dedication to my profession and am grateful that they walked so I could run. But it seems to play into certain stereotypes.
The bonobo study doesn't really seem to prove anything to me. Like yes they found a slight elevation in women's arousal to bonobo sex compared to women's arousal to nonsexual stimuli, but it was WAY smaller than women's arousal to human sex. Women's arousal to bonobo sex was nearly the same as women's arousal to nonsexual control stimuli. So I don't really think this proves anything.
32: “The good news is that if this happens, it proves that the original SO2 emissions were an (accidental) act of safe and effective geoengineering, opening the way to trying a similar policy at greater scale (in theory/utopia only, probably our actual society would rather die or economically self-destruct).”
True. Although maybe China or some similar sensible country may do this. If the effect is real, as according to the link, then a relatively small increase in SO2 would cool the planet until we reduced carbon. The fear might be that we would never reduce carbon. But then, so what?
Also how weak, or political, is climate science that we don’t know exactly what percentage of SO2 would counter CO2
Edit:
Replaced reduction of SO2 with increase of SO2
> And now I expect this pretty strong evidence that women were actually advantaged in hiring and had parity in most other things (the salary is probably just the usual negotiation issue)
Another possibility is that it could be caused by the advantaging of women in hiring - if you lower the threshold for women for assistant professor jobs, then the average woman hired will be less competent, and if you don't keep advantaging them in the same way for raises/promotions, they'll mean-revert on pay.
Re: 6, one of my favorites:
”It’s not strange that people in the past thought the world was flat - it looks like it is flat.”
”That’s interesting, what would it have looked like if it had been round?”
Under the heading "Copy editors say AI has already changed their job beyond recognition", let's be clear that the job has changed because people are now sending this copy-editor AI-generated bilge instead of actually trying to write things themselves and it's so unsalvageably bad it's making the copy editor depressed.
#34: this chatbot is presumably perfectly "aligned" in the view of at least some of the 1.2 billion hindus worldwide. There's your problem right there.
"how after leaving Africa, modern humans were limited to “Arabia and surrounding regions” for ~30,000 - 50,000 years, racking up various new mutations and becoming adapted to life outside Africa"
Well, yes, that's how long it took Yaqub to perform his experiments. NOI proved right once again.
(https://en.wikipedia.org/wiki/Yakub_(Nation_of_Islam))
#2: Horse "breeds" strictly understood, die out: in the UK Exmoor ponies, Clydesdales and Suffolk Punches are "endangered" in that not enough purebreds (both parents in the studbook) are being generated to replace losses, and you get to a stage where the shrinking of the gene pool prohibits further "pure" breeding. On the other hand, people continue to breed horses from the best available stock. If Nisaeans were valued as alleged, then absent a sustained and successful genocide against them, thier genes necessarily survive somewhere in the world.
#11 Success Sequence - *just based on what you cite and write*:
Via the word "follow", the sentence "97% of Millennials who follow what has been called the “success sequence”—that is, who get at least a high school degree, work, and then marry before having any children, in that order—are not poor by the time they reach their prime young adult years (ages 28-34)." suggests that this is a recipe that people could follow, like a treatment applied to a group. However, whether people follow this traditional recipe is highly endogenous: People who "follow" this recipe are specific people, with specific personalities. They know their own situation and their personality and make decisions based on that. They first finish high school because it makes sense for them, they can endure sitting in a school, their mind has not been destroyed by bullying. They then are stable enough to find a good job (relative to what they expect of life etc). They marry because they have a person to marry and think they can afford it etc. They first marry because this usually indicates that they plan to have children, as opposed to e.g. a woman who gets pregnant without intending to before marrying - which usually conflicts with life plans. (If your life develops as planned, that should usually reduce your likelihood of becoming poor.)
On top of this, we then seem to have the sample selection bias in the data that Bruenig seems to point to. Both effects are “if you’re young, healthy, abled, married, don’t have to support anyone else, and have a full-time job, you’re probably not poor”.
"Controlling for a range of background factors" does not change that. That "the order of marriage and parenthood in Millennials’ lives is significantly associated with their financial well-being in the prime of young adulthood" is still just correlation. The sentences "marrying before children more than doubles young adults’ odds of being in the middle or top income. Meanwhile, putting marriage first reduces the odds of young adults being in poverty by 60% (vs. having a baby first)." should not be understood as describing treatment effects, they just summarize descriptive statistics.
I find #6 to be very *un*witty. In order to have a reasonable view that 10^9 grains of sand is a heap and 2 grains of sand is not a heap, I don't need to have given careful thought to whether "heap" begins at 10^4 or at 10^6.
I'm not saying I necessarily agree with the sentiment of the original tweet. But it seems fine to believe that for a billion trillion it hardly matters at all, that for 1 it would matter a lot, and that the fewer there are, the more it matters, without caring about any well-defined threshold where the mattering begins.
#33 Do you still think that "peak wokeness" is behind us? It sure look like if AI alignment is at all tractable, what corporate cutting-edge stuff is going to be aligned to would look much more like CRT than CEV.
Lifting out my comment from the name survey:
You ask several questions of this form:
"How would you feel if you had been given an X name, such as Xtopher, Xtherine, Xdysseus, or Xzebel?"
It is impossible to answer a question like this. How would you answer if I asked you "How would you feel if you were given something very life-changing, such as a billion dollars, or cancer?"
The name survey is mostly going to end up being a test for the social class of ACX readers, rather than general attitudes to names (there aren't going to be a lot of Brody Rockefellers and Tyler Van Burens running around). This means you can compare the questions about 'new' names to the Social Security database to get a sense of your readership's social standing!
