Is there a protocol for drawing Scott’s attention to commenters that should get a warning or banning? It’s also quite possible I’ve missed a FAQ explaining how to do this or why not to do this.
As of August 1, the sale of marijuana becomes legal in Minnesota. Gummies containing 5 mg of THC have been legal for a little over a year.
There are some inexplicable exceptions to the older law though. For example you can buy a canned soda like beverage with 50 mg total. I think of the poor novice to THC intoxication knocking back two or three cans as if that was comparable to drinking 2 or 3 beers. That’s the sort of dose a cancer patient might use for pain and nausea. You need to build up to that level.
Yikes. Happy trails, my friend. I’ll get back to you in 5 hours or so. You might want to lie down for a while. Give me your car keys and I’ll put on some Allman Brothers.
Fortunately the new law will limits the content of canned beverages to 5mg of THC per can.
Sorry for the naiv question: what is the fastest way to find a concrete comment of mine on this site? It was on use of media compared to use of social media in Germany. I think this should contain enough keywords to narrow it down, but I haven't been successful so far.
Have there been any major, damaging computer hack/virus incidents where the perpetrator gained nothing tangible from the act? I'm looking for an example like the "Bank of America Hack of 1998 where Mike Smith hacked into the database and deleted 1 million accounts, destroying $250 million."
Playing with Llama 2, poetry completion seems completely flat compared to Llama 1. Coherence is up but I can't get the 13B model to generate anything better than unimaginative doggerel. Maybe my prompts need serious revision, or I need to move to the 70B model, but so far I am disappointed.
"Some of the best responses were Yes, Students At Elite Schools Are Actually Taught Different Things,"
When my kids were applying to colleges, I researched this very carefully. My conclusion is that the "Ivy League" did very little educationally that could not be found at the flagship State Universities, all of which have "honors" programs and all of which have large graduate programs where the same stuff is taught everywhere.
Since their grandmother wanted to pay their tuition, my kids went to a Midwestern Private R1 university that is always ranked with the Ivies. They got good educations, but, in my estimation, if they had gone to State, they would have wound up the same.
My nephew did go to an Ivy. He wanted to go to medical school, so he took all of his required science courses at a state university branch in his hometown so he could be more assured of getting an "A". He went to Medical school and landed a residency in a famous Eastern hospital.
I am firmly in the Bryan Caplan school on this one.
"several people ... pointing out that smart people want to socialize with other smart people."
Yes, but, there are plenty of smart people everywhere. Let us think this through with numbers.
Most IQ scoring systems have a mean of 100 and a standard deviation of 15. Applying this and a standard deviation calculator, to the approximately 5 million kids who turn 18 every year we find that:
IQ 115 -- 129 (definitely college material) 700,000
IQ 139 --144 (really smart professional/graduate school) 100,000
IQ 145 & up (I don't like the word genius) 7,000
Now the top schools. There are about 70,000 total undergraduates at the Ivy League schools. That would be about 18,000 admissions per year.
A good friend is a senior faculty member at an Ivy. From what he told me, and from what was publicized during the trials leading to the recent SCOTUS decision, Less than half of those admissions are made purely on the basis of academic evidence of intellectual prowess, the rest being, affirmative action, legacy, athletics, etc.
So, I would guess that no more than half of the incoming class of the Ivys consists of IQ >=130. That would be a total of less than 10,000 out of that 100,000. Take another tranche of 10,000 for the the non Ivy top schools (Cal Tech, MIT, Stanford, etc.) and there are still 80,000 really smart kids who have to go to other colleges.
I am willing to wager that any of the top 25 state universities has as many really smart kids (IQ >= 130) as the average Ivy. It is just that they are a smaller percentage of the student bodies at those schools. But, from a social viewpoint, it is not a real problem. They will meet each other in class and find the hang out locations where they can meet more.
Again, Bryan Caplan is right.
My conclusion is that the best reason to want to go to an IVY is the hope of meeting and marrying rich. As my father told me, a man can accumulate more wealth in a 15 minute wedding ceremony that he can in 45 years of honest toil.
Hello smart people of ACX. I'm trying to find the word for a concept: when you do something which you're explaining or talking about. Some examples:
"Oh my god, you don't even know what condescending means", said John condescendingly.
"You're great at manipulating those people by flattering them", said Kim to Tom, causing him to beam.
Google thinks I'm looking for simile, metaphor or analogy because I can't search properly. Whichever Chat GPT is free thinks I want "performative speech". Bard thinks "exemplify", which I think is the closest. Any ideas?
Scott spoke some time ago of TPOT, the postrationalists, and how they never gave any philosophical grounding to their stuff. I'm not really a part of that, but I'm quite taken with their idea of the vibes, and I wrote an anti- or meta-philosophical defense of vibes.
Location: 1970 Port Laurent Place, Newport Beach, CA 92660
Date: Saturday, July 22nd, 2023
Time: 2 PM
This week we are diving into the deep end of the Fermi Paradox, a 70-year-old question that has been surprisingly hard to answer when people have deeply investigated it. The universe seems to be a place that could be filled with intelligent technological life, why haven’t we observed any? There are a lot of simple-sounding solutions to this problem, but upon deep scrutiny, many of them don’t hold up well or rely on factors we just don’t have hard numbers for. Most of the viable solutions point towards disturbing conclusions.
Google Docs
LW/ACX Saturday (7/22/23) The Fermi Paradox (Why don't we see evide...
Re: Aella, but also as a statement of general grumpiness-- what is with the general tendency to treat self-report, self-selected internet polls as meaningful empirical evidence for real-world phenomena?
That goes especially intensely for polls being used to adjudicate culture war questions with strong allegiances among the Very Online, like trans/ poly/ LGBTQ issues. Since it's usually very obvious which answers would yield results that flatter Our Side ("Why, uh, yes, I really definitely *am *a cis woman! And *of course* I get super aroused by the thought of pawing my own boobs while wearing platform heels and a tiny bimbo skirt!"), it seems implausible that those polls wouldn't get a ton of brigading and possibly bot mobilization to give false answers and skew results in the most convenient direction.
Which, frankly, is the type of critique I practically learned to make from reading SSC/ACX. So can anyone explain the serious takes around here?
dating ad: 26 year old woman looking for a slightly older male long-term partner
— am an Australian English/history teacher who enjoys reading (most on Bloom's Western Canon or any text written before the 20th century, miscellaneous articles on anything technical, a lot on Gutenberg and Archive.org, sometimes ArXiv too), classifying transport models, exercising, and travelling. wants children (have worked with them and teenagers for almost a decade, and I think they are the best)
— looking for somebody who is also averse to TikTok and Netflix and other numbing agents. somebody who enjoys reading widely (interest in history, tech, military affairs, and transport e.g. aircraft or trains would be excellent), explaining technically dense processes, travelling, a night out dancing, and Norm Macdonald. strong preference for engineers
Here's an article in the New York Times about the future of Wikipedia in the age of ChatGPT. In addition to its topic, one notable feature for this forum is that the first source who is discussed in the article is presented only by his Wikipedia username, "Barkeep49", and the author explicitly states that he isn't giving this Wikipedian's real name, to avoid harassment.
I'm not sure if the difference in policy is an official change at the New York Times, or the fact that this article is in next Sunday's "New York Times Magazine" rather than a regular daily edition (there could be separate editorial policies for the two contexts), or just about the different attitude of the two authors. (You can see more of their articles at their respective pages. Note that Gertner has his own site, because his articles are published in different venues, while Metz seems to be just at the NYT. https://jongertner.net/category/article-archive/https://www.nytimes.com/by/cade-metz )
"I plan to ask Michael to design the questions for the next survey and demonstrate that they get the same result." Doesn't announcing how you'd like your test subjects to respond mean that your results will be invalid?
If your dream job is to help others do the most good they can, being part of an ambitious start-up, and leverage your expertise in India, then this might be the opportunity you have been waiting for!