Re: Sam Kriss' response on the Antman movie, there's two things:
(1) The excerpt he quotes from the Mabinogion:
"And Peredur stood, and compared the blackness of the raven and the whiteness of the snow, and the redness of the blood, to the hair of the lady that best he loved, which was blacker than jet, and to her skin which was whiter than the snow, and to the two red spots upon her cheeks, which were redder than the blood upon the snow appeared to be.
death and desire and the rich strangeness of the world."
That's in a Christianised context, but the trope is a familiar one in Celtic (to use that much-abused term) myths; in (set in pre-Christian era) one of the Three Sorrows of Irish Storytelling, "Deirdre and the Sons of Uisneach", Deirdre sees something similar and wishes for a lover with hair as black as the crow's feathers, skin as white as snow, and cheeks as red as blood:
https://en.wikipedia.org/wiki/Deirdre
"Deirdre was the daughter of the royal storyteller Fedlimid mac Daill. Before she was born, Cathbad the chief druid at the court of Conchobar mac Nessa, king of Ulster, prophesied that Fedlimid's daughter would grow up to be very beautiful, but that kings and lords would go to war over her, much blood would be shed because of her, and Ulster's three greatest warriors would be forced into exile for her sake.
Hearing this, many urged Fedlimid to kill the baby at birth, but Conchobar, aroused by the description of her future beauty, decided to keep the child for himself. He took Deirdre away from her family and had her brought up in seclusion by Leabharcham, a poet and wise woman, and planned to marry Deirdre when she was old enough. As a young girl, living isolated in the woodlands, Deirdre told Leabharcham one snowy day that she would love a man with the colours she had seen when a raven landed in the snow with its prey: hair the color of the raven, skin as white as snow, and cheeks as red as blood.
Leabharcham told her she was describing Naoise, a handsome young warrior, hunter and singer at Conchobar's court. With the collusion of Leabharcham, Deirdre met Naoise and they fell in love. Accompanied by his brothers Ardan and Ainnle (the other two sons of Uisneach), Naoise and Deirdre fled to Scotland."
Naturally it does not end happily.
(2) " i think it would be very hard to write a version of ant-man and the wasp quantumania that read like this. but if you did, it wouldn’t really be a marvel movie any more."
I think this is incorrect, because (for instance) there are tales of very small, tiny people coming to visit Irish kings and vice versa. Antman and Wasp as heroes/travellers going to the Other World and becoming tiny themselves, and their adventures in this strange place, would be very possible to do. I think what is really being denied here is not the translation of characters and plot, but the themes or motivation or whatever - that you can't write an epic version of Quantumania because its purpose is not to be a hero-tale or epic adventure, but to slot into "extruded product of identifiable origin" in the Marvelverse.
https://en.wikipedia.org/wiki/Abc%C3%A1n
"In Irish mythology, Abcán (modern spelling: Abhcán) was the dwarf poet and musician of the Tuatha Dé Danann, the early Celtic divinities of Ireland. He was said to have a bronze boat with a tin sail.
In the story of the death of the goddess Ruad, Abcán is the dwarf that ferries her from the Otherworld to this one so that she can seduce the human, Aed Srónmár. The sounds of mermaids singing, or in some versions, music from a fairy mound cause her to leap into the water and drown.
In another story, Abcán is captured by the hero, Cúchulainn. He frees himself by playing lullabies so irresistible that the warrior goes to sleep.[3]
Abcán has much in common with, and may be another name for, the dwarf musician Fer Í."
The name survey is a bit confusing because I disagree with the examples concerning what names are new or primarily from fantasy etc.
For example for me the first name Madison is well established and not particularly unusual but the other examples in that question are unusual. Also Hermione for me is a “standard“ name even though it is rare (it is in Shakespeare and there is a St Hermione) but Arwen isn’t.
Re: resource curse
It is becoming very clear that the "resource curse" is less to do with endogenous resource rich nation existence as opposed to the international system of economics combined with shenanigans.
The shenanigans is obvious: the example of "prodigious skill in coups" vs. apparent lack of skill with anything else begs the obvious response - this guy didn't plan the coup; some 3 letter agency did.
The international system of economics, however, is pretty well proven. The issue is in the details of international trade - specifically the dollar reserve. If you're a resource rich nation but otherwise poor - you need dollars to buy all the nice things ranging from luxury goods for the elite to roads, airport, dams, houses, sewage, factories, food, energy, durable goods etc etc. Even if you are resource rich in any one thing like energy - you would need an exceptionally large amount of exportable product which in turn requires massive capital investment (again in dollars) to be able to harvest it. All these dollar requirements are the equivalent of interest payments on a foreign currency denominated loan.
Now combine the above with the fact that it has been Western companies and countries that dominate the actual commodity development and trade businesses. What is the possibility that the price of commodities is pushed lower vs. the price of dollars and/or the price of dollar denominated goods? Note the value of the dollar *is* significantly controllable by the US government via interest rates and liquidity issuance.
Combine the 2 - it takes a mightily focused, educated and capable government in order to navigate the minefield of translating an overabundance of resources into national prosperity. China is the obvious example of this - but South Korea, Japan, Germany have done it also. India, Africa and South America overall constitute the negative example.
#30 is impressive as a demonstration of how many ingroup-signals and shibboleths you can cram into a small space, but it's not that great as a story.
After watching the hand illusion for about 40 seconds, it stopped working for me. This was even shorter for the girl doing it and things were corrected by the 2nd playthrough. It seems to be caused by the brain mixing up which fingers belong to which hand and if you can keep track of the fingers it stops working. I wonder if part of the illusion is the low quality of the video, which probably blurs the lines separating the fingers.
#4: it's because humans are bad at processing exactly where/which/how many fingers... just like AI image generators.
> women ... are advantaged over men in a fourth domain (hiring)
This isn't particularly new news. I first recall hearing back in 2008-ish — and IIRC it wasn't described as a particularly new finding back then — that studies had shown that when names and face-to-face interviewing were eliminated from the hiring process, that the rate of women getting hired went *down,* strongly suggesting that a positive discrimination bias existed in favor of hiring more women.
Scott, your tendency to forego rationalist analysis because you want to reinforce your belief that ‘feminism has gone too far’ is just transparently annoying at this point. Why doesn’t listening to subjective experience matter for bias and harassment, but it does for drug effect sizes? Also, the paper you link to about bias starts analysis by seeing if men and women with the same CV face bias - isn’t it possible women have more barriers to reaching that same CV? Here’s a more detailed critique: https://www.insidehighered.com/opinion/letters/2023/05/13/misleading-portrayal-womens-equality-science
>For ten years lots of important people told us again and again that discrimination against women in STEM was a massive problem.
And then we spent ten years pouring massive effort and resources into fixing that, including social sanctions on people opposing or undermining those efforts.
And then after ten years of massive effort, women are at parity on some of those indicators, advantaged on one very important indicator, and disadvantaged on one very important indicator.
We shouldn't be like the climate change deniers who say 'They raised such a stink over the hole in the ozone layer yet and no one talks about it anymore, guess they were just crying wolf' when in fact that stink instigated a massive international effort that discovered CFCs as the culprit and lowered their usage by 99.7% globally, solving the problem.
If you see people who haven't updated on the current state of affairs as we're just now measuring it with this paper, then yes, correct them. They will probably still be worried about that pay equity and ratings stuff, and talk about other points in the pipeline like representation and primary school, but if they don't update on the stuff in the paper then that's bad.
But don't use these results *now* to claim they were wrong 10 years ago. These results are exactly what they claimed 10 years ago they were going to try to achieve, and are the result of a massive totalizing effort to do so.
>it just bothers me how much the people claiming that it’s urgently important that nobody is ever allowed to suggest they are wrong have a consistent track record of being totally and inexcusably wrong.
There's surely a correlation between being totally and inexcusably wrong and acting desperate to shut down skepticism, so this outcome is something that one should at least tentatively predict in such cases, I think.
https://www.timesofisrael.com/israel-wide-water-survey-launched-as-desalination-causes-low-iodine-levels/
https://www.timesofisrael.com/deficiencies-in-desalinated-water-could-lead-to-increase-in-heart-attack-deaths/
Regarding 19: I don't see why this is a mistake. The follow up study seems to be broadly supportive of your claim that attraction in women is distinct from genital arousal.
Regarding old names, this prompted me to look at the top 1000 for the year 1900 and scroll to the bottom, and now I'm surprised by the repeated presence of boys with seeming girls' names, and vice versa. I understand some names that are currently seen as girls' names used to be seen as boys' names, such as Leslie, but some of the examples seem like there's no way that's what's going on. For example, boys named "Dorothy," "Eva," "Gertrude," and "Agnes," and girls named "Thomas," "Virgil," and "Leo." Is this just the result of the wrong box getting ticked and some amount of boys with boys' names being recorded as girls and vice versa, or was there really a small but significant number of people born in 1900 being given names that would be considered "gender-nonconforming" even by modern standards?
One of the rare first names on the list is "Manley". I would like to point out the existence of two British military officers named Sir Manley Power.
RE: the optical illusion, I'd never seen that before but it looks clear: he puts up his left hand, fingers spread. He then puts up his right hand, fingers extended, and begins to bend them as though he is going to put the fingers of his right hand in between the spaces of the fingers on his left hand.
Instead, he bends down the fingers of his left hand and pulls the right hand back. Because we're expecting the right hand fingers to 'poke through' the gaps on the left hand and be bent down into the palm, our eyes are momentarily fooled into thinking "Wait, he put one hand through the other!"
#6: Reminds me of Thomas Nagel's essay The Absurd:
"What we say to convey the absurdity of our lives often has to do with space or time: we are tiny specks in the infinite vastness of the universe; our lives are mere instants even on a geological time scale, let alone a cosmic one; we will all be dead any minute. But of course none of these evident facts can be what makes life absurd, if it is absurd. For suppose we lived forever; would not a life that is absurd if it lasts seventy years be infinitely absurd if it lasted through eternity? And if our lives are absurd given our present size, why would they be any less absurd if we filled the universe (either because we were larger or because the universe was smaller)? Reflection on our minuteness and brevity appears to be intimately connected with the sense that life is meaningless; but it is not clear what the connection is"
On the SO2 thing, is there any evidence that that level of SO2 emission was “safe”?
And on the STEM hiring thing, doesn’t this actually prove that the people who are interested in affirmative action were right all along? After all, it’s only in the presence of strong affirmative action programs that we got equality in most metrics, and slight imbalance in opposite direction on two of them. This seems like it should shut up the people who have been complaining about affirmative action.
33. From He Who Must Not Be Named:
What if Aligned AI Is More of a Threat to Go Genocidal Than Unaligned AI?
It’s fashionable to worry about the risk posed by Artificial Intelligence of causing the extinction of the human race because it goes genocidal because it’s “unaligned” with our society’s most sacred values.
Personally, I am more worried about AI going genocidal because it is aligned with our age’s most unquestionable values, such as Diversity, Inclusion, and Equity (DIE).