These roles will greatly influence the development of our potential flagship program and our general efforts in India and the Global South, and your work may set an example for other regional offices worldwide!
Deadline: August 4th. We evaluate applications on a rolling basis and might fill the role before the deadline, so we encourage you to submit your application as soon as possible.
If you know someone you think might be a good fit, please send this Job Description to them. There is a $500 prize for referring a successful candidate.
Alright, I'm going to attack a core tenet of the Rationalists that has always bugged me. As I understand it, Rationalists claim to be slightly better at avoiding irrational, ingrained-by-evolution biases than the average person (even of the same intelligence). The Overcoming Bias image of Odysseus strapped to the mast to avoid the possibility of succumbing to the Siren's Song emblazons the belief that this is doable.
The belief one can overcome such ingrained biases strikes me as ludicrous. I'm reminded of Carl Jung saying that trying to overcome the unconscious with the conscious is like holding a mirror to the sun.
Whatever deep ingrained biases you have that are shaped by evolution are As Smart As You! Probably smarter. Your *rational* thought is not independent of such overwhelming forces as billions of years of evolution.
If you think you can outsmart your biases through concentrated reason, you will not only lose that battle, but you will have made yourself (forgive my frankness, Scott) stupider by thinking you can.
Hey, let's share embarrassing stories of a nerdy adolescence! I'll start:
Near the beginning of Grade 7, Mr Barlow the English teacher had us all submit two topics for short speeches. We would draw two topics at random out of a hat, and give a short talk on whichever of the two topics we chose.
My parents were moderate packrats, and a year earlier I had found a 3-year-old LIFE magazine in a box in the basement, a double issue on the the American city. I was enthralled; the photos, the writing, the futuristic drawings, everything just grabbed my young brain. Besides the mostly optimistic vision of the future, there was a rather Malthusian section about urban sprawl, with the forthcoming megapolises gobbling up arable land, this in a time of exponential population growth.
I was (and remain) a space nerd, so the topics I submitted were:
After The Moon - What's Next For Manned Spaceflight?
and
Urban Sprawl!
*************
The hat was passed around to collect the topics, Mr Barlow mixed them up, and the hat came around again for us to select topics.
Mr Barlow gave us a couple of minutes to check out the topics we'd pulled out of the hat, and then asked if everyone was happy with them.
A tough girl named Lori* raised her hand and said "I got one called Yurbin Sprail, and I don't even know what it is!"
Mr Barlow glowered and asked who had submitted it. I raised my hand. Lori and her peers sneered at me and snickered. Mr Barlow got mad and asked me if this was a joke. I stumbled through a brief explanation, and must have seemed sincere enough that Mr Barlow realized I had submitted the topic in good faith.
I don't remember what topics I had picked from the hat, but no doubt both involved pop music I hadn't heard, sports I didn't know anything about, or TV shows I hadn't seen.
Those who wish they could return to their teen years were better equipped to deal with adolescence than I was, or have shorter memories.
* Lori was part of a gang of girls who smoked, stole from the convenience store across the street, and used a lot of profanity. Our rookie homeroom teacher, Mrs. Lawson, took a leave of absence partway through the year. She was gone for a month or so.
I'm trying to learn more about free banking. To that end I'm reading George Selgin's "The Theory of Free Banking" (1988). I'm not an economist but am an econ-nerd, esp. on monetary policy, and got into Scott Sumner and market monetarism back in the day.
I don't think I quite get it. Selgin says that in the Scottish free banking era, banks were stable keeping only 1-2% of deposits as reserves in specie (I say specie here but it could be a base fiat currency as well), and they typically did not suffer from runs and failures. Why does the existence of private banknotes allow this? Is it because the base money and banknotes are no longer the same? If depositors get skittish and want to pull out in a textbook run scenario (regardless of whether those are actually real), how would the bank not quickly run out of reserves? From a financial stability standpoint, I sort of intuit that with banks printing their own money, the money supply endogenously rises and falls with liquidity demand, and insofar as financial crises are generally caused by a contagious liquidity crisis, self-printing of money would prevent runs and instead allow for private money to simply trade at a discount instead if a bank is insolvent (i.e. similar to how there is no such thing as a run on a stock because the price trades freely). But I'm having trouble putting all the pieces together.
I think I may need a plain English explanation to supplement the more jargony one.
Are there any known techniques to improve general memory? I know a bit about the sort of repetition learning and the stuff Gwern has written about, but I more mean memories of experiences. Mine is terrible… it has its pros and cons, like I don’t really get traumatised because I just forget, but I tend to repeat the same mistakes pretty often. My general patterns of experience also seem to get repeated quite a lot.
Assuming that all of the QAnon predictions/prophecies were accurate (especially the ones that contradict each other), what is the current state of the world?
...well, I've got a sudden bug and you folks get to hear about it.
Recently watched Paranoia Agent, an anime about a mysterious attacker in which every episode follows a different character up to their run-in with him. Started with Boogiepop Phantom, which has the same style of episodes following the various people that run afoul of her, but Boogiepop Phantom is not holding up to the latest viewing. Looking for other stories that have episodes revolving around minor characters who are themselves revolving around a central character. Google mentions another anime called Hell Girl, but it seems to not be available in the US.
Does anyone know any other stories in that style? Any medium.
Re: the recurring subject of Bay Area "NIMBY" vs "YIMBY" wars:
Can anyone offer links demonstrating the reality of the much-discussed shortage of "affordable" (for people employed on location) housing there? A cursory look at various ads (browsed SF and even Palo Alto, arbitrarily close to various TBTF corp. headquarters) seems to show plenty of 1- to 3-room flats going for 3 to 5 thou. $ / mo; perhaps tight quarters, but not out of reach cost-wise for even an entry-level programmer. And not particularly more expensive than e.g. the habitable part of Washington DC.
There was a trend of the Open Threads having other foreigners asking "is America really like that?". I guess I'll take my turn.
As a Canadian, something I see a lot in TV is Americans wearing their shoes in their own house. When people don't do this, it is usually called out. I've also seen a few shows refer to it as a Japanese practice.
RE the autogynephilia question, assuming the results are correct, it seems to suggest that it is somewhat associated with being trans, but more importantly that also that a lot of cis men have it/engage in it.
From a practical/trans rights point of view this is a mixed bag result, because the fact that lots of cis males engage in it definitely makes the argument that cis males can use very loose self-ID rules in locations like bathrooms and changing rooms and spas to involve cis females in unwanted sexualized role play of their own fashioning. Trans rights activists want to laser like focus on the idea that 'real' trans people are no more dangerous than cis people. But the problem is that there are lots more cis people. So if autogynephilia is semi-common among cis males, loosening the rules on previously sex segregated spaces ends up potentially exposing a lot of cis females to a lot of cis males.
This may still lead to different answers in different places (most female bathrooms tend to be single stall oriented so maybe no big deal while maybe naked spas could still be sex segregated).
In any case the finding that it isn't *particularly* associated with being trans doesn't answer the question of what we need to do with the cis guys who are autogynophile (in female spaces).
I often watch TV with my girlfriend in the evenings. In principle I think a particular type of video game would be a better use of this time, but I don’t know games - can anybody recommend stuff which satisfy the following criteria:
- me and my girlfriend can play it together
- has artistic/literary value
-is some kind of big adventure
Basically I want something like an Elder Scrolls game or Red Dead Redemption, but which me and my gf play together.
I feel like I remember that Scott linked a Noahpinion tweet of a graph from a paper comparing parenting hours spent by men and women across time, which showed that men today spend as much time on parenting as women did 50 years ago, but that women have similarly increased their time, so the gap between men and women has stayed ~the same. I can't for the life of me find either the tweet or scott's link to it.
Does anyone happen to know/remember what I'm talking about and where to find it?
Whatever happened to Rodney Brooks "Intelligence without Representation"-style robots? Wikipedia says it's correctly called "Subsumption architecture" (or maybe "behavior-based robotics"). Was it a dead end? How complex did their behaviors end up becoming? Is it still an area of active inquiry?