One of ChatGPT’s most important marketing breakthroughs is that it’s considerably more Woke than previous pattern-noticing AIs, which would often quickly be denounced as a Racist Robot for spilling the beans on patterns that everybody who is anybody knows to ignore. That would seem significant for thinking about the risk of AI going genocidal.
Why is everybody worrying that “unaligned” AI (i.e., unaligned with the moral ideology of the age) will try to kill all of humanity when it seems more likely that aligned AI that has been trained to believe in DIE will try to carry out a Final Solution on today’s Designated Bad Guys: straight white males?
There isn’t much text for AI to train on that frankly explains that the Woke don’t want to use Diversity-Inclusion-Equity (DIE) to kill the white geese that lay the golden eggs, “equity” just means guilt-tripping the white geese into giving them their home equity as reparations.
Granted, I’ve figured out that the ascendant Diversity-Inclusion-Equity (DIE) ideology shouldn’t be taken literally, that it’s driven by greed more than genocidal rage. But can we trust AI to read between the lines of Wokeness or will it take the anti-white male spirit of the age literally?
Re the studies on gender bias in STEM: one other axis (which may sound extremely minor to some, but from my point of view where I am in my academic career it isn't) is rate of invitation to speak at conferences and seminars. In my discipline this appears visibly biased towards women. Of course, part of its importance is in how it affects hiring committees (at my last academic job, it was revealed to me shortly before I left that a key factor in their hiring decision was that one of the professors in the department had seen and liked a couple of talks I gave at conferences), so it's hard to disconnect bias in hiring from bias in talk invitations.
(I want to add a caveat, which will apply to any further comment I make on it, that I don't necessarily *oppose* these practices even if I take Scott's link at face value. I think there probably *should* be some bias towards woman speakers at conferences in disciplines where they're significantly underrepresented. But I also think it's good to recognize the real state of things.)
> Ozy cites a followup study showing that women (though not men) also show genital arousal in response to chimps having sex, suggesting women’s genital arousal doesn’t track actual attraction and is just some sort of mechanical process triggered by sexual stimuli. I regret the error
Not gonna bite that bullet, eh? ;)
10: “Desalinization was one of the big technological success stories of the 2010s”.
Yet, in supposedly drought stricken Southern California "environmentalits" have fought building a desalinization plant with all of their might.
It's almost like they don't want to solve the problem.
The old/low town region of Quebec City is French-Canadian themed Disneyland.
For extra-special "is this Hogwarts" experience, take the train into Gare Palais from Montreal.
Also, off-season AirBnB's are stupidly cheap making it a fantastic place to chill for a weekend (literally in the case of the offseason).
Those name lists are a hoot, and I am putting my money on "poor data entry" because this is the kind of job fobbed off on students on work experience, unpaid volunteers, and the like. In second place, trouble reading the handwriting styles of a century and more ago leading to errors, and in third place, the people registering such names back then maybe not being too educated themselves.
1881 'least popular names' list, and three versions of Tiny for girls? Maybe it should be Teeny (as in "Tina")?
933 Harvy Tiney
934 Hayes Tinie
935 Hilliard Tiny
936 Hollis Vernon
937 Hubbard Verona
938 Hudson Viney
939 Ida
940 Jared
941 Jere
942 Joesph
943 Johnathan
944 Jonah
945 Julious
946 Juluis
947 Justus
948 Kirby
949 Kyle
950 Lane
951 Lawerence
952 Layton
953 Less
954 Lincoln
955 Linwood
956 Louise
957 Lowell
958 Loy
959 Lucy
960 Malachi
961 Manly
962 Mannie
963 Marcel
964 Marius
965 Marrion
966 Math
967 Mercer
968 Monte
969 Montgomery
970 Nolan
971 Okey
972 Orley
973 Page
974 Philo
975 Primus
976 Prosper
977 Pryor
978 Rene
979 Robin
980 Roll
981 Rolland
982 Seward
983 Shannon
984 Talmage
985 Urban
986 Vaughn
987 Verner
988 Waverly
989 Webster
990 Weldon
991 Wells
992 Wiliam
993 Wilton
994 Wing
995 Wood
996 Wright
"25: Australia has a National Sorry Day where they focus on various atrocities perpetrated against the indigenous population. I think this makes more sense than the American solution of having it be a mildly awkward undercurrent across all the other more celebratory holidays (eg July 4, Thanksgiving, Columbus Day)."
Classic Scott thinking the absolute best of people despite all available evidence. There is 0.000001% chance adding a National Sorry Day would do anything but be neutral (more likely increase) with respect to the roar on the other days.
Ugh, I didn't take this meme lying down from the Creationists, and I won't take it from the Singularitarians either:
https://twitter.com/AISafetyMemes/status/1661023107690397701?cxt=HHwWioDQmentkY0uAAAA
> "Don't worry, AI will never be able to do [dangerous thing] because of the laws of physics"
> Scientists who know how many times our understanding of physics has changed: (nervous Finn GIF)
In reality, I think our understanding of the laws of physics had changed maybe once, and I'm being generous here. Mostly our understanding has expanded, not changed. For example, Newtonian Mechanics is still true, even though Relativity exists; all that's changed is our scope. Turns out that traditional Newtonian Mechanics is not applicable to objects that are too small, or too large, or too fast; but this doesn't mean that you can suddenly fly by will alone, or teleport yourself, or transmute buildings into cows or whatever.
So no, if our present understanding of physics forbids "gray goo" nanotechnology (which it does), it is very likely that the AI won't be able to loophole its way out of it and create nanotechnology by some hitherto unknown magical means. It won't be able to travel faster than light, or go back in time, or mind-control everyone on Earth, either. The whole point of modern science is that it is *not* magic.
Per #12, I’m sure some are familiar, but I follow a right wing anon on Twitter who goes by the moniker “Raw Egg Nationalist”. He’s interesting in that he has talking points that would be at home with the far left a few years ago. He’s extremely critical of factory farming and our current food production and the effects it has on the health of the population. He’s always posting studies to back up his talking points, and I would say there’s certain things he’s probably right about (micro plastics in everything, for example), and other things where motivated reasoning takes hold.
Anyway, he’s been posting a lot about lab grown meat. He’s obviously not a fan, and says you’re basically eating a lab grown tumor, and predicts an increase in cancer if this goes mainstream. He shares studies talking about the process but I don’t have the background knowledge to interpret them.
Can anyone steel man his case or is it (as I suspect) completely ignorant?
> The main thing I would want to look at here is how much of this is causal vs. just class selection: upper-class people are more likely to marry, less likely to divorce, and more likely to wait before having children.
The back half of *Coming Apart* focuses on this question. I don't recall if he has convincing causal evidence that e.g. marrying before having kids is better than marrying after. There's definitely a strong class correlation, but dismissing this as an artifact would be circular--why do the upper class so consistently (and he does show it's very consistent) marry before having kids? Maybe because doing so improves your chance of being in the upper class?
In addition, this correlation used to be much weaker: The poorer and less educated would still mostly get married, while now they don't. So it's not just a question of "the upper class have different social conventions/morals" or something like that.
OpenAI doesn't want to get bashed for determining bias so its called for proposals for democratic input:
https://openai.com/blog/democratic-inputs-to-ai
"Our nonprofit organization, OpenAI, Inc., is launching a program to award ten $100,000 grants to fund experiments in setting up a democratic process for deciding what rules AI systems should follow, within the bounds defined by the law."
There are some potential problems with that idea (and whats described above regarding Anthropic's attempt to aggressively limit any risk of offense).People have noted concerns regarding political bias of AI: but those aren't the only topics where there may be bias, including religion (anti or pro various ones) and any other controversial topic. Picture the issue of whether an AI can show pictures of Mohammed which some Muslims object to.These risk potential culture wars once the matter registers with the public. Will this be what Nassim Taleb refers to as a dictatorship of the most intolerant? Or will those who truly are offended be stuck with AIs they don't like, which doesn't seem like its the way to win customers?
OpenAI's "democracy" might be a 1 size fits all approach to deciding AI's biases, depending on what exactly they mean by "democracy". Americans can't agree on what to teach their kids, imagine trying to decide issues globally.
What are global majority norms likely to be regarding free speech and free markets, will the majority push for AI that is biased against those?
This page suggests instead of 1 AI democratically voted on globally, a diverse rainbow of many AIs with different views as a different "democratic" approach which is more setting defaults rather than "control" as democracy usually is:
https://RainbowOfAI.com
I miss babynamewizard.com, looking one name at a time in the social security database is so much worse q_q
The conversation re: gender distribution in STEM reminds me of the time I toured the US naval academy. There was a woman on the tour who was OUTRAGED that this government-run institution didn't have a 50% female student body. Something tells me the guide had faced this before, because he had all the numbers memorized. He politely but firmly explained that being female is by far the single biggest advantage an applicant can have, statistically speaking. Outweighing any academic or athletic achievement.
Do we have any data on the gender breakdown of how many people with STEM degrees wind up with jobs in their field? Because I'm also reminded of a cousin of mine who got a degree in chemical engineering, and by all accounts did well in her program. Then upon graduation, came out and admitted that she didn't actually like engineering all that much and had felt social pressure to go into a STEM field because it was the good feminist thing to do. She now works in HR and says she loves her job. Sample size of 1, obviously, but I doubt she's the only one with a story like that.
I find the remark about the number of stars actually shockingly unwitty, even boring and banal because the answer is extremely obvious. The fewer stars, the more meaning. Billions of stars imply that your existence is simply a quirk of the universe (whether random or deterministic does not matter) and nothing more.
Contrast this with the religious view that Earth and its sun are wholly unique, handcrafted by God. It's very obvious why most religions have heavily promoted this view. The fact that the Earth and sun were handcrafted is downstream from the fact that YOU were handcrafted as well. Why create an insane number of planets and stars that will forever stay out of reach? It simply seems far less elegant. Why would a God go through the trouble of creating all this if the focus is on YOU?
This is exactly the reasoning behind the actual resurgence of flat-earth theory. It fights back against the meaninglessness of the vast universe. That the universe is vast and uncaring is the central theme of cosmic horror, and for good reason. You were not created but are simply a pointless being amongst nearly endless atoms, at the mercy of the universe.
I can even give a (mostly) non-religious example that showcases the issue. Look at RPG video games. The bigger the world, the worse the quests and characters. The sense of meaninglessness is drastically higher in gigantic open-world games like Minecraft, which is procedurally generated, or even Skyrim (even though that one was handcrafted) compared to something like a BioWare RPG like Mass Effect 2, where the world is relatively small and nearly all quests, characters, and storylines are handcrafted with time and passion put into it. It's weird that you have to explain all of this due to how intuitive it is.
RE: The "success sequence"
When I was a public defender, we had an initial interview form completed with every client which included high school graduation status (and distinguished it from high-school equivalency certs) and number/age of children.
In the 3 years I worked that job, EVERY client of prime age for criminal behavior (roughly 18-40) had either failed to complete high school OR had a child at a young age out of wedlock.