How important is proofreading in languages other than English?
I was reading the newest book review on The Educated Mind and noticed a place where the reviewer mixed up the words "good" and "well." This on it's own isn't very interesting; even extremely intelligent, native English speaking, professional writers who make a conscious effort to proofread their work make mistakes like this with such frequency that most of us sail right past them without a second thought. But that normalcy was striking to me. I am a native English speaker, with 19 years of education mostly spent on improving my reading and writing ability under my belt, who makes his living using the written word to persuade, and I would be shocked if this short paragraph didn't contain at least one error in grammar or punctuation. I can't remember the last blog post or newspaper article I read that didn't include multiple typos. I have no fewer than three separate books on my shelf on English style, grammar, and composition, which my employer considered important enough to my job to pay for as a business expense.
I've always taken this state of affairs for granted, where nobody really knows a lot of the rules and even the most literate among us mostly get by on extremely-error-prone intuition and are forced to spend significant amounts of time proofreading if they want to be clearly understood, but I'm also aware of just how weird/complex/idiosyncratic the English language is compared to other world languages.
For those of you whose native language is not English, do fluent and even skilled writers in your native language seem to struggle with grammar, punctuation, and spelling to a similar extent English writers do? How common are the sort of minor errors that most of us have learned to politely ignore when reading English-language newspapers? Are there different sorts of pitfalls than the ones I'm familiar with as a native English speaker? I'm especially interested in non-alphabetical writing systems like Chines, Korean, Japanese, etc.
The synthesis of your post regarding admissions and your link from reddit about advanced math classes, is this:
Harvard et al really do have a big share of the smartest, most driven students. But that isn't the bulk of the student body. The bulk is kids who are less smart/driven and there in part due to privilege.
Sure, great that you can take grad-level math classes as a junior, but realistically most Harvard kids don't do that.
Part of the advantage of going to Harvard is the ambiguity. Are you "pothead child of hedge fund billionaire", "somewhat smart kid of two doctor/lawyer/etc types", or "Mr. math genius, take-multiple-graduate-level-math-classes-as-a-junior"?
Also - just to say it - I think the last group will be disproportionately from well-off families, just because that sort of ability isn't only about talent, but also about having it cultivated from a young age. There are perhaps "doing calculus in preschool" level prodigies out there for whom it doesn't matter, but they're a *very* small slice, even smaller than the "Mr. math genius" group.
On my Substack reader app, there is now an annoying feed. They try to put in links to blogs they think I might like. And one-line comments much like tweets keep popping up. It is distracting and annoying. I cannot find what I subscribed to and want to read.
It may be entertaining to watch, but I don't expect it to be the caliber of Willy Wonka's Chocolate Factory, which remains one of my favorite movies, and which I consider better than the book. In fact, watching the trailer made me think that the script could have been generated by an AI. So...I asked ChatGPT, which only has training data before 2022, to tell me the plot of such a prequel.
I actually like ChatGPT's plot a little better, but found, as expected, it didn't really have much new. It told of Willy's adventures, with his friend Matilda, went in search of magical confections, including Vanilla Sprites, Sugar Sylphs, and the finding of oompa-loompas.
I asked it to expand on Sugar Sylphs, and found they lived in clouds, which Willy and Matilda reached using some kind of magic rituals. Though I kind of liked this idea, I found it missing the essential ingredient, as it were, of what made the original movie so great, which was the zaniness. I suggested that Willy and Matilda should, instead, reach the Sugar Sylphs' clouds by eating the right combination of specially-crafted sweets, to obtain a sugar high. This seems in-line with Willy mixing a coat into a batch of candy because it was too cold, and some shoes in another batch to give it a little kick.
This illustrates to me why LLMs like ChatGPT can't yet replace humans. Hollywood currently has a writer's strike going on, protesting AIs. If the writers only produce what AIs can produce, then Hollywood has a point. If they can do better, generating actually good content, even with the help of AI, then Hollywood doesn't realize their true value.
Though I have never written a screenplay, I bet I could do better than the prequel coming out in December, with help from ChatGPT. A professional screenwriter ought to, on average, do better than me.
How would the US military deal with minefields, like what the Ukrainian military is dealing with now? Like, how would they clear them? I understand US dominance is based on air superiority, but you do have to cross the actual minefield at some point- right? I read the Wiki page on demining, but it just lists general tactics.
Inspired by the Wiki page and the Ukrainian struggle, I (a non-engineer) came up with a couple of random cheap demining tactics. These are more in the category of 'shower thoughts' and not like a serious proposal, but just for fun:
How about a pressure washer to trigger them? A cheap remotely-controlled vehicle like a tractor or excavator can carry a commercial pressure washer that sprays the ground in front of it. The amount of pressure a pressure washer can generate is amazing- like more than a human's weight- so should be enough to trigger the mines. Just keep spraying the ground and advancing. Will some of them get damaged? Sure- just roll out the next cheap tractor/pressure washer combo and keep going. Water supplies might be a hard limit.
Alternately you could just do an agricultural-style controlled burn on vegetation, where the heat should trigger all the mines. Do a spray of a cheap propellant (again via a remote-controlled tractor), the burn will detonate buried mines, if the tractor gets damaged replace with another, and keep advancing. Lots of farmers do controlled burns in their fields, not a high-tech problem
David Brooks had an interesting piece in the NYT about Douglas Hofstadter changing his mind about AI. He links to a LW article that Gwern posted with a transcript of an interview with Hofstadter.
DH: It “just renders humanity a very small phenomenon compared to something else that is far more intelligent and will become incomprehensible to us, as incomprehensible to us as we are to cockroaches.”
Brooks was surprised by the interview and the change in thinking and called Hofstadter to talk about the change.
The standard evolutionary explanation for morning sickness is that, in combination with hyperosmia, it deters pregnant humans from eating questionable food during the first trimester when embryos are sensitive to teratogens, thereby serving an adaptive role by compensating for the cost of not being able to eat very much or do much of anything except lie in your cave feeling miserable.
I think this is a plausible explanation for hyperosmia, but fails to explain species-specificity and timing of nausea and vomiting. Routine vomiting during gestation has barely been reported in non-human animals, including other omnivores (rats don't vomit at all, some reports of mild food aversion in gravid dogs and cats); I don't know of any evidence that human embryos are more vulnerable to environmental teratogenesis, and if we were I'd guess it would be during the the late first and throughout the second trimester during the most intense periods of neurogenesis and neuron migration to make our abnormally complex brains, whereas morning sickness is usually worst around weeks 6–14.
My alternative hypothesis is that the adaptive value is as a stress test of the social caloric support system. Humans are so social and invest so heavily in each offspring that it's almost impossible to survive and successfully complete pregnancy and lactation unless you have conspecifics bringing you lots of food. Therefore, instead of being selected to minimize hormonally-induced nausea like other mammals, early humans found that it increased lifetime fitness if they stopped being able to feed themselves for a month or so right at the beginning of pregnancy—and, if this test revealed that they were in a situation where nobody was able or willing to feed them, to cut their losses by having malnutrition-induced miscarriages before going through the extremely dangerous process of late gestation and childbirth, and to try again later.
This alternative hypothesis explains not only why our bodies would do something with such obvious fitness costs, but also why other animals don't do it (except perhaps those that rely on humans feeding them when they feel bad!) and why it happens essentially as soon as the body is confident that it's pregnant.
I write a Substack newsletter where I share three things I find interesting, once a week.
In the latest issue I shared:
1. A study finding an unexpected overlap in the demographics of BLM and anti lockdown protests in the USA
2. A chart demonstrating the U shaped relationship between fertility and household income
3. My mapping of median income per capita against average yearly working hours across a number of countries. This shows among other things that the higher the incomes, the lower the average working hours, with the exception of Korea.
You know, something that I haven't heard mentioned in the autogynaphilia debate, but which I think supports the mainstream trans-as-dysmorphia story, is furries.