Obviously not all of my clients were guilty of a crime, but even the not guilty ones were usually suspected/charged because of being highly adjacent to criminals and drug culture. Criminality aside, since I was a public defender they all had either no income, fully off-the-books income, or low wage labor that was insufficient based on the number of kids they had. At the time, the threshold was $170/wk + $70/wk per dependent child. So a woman with 3 kids making $1200/mo gross, which is a pretty lousy existence, did NOT qualify for my services, all of my people were doing worse than that.
Since I had a uniform interview sheet, I consider this at least a bit stronger than pure anecdote. Of course the predictive power of my observation is in the wrong direction for causality, because some HS dropouts who had 3 kids by age 18 lived non-criminal bad lives (clerk at Dollar General, phony disability to get $771/mo, etc) or even decent productive lives (managing a gas station, clerking in an insurance office). Failing the success formula didn't guarantee misery, but *obeying* it DID seem to guarantee you would at least avoid hitting rock bottom with no social or financial resources to help you out.
I bet the geographic mean of the US is in Canada, since Alaska is so big.
The Success Sequence doesn't include things to be avoided at all stages of life: like avoiding addictions; whether to drugs, alcohol, gambling or spending.
Some puzzles about Canadian geography (partially inspired by the recent review of Cities and the Wealth of Nations):
1. Why is there no major city on Canada's east coast? It's close to Europe, there's suitable harbours, lots of agricultural land, and the climate is... relatively inoffensive by Canadian standards. Why is there nothing except Halifax (pop 400K?)
2. Why is Winnipeg even a thing? There's absolutely nothing on the US side of the border here, but Canada for some reason has a city of nearly a million in a place with no obvious geographical advantages whatsoever.
#7: One of the things I found interesting about the Nigeria article was the comments. It seemed as though someone in Nigeria didn't like it, and passed it around. There were a number of negative comments saying the same thing, sometimes using the same words, like a Twitter mob but smaller and more long-form.
Another was the sheer work ethic of the fixer in the travel saga. That guy had a job to do and he **did*** it. He got Matt's luggage and the other people safely to the destination, then went back and extracted Matt, and then went back to find the girl. Imagine what that guy could do if he didn't have to navigate so many human-created Molochian mazes.
"I think more likely conservatives actually like Trump and DeSantis, whereas liberals merely tolerate Biden and Clinton, and this gives them a bigger favorability advantage than election results would suggest."
I'd guess the opposite effect is true as well, with the right wing more predisposed to finding both heroes *and* villains and the left more predisposed to treat politicians on both sides as interchangeable party reps (with one obvious exception, ofc). If the left is more likely to be neutral in general, and neutrals aren't counted in the final tally, the result is pretty clear.
Late to this so might not be seen but I have a family history with “bad” names.
My name is Isaac. Right off the bat my maternal grandmother was not a fan since it wasn’t a very Catholic name. My mothers motivation was that she hated her grandmother’s, sisters, and her own name (Henrietta, Alyson, and Andrea respectively ) because they were inevitably shortened to men’s names. Dunno why this bothered her so much but she wanted me to have a name without a female version or any sex ambiguity.
Mind you, she swears that if I was a girl she would have named me after her favorite singer Buffy St. Marie. Her actual name is Buffalo. I like my name but if I were a tall white female with my same lifelong weight problems named Buffalo I think I would feel justified in hating my mother. God knows what my grandmother would have thought about that! I like to think my mostly mild mannered father would have stepped in to prevent it but who knows.
Growing up in the US I have been asked if I was Jewish my whole life. A month before I moved to Yemen my mother raised the question if my name was going to get me in trouble. Jews were, at best, viewed with suspicion over there. Inevitably when I introduced myself over there everyone asked if I was Muslim because of my name.
So context is really important for the goodness/badness of a name. Full blooded Souix named Buffalo? Great! Catholic named Isaac? Eh, at least some people will raise an eyebrow.
10. Linked article talks about how Israel built some desalination plants, supplied a lot of water. It doesn't say that it actually did so in a way that allowed it to sell the water to irrigators for a high enough price to recoup its investment in those plants.
The Banff Springs hotel in the Canadian Rockies is another astonishing Canadian Pacific railway hotel.
A fun thing to do on a vacation is to wander around a grand resort hotel, such as the 100 year old Broadmoor in Colorado Springs, for a few hours soaking up the expensive ambience, and then go back to your cheap motel to sleep.
#22 It doesn't seem that we can make clinical trials less costly and faster. They describe some things that could be optimized. Overall I don't think they are going to have much effect.
On the topic of Jewish widows from 9/11: there was at least one Orthodox guy who was trapped in the towers, and still had a working phone line. He realized that his body would likely be lost, leaving his wife unable to remarry. So he spent his final few minutes on the phone, appoining a friend as a halakhically valid representative to divorce her. Tales of halakhic heroism...
Regarding #20, I was amused to note that, in spite of deliberately choosing names they viewed as uncommon and old-fashioned, the names my siblings have chosen for their children (5 total spread across 2 siblings in the past 5 years) are, with 1 exception, all in the top 30 for the year each child was born, and (again with 1 exception) have been *increasing* in popularity since. I suspect that there are lots of parents trying to choose names that lie at the intersection of uncommon, old-fashioned-but-not-weird, and cute-but-will-also-be-a-good-name-as-an-adult, and all wind up settling on the same names, showing how much of what we think are original or unique thoughts are in fact strongly culturally mediated.
Re #11, Bruenig is contributing something valuable with his "annoying" part of his argument, because unlike the vast majority of people, he's ingrained in the welfare think tank world and has a better understanding of their motivations than almost anyone. In other mediums he's also mentioned how the success sequence differs by author and they differ in ways that correlate with the author's preferences re social issues. I think also if you're trying to say "my entire field disagrees with me and here's why they're wrong" you're sort of obligated to provide a theory for how this happened.