The furry community plays host to people with a wide variety of paraphilias- attraction to characters with animal features, transformation, vore, micro and macro, etc. But what really connects people in the community isn't the kinky stuff- it's the common longing to be something other than human. Almost everything that makes a furry a furry- the fursuits, the OCs, the role-playing- in an expression of that desire, and finding people who share it is the main reason people become furries. Functionally, the furry community is a social network for people suffering from species dysmorphia.
So how does the autogynaphilia model account for that? Are they supposed to secretly be zoophiles pretending to be dysmorphic? Obviously, that doesn't make sense. Why should one socially unaccepted paraphilia masquerade as a different one? Also, many of the fetishes associed with the community, ilke micro and macro, have nothing to do with animals- instead, the common thread is an attraction to being something not quite human.
It's generally assumed in the community that the fetishes arise from the dysmorphia. That's supported by the fact that most furries' obsession with being non-human started before puberty. It's also just a lot more parsimonious than the alternative.
Finally, furries make no public claims of dysmorphia- most still try to insist to outsiders that it's just an anthropomorphic animal fandom. So, if dysmorphia is supposed to just be posturing for social acceptance, why would they only express it in private?
I'm one of the straight cis male respondents who gave a non-zero answer to the autogynephilia question, and I still don't think I have autogynephilia. Let me explain why.
In a question of the form "Picture a very beautiful woman. How sexually arousing would you find it to imagine <X, where X involves her>?" I am never not going to be aroused by that, unless X is actively disgusting. X could be "her standing in the corner holding textbooks and pretending to be a bookshelf", "her attached to the end of your foot like a shoe" (sorry), or "her as a gigantic godlike figure holding up the night sky" and I wouldn't answer zero. I'm not attracted to X, but I am attracted to beautiful women doing any old thing.
The answers of respondents probably differed more by how they read the question than whether they really have autogynephillia. Some of the cis attracted-to-women respondents are either disgusted by the premise, or answering the question relative to their base rate of being aroused by things. Some of them are answering it literally. Most/all of the trans respondents are disgusted. This entirely explains the weirder results and also invalidates any conclusions drawn from them.
Last year, after reading one of the ACX book reviews, I built a short survey about inner voice and memory (mostly from existing scales), to test a hypothesis that seemed likely to me after reading this review. After analysing the data, well my hypothesis was totally wrong, but there seems to be something unexpected and interesting going on in the data. But the number of responses I got is slightly too low for me to draw firm conclusions.
If you have a few minutes to spare (about 5/7 minutes), would you agree to answer the survey below? If I get a few more responses, I can finish the analyses and post the results if people are interested. Thank you so much :-)
I finished 'Of Boys and Men' By Richard Reeves. I didn't like it so much. Here's my Goodreads review.
2/5 Stars
The first part of the book deals with the gender gap in outcomes between men and women. There are some interesting figures here, but the basic gist of the first part succumbs to "gapism" ie., listing a bunch of gaps in outcomes and declaring that an argument was made. "There aren't as many male nurses, there aren't as many male teachers!" Reeves, says. I'm sure he's right. But he never questions the medical or educational systems directly - not their organization, not their waste and mismanagement, not the government meddling, not their enormous subsidy.
Reeves spends the chapter on solutions basically arguing for even more subsidy, even more affirmative action, and even more meddling. He anticipates that the reader will criticize him from the left, "Maybe you'll want to report me to the EEOC." Yea, maybe. "What about all of the affirmative action for women!" He says. Yea, I don't think so.
The question from the outset should be: What are the structural problems in the education and medical systems that can be improved at the benefit of everyone? I think redshirting sounds like an ok idea, but the rest of the solutions discussed in this book border on total social and labor market planning - which got us into this mess in the first place. I'm skeptical that more of it will get us out.
I ask because America markets itself as this rugged individualistic country.
But then I hear of people calling the police because they say a kid walking to school or what’s weirdly called a “welfare check” where if someone doesn’t answer your calls you claim they are suicidal and police smash their door in(yes I’m exaggerating but only a litttle; look at all the swatting incidents).
In the UK and most of Europe calls like this would be treated as wasting police time and you would get a warning. So why are they common in America?
Is it because , contrary to the marketing , America is actually a nanny state?
Ha, I'm out in Berkeley this week, but will have to miss the 7/23 meetup... as that's the day I'm returning to New York! So, see you in New York, I guess. :)
Has anyone observed a correlation between niceness/kindness/generosity and lower IQ? I’m not talking about people who are nice enough. I’m thinking of someone who would be first and foremost described as nice or kind by their friends.
On the other hand of the equation, the “term” calculating is associated with being unkind or inauthentic. But from a rationalist pov being calculating seems like a compliment. Why shouldn’t I use my intelligence to make smart decisions about my relationships?
I was thinking this might have to do with a social expectation that relationships are more authentic when they are based on emotions instead of logic.
I've added a fun Friday type entry for my "blog" https://icdstories.substack.com/. Now every Friday I ask GPT4 to make up an ICD-10-CM code and a vignette to go with it!
Whom will Biden appoint to the UFO Disclosure Review Board as economist and sociologist?
(4) QUALIFICATIONS.—Persons nominated to the Review Board— (A) shall be impartial citizens, none of whom shall have had any previous or current involvement with any legacy program or controlling authority relating to the collection, exploitation, or reverse engineering of technologies of unknown origin or the examination of biological evidence of living or deceased nonhuman intelligence; (B) shall be distinguished persons of high national professional reputation in their respective fields who are capable of exercising the independent and objective judgment necessary to the fulfillment of their role in ensuring and facilitating the review, transmission to the public, and public disclosure of records related to the government’s understanding of, and activities associated with unidentified anomalous phenomena, technologies of unknown origin, and non-human intelligence and who possess an appreciation of the value of such material to the public, scholars, and government; and (C) shall include at least — (i) 1 current or former national security official; (ii) 1 current or former foreign service official; (iii) 1 scientist or engineer; (iv) 1 economist; (v) 1 professional historian; and (vi) 1 sociologist.
Last week I published a report on the anonymous salary survey in Berlin, Germany. Tech bias, but maybe still useful for some of you. It also has an interactive dashboard. I am also open to suggestions, collaborations and other improvements.
I only started reading ACX after the move from SSC. I have gone back and read the top posts from SSC but still felt like I was missing out on some of the old classics. I built a website to resurface old content from blogs by sending weekly emails. Let me know if you have any suggestions of other blogs / content you’d like to see, hope you find it helpful!
Have you ever felt like you have a song "stuck in your head" while you're doing other things?
If you usually think in words, and the song had lyrics, does that mean you felt like the song lyrics were going through your head at the same time other things were going through your head, ie you had two parallel internal audio streams?
Can you voluntarily give yourself two parallel internal audio streams, ie play two songs with lyrics in your head at the same time?
Can you voluntarily give yourself two parallel internal visual streams, ie visualize a tiger and a sunflower at the same time? If yes, how sure are you that you're not switching back and forth between them very quickly? How many internal visual streams can you maintain at once?
Can you focus on two different areas of your (internal) visual field? IE can you close your eyes and imagine some text in the far top-left corner of your visual field, and some other text in the far bottom-right corner of your visual field, and perceive both clearly, in a way you wouldn't be able to focus on two pieces of text in opposite parts of the visual field in the external world?
What if it's one piece of text, but it goes all the way across your visual field (ie it's in very big font)? Does that make it easier, or not?
Haven't been here in a couple years, is there still culture war thread here or is that totally gone now?
Is there a protocol for drawing Scott’s attention to commenters that should get a warning or banning? It’s also quite possible I’ve missed a FAQ explaining how to do this or why not to do this.
As of August 1, the sale of marijuana becomes legal in Minnesota. Gummies containing 5 mg of THC have been legal for a little over a year.
There are some inexplicable exceptions to the older law though. For example you can buy a canned soda like beverage with 50 mg total. I think of the poor novice to THC intoxication knocking back two or three cans as if that was comparable to drinking 2 or 3 beers. That’s the sort of dose a cancer patient might use for pain and nausea. You need to build up to that level.
Yikes. Happy trails, my friend. I’ll get back to you in 5 hours or so. You might want to lie down for a while. Give me your car keys and I’ll put on some Allman Brothers.
Fortunately the new law will limits the content of canned beverages to 5mg of THC per can.
Sorry for the naiv question: what is the fastest way to find a concrete comment of mine on this site? It was on use of media compared to use of social media in Germany. I think this should contain enough keywords to narrow it down, but I haven't been successful so far.
Have there been any major, damaging computer hack/virus incidents where the perpetrator gained nothing tangible from the act? I'm looking for an example like the "Bank of America Hack of 1998 where Mike Smith hacked into the database and deleted 1 million accounts, destroying $250 million."
So Barbie is Eve in the garden of Eden.
Wish we had a kabbalistic review of that movie (and a kaballistic one of Oppenheimer too)
Playing with Llama 2, poetry completion seems completely flat compared to Llama 1. Coherence is up but I can't get the 13B model to generate anything better than unimaginative doggerel. Maybe my prompts need serious revision, or I need to move to the 70B model, but so far I am disappointed.
"Some of the best responses were Yes, Students At Elite Schools Are Actually Taught Different Things,"
When my kids were applying to colleges, I researched this very carefully. My conclusion is that the "Ivy League" did very little educationally that could not be found at the flagship State Universities, all of which have "honors" programs and all of which have large graduate programs where the same stuff is taught everywhere.
Since their grandmother wanted to pay their tuition, my kids went to a Midwestern Private R1 university that is always ranked with the Ivies. They got good educations, but, in my estimation, if they had gone to State, they would have wound up the same.
My nephew did go to an Ivy. He wanted to go to medical school, so he took all of his required science courses at a state university branch in his hometown so he could be more assured of getting an "A". He went to Medical school and landed a residency in a famous Eastern hospital.
I am firmly in the Bryan Caplan school on this one.
"several people ... pointing out that smart people want to socialize with other smart people."
Yes, but, there are plenty of smart people everywhere. Let us think this through with numbers.
Most IQ scoring systems have a mean of 100 and a standard deviation of 15. Applying this and a standard deviation calculator, to the approximately 5 million kids who turn 18 every year we find that:
IQ 115 -- 129 (definitely college material) 700,000
IQ 139 --144 (really smart professional/graduate school) 100,000
IQ 145 & up (I don't like the word genius) 7,000
Now the top schools. There are about 70,000 total undergraduates at the Ivy League schools. That would be about 18,000 admissions per year.
A good friend is a senior faculty member at an Ivy. From what he told me, and from what was publicized during the trials leading to the recent SCOTUS decision, Less than half of those admissions are made purely on the basis of academic evidence of intellectual prowess, the rest being, affirmative action, legacy, athletics, etc.
So, I would guess that no more than half of the incoming class of the Ivys consists of IQ >=130. That would be a total of less than 10,000 out of that 100,000. Take another tranche of 10,000 for the the non Ivy top schools (Cal Tech, MIT, Stanford, etc.) and there are still 80,000 really smart kids who have to go to other colleges.
I am willing to wager that any of the top 25 state universities has as many really smart kids (IQ >= 130) as the average Ivy. It is just that they are a smaller percentage of the student bodies at those schools. But, from a social viewpoint, it is not a real problem. They will meet each other in class and find the hang out locations where they can meet more.
Again, Bryan Caplan is right.
My conclusion is that the best reason to want to go to an IVY is the hope of meeting and marrying rich. As my father told me, a man can accumulate more wealth in a 15 minute wedding ceremony that he can in 45 years of honest toil.
Hello smart people of ACX. I'm trying to find the word for a concept: when you do something which you're explaining or talking about. Some examples:
"Oh my god, you don't even know what condescending means", said John condescendingly.
"You're great at manipulating those people by flattering them", said Kim to Tom, causing him to beam.
Google thinks I'm looking for simile, metaphor or analogy because I can't search properly. Whichever Chat GPT is free thinks I want "performative speech". Bard thinks "exemplify", which I think is the closest. Any ideas?
Scott spoke some time ago of TPOT, the postrationalists, and how they never gave any philosophical grounding to their stuff. I'm not really a part of that, but I'm quite taken with their idea of the vibes, and I wrote an anti- or meta-philosophical defense of vibes.
The Truth Is Literally A Vibe
https://squarecircle.substack.com/p/the-truth-is-literally-a-vibe
LW/ACX Saturday (7/22/23) The Fermi Paradox (Why don't we see evidence of alien life)
https://docs.google.com/document/d/1RirLmvmvVxzVS_m1zQ2jz2W1fgPyeHGAZ964CciCizU/edit?usp=sharing
Hello Folks!
We are excited to announce the 35th Orange County ACX/LW meetup, happening this Saturday and most Saturdays thereafter.
Host: Michael Michalchik
Email: michaelmichalchik@gmail.com (For questions or requests)
Location: 1970 Port Laurent Place, Newport Beach, CA 92660
Date: Saturday, July 22nd, 2023
Time: 2 PM
This week we are diving into the deep end of the Fermi Paradox, a 70-year-old question that has been surprisingly hard to answer when people have deeply investigated it. The universe seems to be a place that could be filled with intelligent technological life, why haven’t we observed any? There are a lot of simple-sounding solutions to this problem, but upon deep scrutiny, many of them don’t hold up well or rely on factors we just don’t have hard numbers for. Most of the viable solutions point towards disturbing conclusions.
Google Docs
LW/ACX Saturday (7/22/23) The Fermi Paradox (Why don't we see evide...
https://docs.google.com/document/d/1RirLmvmvVxzVS_m1zQ2jz2W1fgPyeHGAZ964CciCizU/edit?usp=sharing Hello Folks! We are excited to announce the 35th Orange County ACX/LW meetup, happening this Saturday and most Saturdays thereafter. Host: Michael Michalchik Email: michaelmichalchik@gmail.com (Fo...
[3:55 PM]
Follow the link for the full message
Re: Aella, but also as a statement of general grumpiness-- what is with the general tendency to treat self-report, self-selected internet polls as meaningful empirical evidence for real-world phenomena?
That goes especially intensely for polls being used to adjudicate culture war questions with strong allegiances among the Very Online, like trans/ poly/ LGBTQ issues. Since it's usually very obvious which answers would yield results that flatter Our Side ("Why, uh, yes, I really definitely *am *a cis woman! And *of course* I get super aroused by the thought of pawing my own boobs while wearing platform heels and a tiny bimbo skirt!"), it seems implausible that those polls wouldn't get a ton of brigading and possibly bot mobilization to give false answers and skew results in the most convenient direction.
Which, frankly, is the type of critique I practically learned to make from reading SSC/ACX. So can anyone explain the serious takes around here?
dating ad: 26 year old woman looking for a slightly older male long-term partner
— am an Australian English/history teacher who enjoys reading (most on Bloom's Western Canon or any text written before the 20th century, miscellaneous articles on anything technical, a lot on Gutenberg and Archive.org, sometimes ArXiv too), classifying transport models, exercising, and travelling. wants children (have worked with them and teenagers for almost a decade, and I think they are the best)
— looking for somebody who is also averse to TikTok and Netflix and other numbing agents. somebody who enjoys reading widely (interest in history, tech, military affairs, and transport e.g. aircraft or trains would be excellent), explaining technically dense processes, travelling, a night out dancing, and Norm Macdonald. strong preference for engineers
lilyreadsyouremail @ gmail . com
What is the most negative Rick Steves travel video review?
The following is a proof-of-concept: a quote-supported opinion generated by a PIM App.
Looking for feedback on both content and format. Thanks!