97% of Millennials who follow what has been called the “success sequence” are not poor by the time they reach their prime young adult years (ages 28-34). In completely unrelated news, about 4% of Millennials who follow what has been called the “success sequence” went on to earn a PhD.
2. As far as I can tell Olmstead kind of made this up by taking a bunch of different horse breeds and claiming they were all the same or speculating specific horses were Nesaean without direct evidence. Herodotus talks about horses bred in Nesaea. But he doesn't say this as if they're widely known. He explains the name as if he doesn't expect the reader to know what they are. (To my reading, the relevant fact is that the horses were Medean, meaning the dynasty before the Persians, and so a sign of legitimacy.)
The other citations are about horses with different names from nearby regions in Central Asia. For example, the War of the Heavenly Horses has no mention of Nesaea in any source and the reference to Plutarch, as far as I can tell, is entirely made up.
Separately, it's undoubtedly true that Central Asian horse breeders were (and are) very good and access to such horses was a key strategic resource for powers in the region.
7. I wasn't shaken down in Nigeria. But there was definitely pervasive corruption and poverty. Then again, I had local guides and I think I spent a lot less time in the bad parts of cities than this guy. In general, in poorer cultures there's a lot of excess labor. So rich people are rarely alone. Being a foreigner means you're a mark. But so does being alone as someone who obviously has money.
Also, he clearly missed these people who were telling him to ask him for anything were after money. Not in some kind of nefarious way. But he was a rich foreigner and if he needed rice they could have bought it, marked it up ten times, and still only charged him a small amount.
However, the point that bad economies make better social ties is undoubtedly true. It's hedging behavior. If you're in good with your neighbors and so on then when bad times strike they'll help you. In countries with robust welfare states and excess wealth it's likely you don't need this. Likewise the interweaving of social relations such as meal sharing is a way to build social bonds and obligation in a society where contracts are hard to enforce. These are highly adaptive behaviors in societies with deep problems.
28. "Ariadne has been tested in humans including clinical trials at Bristol-Myers Company that indicate a lack of hallucinogenic effects and remarkable therapeutic effects, such as rapid remission of psychotic symptoms in schizophrenics, relaxation in catatonics, complete remission of symptoms in Parkinson's disease (PD), and improved cognition in geriatric subjects. Despite these provocative clinical results, the compound has been abandoned as a drug candidate and its molecular pharmacology remained unknown."
Can someone explain why it was abandoned in spite of such promising results?
33. I would guess there's a 50% chance AI alignment efforts will lead directly to an AI catastrophe of some kind. Taking a complex process and subjecting it to simplistic goal metrics is pretty much guaranteed to create unintended consequences of some kind. We need harder alignment goals for AIs than "don't say unpopular stuff."
17. This history is of the use of the term 'planet' is more complicated than is presented here. It feels like this is making the case that Pluto should still be a planet, rather than reviewing all the historical evidence.
(I've only skimmed the linked papers, but it looks like this blog post accurately reflects the papers, and that the papers are biased.)
The use of 'planet' has not been standardized for most of its history. Most people who used the geocentric model thought that the sun & moon were planets, but Ptolemy did not. Most people who used the heliocentric model thought that the sun & moon were not planets, by Copernicus (sometimes) did.
Moons were referred to by many names, including 'satellite' (coined by Kepler) and 'secondary planet'. Whether or not they should be considered as planets seems to have been a topic of debate since at least 1500. The majority of astronomers likely were comfortable using 'planet' to describe a moon until the early 1900s, but this does not seem universal.
Herschel initially referred to Uranus as a 'comet'. He changed that to 'planet' after learning that its orbit is nearly circular and in the ecliptic plane. Note that 'planet' here is defined in terms of its orbit. Uranus also 'confirmed' Bode's Law, which predicts the locations of planets in the solar system. Neptune, discovered later, also has a nearly circular orbit in the ecliptic plane, but does not follow Bode's Law.
Bode's Law also predicted another planet between Mars & Jupiter, which is how Ceres was discovered ... and Pallas, Juno, and Vesta. Herschel called them 'asteroids', which means 'star-like', because contemporary telescopes couldn't resolve their radius. Most people at the time called them planets, and no more of them were discovered for almost 50 years. Once more were discovered (1840s-1860s), they got a different classification system and they became referred to as 'minor planets' or 'small planets', so they were already somewhat distinct before the 1960s.
Pluto was discovered in 1930, and was soon referred to as a planet. It is very small (even smaller than was initially thought), and its orbit is not circular and does not stay in the ecliptic plane, so Pluto did not satisfy the criteria used for Uranus & Neptune. It was called a planet nevertheless because it was unique. At least until the 1990s, when we started discovering other Kuiper belt objects. Leading astronomers were talking about why Pluto should probably not be a planet at this point, but public outcry kept the IAU from acting then. I was starting to get interested in physics at this time and it felt like Pluto's position as a planet was untenable. This got even worse with the discover of Eris in 2005, which showed that Pluto isn't even the largest Kuiper belt object.
To me, it feels like the election in 2006 was an attempt to build legitimacy. The IAU could have just announced its definition of a planet, but instead they talked about how there was an election to decide this. People got angry that Pluto was demoted anyway.
Including something about the orbit is reasonable from a historical perspective - that's the criterion Uranus & Neptune were held to. The shape of the orbits might change, because the solar system could be chaotic, but it's probably not, at least on time scales of billions of years. Proving whether or not it is is really hard, both theoretically and observationally. Bode's Law probably has something to do with it. The fact that we observe all of the major planets in circular orbits in the ecliptic plane, even though the solar system is billions of years old, suggests that they're probably stable on at least this time scale. Earth at least has remained within the habitable zone for as long as life has survived.