A Balanced Sense of Self
https://1drv.ms/b/s!Av3DdRPJXjSngTvaHplrV0yFbGlz?e=gs89Fi
[MS OneDrive, recommend downloading]
Dear ACX readers,
I've just written a post on how to foster Antifragility - I think you will all enjoy it:
https://open.substack.com/pub/zantafakari/p/17-be-antifragile-by-expanding-awareness?r=p7wqp&utm_campaign=post&utm_medium=web
Please leave feedback so that I can improve. And if you like my writing - share, like and subscribe!
Here's an article in the New York Times about the future of Wikipedia in the age of ChatGPT. In addition to its topic, one notable feature for this forum is that the first source who is discussed in the article is presented only by his Wikipedia username, "Barkeep49", and the author explicitly states that he isn't giving this Wikipedian's real name, to avoid harassment.
Notably, this article is by Jon Gertner, rather than Cade Metz, who is the one that wrote the article that gave Scott's legal last name (https://www.nytimes.com/2021/02/13/technology/slate-star-codex-rationalists.html).
I'm not sure if the difference in policy is an official change at the New York Times, or the fact that this article is in next Sunday's "New York Times Magazine" rather than a regular daily edition (there could be separate editorial policies for the two contexts), or just about the different attitude of the two authors. (You can see more of their articles at their respective pages. Note that Gertner has his own site, because his articles are published in different venues, while Metz seems to be just at the NYT. https://jongertner.net/category/article-archive/ https://www.nytimes.com/by/cade-metz )
https://www.nytimes.com/2023/07/18/magazine/wikipedia-ai-chatgpt.html
"I plan to ask Michael to design the questions for the next survey and demonstrate that they get the same result." Doesn't announcing how you'd like your test subjects to respond mean that your results will be invalid?
Impact Academy is looking for a Regional Director and a Regional Associate for our upcoming programs: https://www.impactacademy.org/joinus
If your dream job is to help others do the most good they can, being part of an ambitious start-up, and leverage your expertise in India, then this might be the opportunity you have been waiting for!
These roles will greatly influence the development of our potential flagship program and our general efforts in India and the Global South, and your work may set an example for other regional offices worldwide!
Deadline: August 4th. We evaluate applications on a rolling basis and might fill the role before the deadline, so we encourage you to submit your application as soon as possible.
If you know someone you think might be a good fit, please send this Job Description to them. There is a $500 prize for referring a successful candidate.
Apply for both roles here: https://www.impactacademy.org/joinus
I'm running low on non-fiction books I'm interested in reading, wondering if anyone has any suggestions.
In the west, smarter people tend to be left wing. Are there countries where high IQ is correlated with being conservative/on the right ?
Alright, I'm going to attack a core tenet of the Rationalists that has always bugged me. As I understand it, Rationalists claim to be slightly better at avoiding irrational, ingrained-by-evolution biases than the average person (even of the same intelligence). The Overcoming Bias image of Odysseus strapped to the mast to avoid the possibility of succumbing to the Siren's Song emblazons the belief that this is doable.
The belief one can overcome such ingrained biases strikes me as ludicrous. I'm reminded of Carl Jung saying that trying to overcome the unconscious with the conscious is like holding a mirror to the sun.
Whatever deep ingrained biases you have that are shaped by evolution are As Smart As You! Probably smarter. Your *rational* thought is not independent of such overwhelming forces as billions of years of evolution.
If you think you can outsmart your biases through concentrated reason, you will not only lose that battle, but you will have made yourself (forgive my frankness, Scott) stupider by thinking you can.
Some of my online friends believe that the solution to global warming is to eliminate capitalism and/or eliminate civilization.
Do you want to eliminate capitalism?
Do you want to eliminate civilization?
Please answer in any way you choose.
Peter
Hey, let's share embarrassing stories of a nerdy adolescence! I'll start:
Near the beginning of Grade 7, Mr Barlow the English teacher had us all submit two topics for short speeches. We would draw two topics at random out of a hat, and give a short talk on whichever of the two topics we chose.
My parents were moderate packrats, and a year earlier I had found a 3-year-old LIFE magazine in a box in the basement, a double issue on the the American city. I was enthralled; the photos, the writing, the futuristic drawings, everything just grabbed my young brain. Besides the mostly optimistic vision of the future, there was a rather Malthusian section about urban sprawl, with the forthcoming megapolises gobbling up arable land, this in a time of exponential population growth.
I was (and remain) a space nerd, so the topics I submitted were:
After The Moon - What's Next For Manned Spaceflight?
and
Urban Sprawl!
*************
The hat was passed around to collect the topics, Mr Barlow mixed them up, and the hat came around again for us to select topics.
Mr Barlow gave us a couple of minutes to check out the topics we'd pulled out of the hat, and then asked if everyone was happy with them.
A tough girl named Lori* raised her hand and said "I got one called Yurbin Sprail, and I don't even know what it is!"
Mr Barlow glowered and asked who had submitted it. I raised my hand. Lori and her peers sneered at me and snickered. Mr Barlow got mad and asked me if this was a joke. I stumbled through a brief explanation, and must have seemed sincere enough that Mr Barlow realized I had submitted the topic in good faith.
I don't remember what topics I had picked from the hat, but no doubt both involved pop music I hadn't heard, sports I didn't know anything about, or TV shows I hadn't seen.
Those who wish they could return to their teen years were better equipped to deal with adolescence than I was, or have shorter memories.
* Lori was part of a gang of girls who smoked, stole from the convenience store across the street, and used a lot of profanity. Our rookie homeroom teacher, Mrs. Lawson, took a leave of absence partway through the year. She was gone for a month or so.
I'm trying to learn more about free banking. To that end I'm reading George Selgin's "The Theory of Free Banking" (1988). I'm not an economist but am an econ-nerd, esp. on monetary policy, and got into Scott Sumner and market monetarism back in the day.
I don't think I quite get it. Selgin says that in the Scottish free banking era, banks were stable keeping only 1-2% of deposits as reserves in specie (I say specie here but it could be a base fiat currency as well), and they typically did not suffer from runs and failures. Why does the existence of private banknotes allow this? Is it because the base money and banknotes are no longer the same? If depositors get skittish and want to pull out in a textbook run scenario (regardless of whether those are actually real), how would the bank not quickly run out of reserves? From a financial stability standpoint, I sort of intuit that with banks printing their own money, the money supply endogenously rises and falls with liquidity demand, and insofar as financial crises are generally caused by a contagious liquidity crisis, self-printing of money would prevent runs and instead allow for private money to simply trade at a discount instead if a bank is insolvent (i.e. similar to how there is no such thing as a run on a stock because the price trades freely). But I'm having trouble putting all the pieces together.
I think I may need a plain English explanation to supplement the more jargony one.
Are there any known techniques to improve general memory? I know a bit about the sort of repetition learning and the stuff Gwern has written about, but I more mean memories of experiences. Mine is terrible… it has its pros and cons, like I don’t really get traumatised because I just forget, but I tend to repeat the same mistakes pretty often. My general patterns of experience also seem to get repeated quite a lot.
Any ideas?
Assuming that all of the QAnon predictions/prophecies were accurate (especially the ones that contradict each other), what is the current state of the world?
...well, I've got a sudden bug and you folks get to hear about it.
Recently watched Paranoia Agent, an anime about a mysterious attacker in which every episode follows a different character up to their run-in with him. Started with Boogiepop Phantom, which has the same style of episodes following the various people that run afoul of her, but Boogiepop Phantom is not holding up to the latest viewing. Looking for other stories that have episodes revolving around minor characters who are themselves revolving around a central character. Google mentions another anime called Hell Girl, but it seems to not be available in the US.
Does anyone know any other stories in that style? Any medium.
Re: the recurring subject of Bay Area "NIMBY" vs "YIMBY" wars:
Can anyone offer links demonstrating the reality of the much-discussed shortage of "affordable" (for people employed on location) housing there? A cursory look at various ads (browsed SF and even Palo Alto, arbitrarily close to various TBTF corp. headquarters) seems to show plenty of 1- to 3-room flats going for 3 to 5 thou. $ / mo; perhaps tight quarters, but not out of reach cost-wise for even an entry-level programmer. And not particularly more expensive than e.g. the habitable part of Washington DC.
There was a trend of the Open Threads having other foreigners asking "is America really like that?". I guess I'll take my turn.
As a Canadian, something I see a lot in TV is Americans wearing their shoes in their own house. When people don't do this, it is usually called out. I've also seen a few shows refer to it as a Japanese practice.
Do all Americans do this? Is it a regional thing?
RE the autogynephilia question, assuming the results are correct, it seems to suggest that it is somewhat associated with being trans, but more importantly that also that a lot of cis men have it/engage in it.
From a practical/trans rights point of view this is a mixed bag result, because the fact that lots of cis males engage in it definitely makes the argument that cis males can use very loose self-ID rules in locations like bathrooms and changing rooms and spas to involve cis females in unwanted sexualized role play of their own fashioning. Trans rights activists want to laser like focus on the idea that 'real' trans people are no more dangerous than cis people. But the problem is that there are lots more cis people. So if autogynephilia is semi-common among cis males, loosening the rules on previously sex segregated spaces ends up potentially exposing a lot of cis females to a lot of cis males.
This may still lead to different answers in different places (most female bathrooms tend to be single stall oriented so maybe no big deal while maybe naked spas could still be sex segregated).
In any case the finding that it isn't *particularly* associated with being trans doesn't answer the question of what we need to do with the cis guys who are autogynophile (in female spaces).
What are some theories for why top American politicians are so goddamn old now?
My spur of the moment hypothesis is that, lacking other qualities, oldness is the closest proxy left for presidential.
Another is: in a feminized era where increasingly more men speak with high-pitched voices (see: Ron DeSantis), only really old guys have any gravitas.
I often watch TV with my girlfriend in the evenings. In principle I think a particular type of video game would be a better use of this time, but I don’t know games - can anybody recommend stuff which satisfy the following criteria:
- me and my girlfriend can play it together
- has artistic/literary value
-is some kind of big adventure
Basically I want something like an Elder Scrolls game or Red Dead Redemption, but which me and my gf play together.
Thanks
I feel like I remember that Scott linked a Noahpinion tweet of a graph from a paper comparing parenting hours spent by men and women across time, which showed that men today spend as much time on parenting as women did 50 years ago, but that women have similarly increased their time, so the gap between men and women has stayed ~the same. I can't for the life of me find either the tweet or scott's link to it.
Does anyone happen to know/remember what I'm talking about and where to find it?
Whatever happened to Rodney Brooks "Intelligence without Representation"-style robots? Wikipedia says it's correctly called "Subsumption architecture" (or maybe "behavior-based robotics"). Was it a dead end? How complex did their behaviors end up becoming? Is it still an area of active inquiry?
Blog post on a novel but flawed libertarian exegesis of Romans 13:
https://thearchy.substack.com/p/the-need-for-good-libertarian-exegesis
New blog post exploring the case against the FDA regulating general purpose medical AI and the growing need for decentralized approaches to oversight.
"Why AI doctors should not be FDA regulated"
https://moreisdifferent.substack.com/p/why-ai-doctors-should-not-be-fda
How important is proofreading in languages other than English?
I was reading the newest book review on The Educated Mind and noticed a place where the reviewer mixed up the words "good" and "well." This on it's own isn't very interesting; even extremely intelligent, native English speaking, professional writers who make a conscious effort to proofread their work make mistakes like this with such frequency that most of us sail right past them without a second thought. But that normalcy was striking to me. I am a native English speaker, with 19 years of education mostly spent on improving my reading and writing ability under my belt, who makes his living using the written word to persuade, and I would be shocked if this short paragraph didn't contain at least one error in grammar or punctuation. I can't remember the last blog post or newspaper article I read that didn't include multiple typos. I have no fewer than three separate books on my shelf on English style, grammar, and composition, which my employer considered important enough to my job to pay for as a business expense.
I've always taken this state of affairs for granted, where nobody really knows a lot of the rules and even the most literate among us mostly get by on extremely-error-prone intuition and are forced to spend significant amounts of time proofreading if they want to be clearly understood, but I'm also aware of just how weird/complex/idiosyncratic the English language is compared to other world languages.
For those of you whose native language is not English, do fluent and even skilled writers in your native language seem to struggle with grammar, punctuation, and spelling to a similar extent English writers do? How common are the sort of minor errors that most of us have learned to politely ignore when reading English-language newspapers? Are there different sorts of pitfalls than the ones I'm familiar with as a native English speaker? I'm especially interested in non-alphabetical writing systems like Chines, Korean, Japanese, etc.
I have been working on a new *style* of blogging, and it might be ready for previews: https://shragafeivel.com/
The synthesis of your post regarding admissions and your link from reddit about advanced math classes, is this:
Harvard et al really do have a big share of the smartest, most driven students. But that isn't the bulk of the student body. The bulk is kids who are less smart/driven and there in part due to privilege.
Sure, great that you can take grad-level math classes as a junior, but realistically most Harvard kids don't do that.
Part of the advantage of going to Harvard is the ambiguity. Are you "pothead child of hedge fund billionaire", "somewhat smart kid of two doctor/lawyer/etc types", or "Mr. math genius, take-multiple-graduate-level-math-classes-as-a-junior"?
Also - just to say it - I think the last group will be disproportionately from well-off families, just because that sort of ability isn't only about talent, but also about having it cultivated from a young age. There are perhaps "doing calculus in preschool" level prodigies out there for whom it doesn't matter, but they're a *very* small slice, even smaller than the "Mr. math genius" group.
Substack has become extremely annoying.
On my Substack reader app, there is now an annoying feed. They try to put in links to blogs they think I might like. And one-line comments much like tweets keep popping up. It is distracting and annoying. I cannot find what I subscribed to and want to read.
I deleted the app.
(I got here from my browser. )
So they're coming out with a new Willy Wonka movie, a prequel: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwjkod2f9ZWAAxUyFTQIHeDqC_EQ3yx6BAgQEAI&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DotNh9bTjXWg&usg=AOvVaw1RftwKs8aMeFnIOtj_DKoV&opi=89978449
It may be entertaining to watch, but I don't expect it to be the caliber of Willy Wonka's Chocolate Factory, which remains one of my favorite movies, and which I consider better than the book. In fact, watching the trailer made me think that the script could have been generated by an AI. So...I asked ChatGPT, which only has training data before 2022, to tell me the plot of such a prequel.
I actually like ChatGPT's plot a little better, but found, as expected, it didn't really have much new. It told of Willy's adventures, with his friend Matilda, went in search of magical confections, including Vanilla Sprites, Sugar Sylphs, and the finding of oompa-loompas.
I asked it to expand on Sugar Sylphs, and found they lived in clouds, which Willy and Matilda reached using some kind of magic rituals. Though I kind of liked this idea, I found it missing the essential ingredient, as it were, of what made the original movie so great, which was the zaniness. I suggested that Willy and Matilda should, instead, reach the Sugar Sylphs' clouds by eating the right combination of specially-crafted sweets, to obtain a sugar high. This seems in-line with Willy mixing a coat into a batch of candy because it was too cold, and some shoes in another batch to give it a little kick.
This illustrates to me why LLMs like ChatGPT can't yet replace humans. Hollywood currently has a writer's strike going on, protesting AIs. If the writers only produce what AIs can produce, then Hollywood has a point. If they can do better, generating actually good content, even with the help of AI, then Hollywood doesn't realize their true value.
Though I have never written a screenplay, I bet I could do better than the prequel coming out in December, with help from ChatGPT. A professional screenwriter ought to, on average, do better than me.
How would the US military deal with minefields, like what the Ukrainian military is dealing with now? Like, how would they clear them? I understand US dominance is based on air superiority, but you do have to cross the actual minefield at some point- right? I read the Wiki page on demining, but it just lists general tactics.