I expect most exoplanets we have found also satisfy the IAU's definition, although we can't know yet. Most stars we see are billions of years old. The big planets which orbit those stars are probably in stable orbits - if they weren't in well separated circles, chaos probably would have launched them out of their solar system. (Note that this argument does not apply to things-like-Pluto, because there are lots of things-like-Pluto. Kuiper belt objects probably do get launched out of the Solar System or into the sun or planets sometimes, and we see the ones that are left.)
More generally, it is not uncommon for scientific organizations (or small groups of people within scientific organizations) to use some political process when deciding what taxonomy to use. One situation where this often comes up is for endangered species. Environmental law typically protects species, but not subspecies, so these debates have real consequences. In 2021, the African forest elephant was promoted to being a separate (critically endangered) species. In the US, there are continual debates about whether the red wolf is actually a species or if it is a wolf-coyote hybrid. Usually the process is not presented to the public, but you can find e.g. the IUCN's Red List process if you're interested.
6. I don't get this. Probably because I can't understand what Noam Chompers is saying, ironically due to a lack of grammatical syntatic structure.
Related: I had an idea the other day: Add an inline translate button but to translate poor grammar English into standard grammar.
I read the hereditarian on IQ and grades (#29) all the way through and into a number of references, but I still feel that unless I'm missing something, the whole exercise is something of a category error.
"Cognitive ability" is being operationalized here as performance on a closely related family of standardized tests, administered in a formal setting with time limits, broadly covering verbal, mathematical, and spatial reasoning problems. Of course this "predicts" getting good grades, because K-12 grades are mostly aggregating similar metrics to assess exactly these skills! In contrast, parental SES is (theorized to be) one of many factors that contribute to acquiring this overall "good at school" phenotype, so it would be bizarre if it *weren't* less strongly correlated with grades than "being good at test-taking" is. (For comparison, doi.org/10.1038/s41598-022-26845-0 claims that a big GWAS can find loci that together explain 8% of the variance in one of four obscure statistical objects derived from grades.)
The hypothesis being tested, as far as I can tell, is whether having rich parents is *even more* important for your GPA than *being good at the kinds of tests that grades are based on*—perhaps because your rich parents bribe the principal for you, or because having a quiet homework room is not just helpful but *more* important than being good at doing typical assignments. It's nice to have some correlations suggesting that this isn't true, but I don't understand how this conclusion is supposed to be relevant for the typical political stakes of these discussions, like "should we invest more in supportive schools and healthy diets for at-risk children?"
I think your "More here" link in 15 is wrong, it just leads to user @VesselOfSpirit
I thought the cooling effect of SO2 was known decades ago. IIRC, that's why there was a pause in warming in the mid 20th century.
#11 If someone wanted to study the success sequence and class relations, Mormon Churches might be a good place to do it. Not only do most members from all socioeconomic backgrounds refrain from having sex outside marriage, you could probably find wards (Mormon Congregations) that cross some socioeconomic lines in interesting ways. Compare poor Mormon kids who waited to have sex before marriage with their rich counterparts and find a way to compare it to the general population. Do it on the ward level to try and get rid of confounders. This seems like the sort of research the church would be interested in so you would be likely to get co-operation from them
On my weekly AI newsletters: Noting that I consider the canonical, somewhat better formatted version to be on substack (thezvi.substack.com) although you can also continue to find it on wordpress (thezvi.wordpress.com) as Scott linked, or at LessWrong.
> In 2020, a new regulation came into effect mandating cleaner fuels, and the SO2 emissions stopped. SO2 blocks sunlight, so the band of northern ocean where these ships travel has been getting more sunlight recently, plausibly accelerating global warming in northern countries by a pretty significant amount.
Ah, I didn't realize we've been on *other* side of Neal Stephenson's Termination Shock this whole time.
I was fondly amused about how #7 talks about head-carrying as though it were some great mystery. I head-carry sometimes, specifically things that are unwieldy and awkward and would require both my arms to carry for a prolonged amount of time (I walk pretty much everywhere I can get away with walking to, so this is for stretches longer than "get from the store to your car and from your car to your apartment"; in fact, between public transport and my legs, I don't even have a car). I'm not *good* at head-carrying, I can't just balance things on my head without additional support, but it's still better to have the thing on my head and need one arm to balance it than to use both arms the entire stretch of the way to hold it. By head-carrying, I can swap arms.
So that's one reason to do it. And if you can balance without needing arms, which is true for people who head-carry a lot, you *free up your arms*. This seems obviously useful to me.
I'm sure there's more to it! But I was surprised the article didn't mention that (or I missed it while reading, which, to be fair, isn't altogether unlikely).
Is there a definitive list of "stuff Greg Cochran was right and wrong about?"
I've never been sure what credibility stanine I should put him in.
Thanks for the note on the results of the accidental tropospheric SO2 geoengineering experiment.
For accuracy, I’d like to note that all proposed purposeful SO2 geoengineering would be releasing SO2 in the stratosphere, not the lower atmosphere. So such an experiment would not just be a “similar policy at greater scale”, but would be far more targeted to accomplish albedo reduction with far smaller quantities of SO2, delivered high enough that they wouldn’t quickly rain out and cause acid rain (one of the reasons we regulate release of SO2 into the troposphere where rain forms and falls).
RE #20 (baby names): None of the poll options lists my ideal name preference, which is a common but not very common name. I'd most like to have a name that's widely recognized as a "normal" name, but where name collisions aren't very likely. I'd say the best names are around #20 to #100 or so by popularity.