Inspired by the Wiki page and the Ukrainian struggle, I (a non-engineer) came up with a couple of random cheap demining tactics. These are more in the category of 'shower thoughts' and not like a serious proposal, but just for fun:
How about a pressure washer to trigger them? A cheap remotely-controlled vehicle like a tractor or excavator can carry a commercial pressure washer that sprays the ground in front of it. The amount of pressure a pressure washer can generate is amazing- like more than a human's weight- so should be enough to trigger the mines. Just keep spraying the ground and advancing. Will some of them get damaged? Sure- just roll out the next cheap tractor/pressure washer combo and keep going. Water supplies might be a hard limit.
Alternately you could just do an agricultural-style controlled burn on vegetation, where the heat should trigger all the mines. Do a spray of a cheap propellant (again via a remote-controlled tractor), the burn will detonate buried mines, if the tractor gets damaged replace with another, and keep advancing. Lots of farmers do controlled burns in their fields, not a high-tech problem
David Brooks had an interesting piece in the NYT about Douglas Hofstadter changing his mind about AI. He links to a LW article that Gwern posted with a transcript of an interview with Hofstadter.
DH: It “just renders humanity a very small phenomenon compared to something else that is far more intelligent and will become incomprehensible to us, as incomprehensible to us as we are to cockroaches.”
Brooks was surprised by the interview and the change in thinking and called Hofstadter to talk about the change.
Brooks OpEd
‘Human Beings Are Soon Going to Be Eclipsed'
’https://archive.ph/3kref
Brooks reads LW??
:wave: The Roots of Progress founder here, AMA about the blogging program.
The standard evolutionary explanation for morning sickness is that, in combination with hyperosmia, it deters pregnant humans from eating questionable food during the first trimester when embryos are sensitive to teratogens, thereby serving an adaptive role by compensating for the cost of not being able to eat very much or do much of anything except lie in your cave feeling miserable.
I think this is a plausible explanation for hyperosmia, but fails to explain species-specificity and timing of nausea and vomiting. Routine vomiting during gestation has barely been reported in non-human animals, including other omnivores (rats don't vomit at all, some reports of mild food aversion in gravid dogs and cats); I don't know of any evidence that human embryos are more vulnerable to environmental teratogenesis, and if we were I'd guess it would be during the the late first and throughout the second trimester during the most intense periods of neurogenesis and neuron migration to make our abnormally complex brains, whereas morning sickness is usually worst around weeks 6–14.
My alternative hypothesis is that the adaptive value is as a stress test of the social caloric support system. Humans are so social and invest so heavily in each offspring that it's almost impossible to survive and successfully complete pregnancy and lactation unless you have conspecifics bringing you lots of food. Therefore, instead of being selected to minimize hormonally-induced nausea like other mammals, early humans found that it increased lifetime fitness if they stopped being able to feed themselves for a month or so right at the beginning of pregnancy—and, if this test revealed that they were in a situation where nobody was able or willing to feed them, to cut their losses by having malnutrition-induced miscarriages before going through the extremely dangerous process of late gestation and childbirth, and to try again later.
This alternative hypothesis explains not only why our bodies would do something with such obvious fitness costs, but also why other animals don't do it (except perhaps those that rely on humans feeding them when they feel bad!) and why it happens essentially as soon as the body is confident that it's pregnant.
I write a Substack newsletter where I share three things I find interesting, once a week.
In the latest issue I shared:
1. A study finding an unexpected overlap in the demographics of BLM and anti lockdown protests in the USA
2. A chart demonstrating the U shaped relationship between fertility and household income
3. My mapping of median income per capita against average yearly working hours across a number of countries. This shows among other things that the higher the incomes, the lower the average working hours, with the exception of Korea.
If this sounds like the sort of thing you’d be interested in, you can check it out and subscribe using the following link: https://open.substack.com/pub/interessant3/p/interessant3-45
"Progress" is *such* a loaded word these days.
You know, something that I haven't heard mentioned in the autogynaphilia debate, but which I think supports the mainstream trans-as-dysmorphia story, is furries.
The furry community plays host to people with a wide variety of paraphilias- attraction to characters with animal features, transformation, vore, micro and macro, etc. But what really connects people in the community isn't the kinky stuff- it's the common longing to be something other than human. Almost everything that makes a furry a furry- the fursuits, the OCs, the role-playing- in an expression of that desire, and finding people who share it is the main reason people become furries. Functionally, the furry community is a social network for people suffering from species dysmorphia.
So how does the autogynaphilia model account for that? Are they supposed to secretly be zoophiles pretending to be dysmorphic? Obviously, that doesn't make sense. Why should one socially unaccepted paraphilia masquerade as a different one? Also, many of the fetishes associed with the community, ilke micro and macro, have nothing to do with animals- instead, the common thread is an attraction to being something not quite human.
It's generally assumed in the community that the fetishes arise from the dysmorphia. That's supported by the fact that most furries' obsession with being non-human started before puberty. It's also just a lot more parsimonious than the alternative.
Finally, furries make no public claims of dysmorphia- most still try to insist to outsiders that it's just an anthropomorphic animal fandom. So, if dysmorphia is supposed to just be posturing for social acceptance, why would they only express it in private?
I'm one of the straight cis male respondents who gave a non-zero answer to the autogynephilia question, and I still don't think I have autogynephilia. Let me explain why.
In a question of the form "Picture a very beautiful woman. How sexually arousing would you find it to imagine <X, where X involves her>?" I am never not going to be aroused by that, unless X is actively disgusting. X could be "her standing in the corner holding textbooks and pretending to be a bookshelf", "her attached to the end of your foot like a shoe" (sorry), or "her as a gigantic godlike figure holding up the night sky" and I wouldn't answer zero. I'm not attracted to X, but I am attracted to beautiful women doing any old thing.
The answers of respondents probably differed more by how they read the question than whether they really have autogynephillia. Some of the cis attracted-to-women respondents are either disgusted by the premise, or answering the question relative to their base rate of being aroused by things. Some of them are answering it literally. Most/all of the trans respondents are disgusted. This entirely explains the weirder results and also invalidates any conclusions drawn from them.
Hello everyone,
Last year, after reading one of the ACX book reviews, I built a short survey about inner voice and memory (mostly from existing scales), to test a hypothesis that seemed likely to me after reading this review. After analysing the data, well my hypothesis was totally wrong, but there seems to be something unexpected and interesting going on in the data. But the number of responses I got is slightly too low for me to draw firm conclusions.
If you have a few minutes to spare (about 5/7 minutes), would you agree to answer the survey below? If I get a few more responses, I can finish the analyses and post the results if people are interested. Thank you so much :-)
https://shorturl.at/jnG19
I finished 'Of Boys and Men' By Richard Reeves. I didn't like it so much. Here's my Goodreads review.
2/5 Stars
The first part of the book deals with the gender gap in outcomes between men and women. There are some interesting figures here, but the basic gist of the first part succumbs to "gapism" ie., listing a bunch of gaps in outcomes and declaring that an argument was made. "There aren't as many male nurses, there aren't as many male teachers!" Reeves, says. I'm sure he's right. But he never questions the medical or educational systems directly - not their organization, not their waste and mismanagement, not the government meddling, not their enormous subsidy.
Reeves spends the chapter on solutions basically arguing for even more subsidy, even more affirmative action, and even more meddling. He anticipates that the reader will criticize him from the left, "Maybe you'll want to report me to the EEOC." Yea, maybe. "What about all of the affirmative action for women!" He says. Yea, I don't think so.
The question from the outset should be: What are the structural problems in the education and medical systems that can be improved at the benefit of everyone? I think redshirting sounds like an ok idea, but the rest of the solutions discussed in this book border on total social and labor market planning - which got us into this mess in the first place. I'm skeptical that more of it will get us out.
Is America a nanny state?
I ask because America markets itself as this rugged individualistic country.
But then I hear of people calling the police because they say a kid walking to school or what’s weirdly called a “welfare check” where if someone doesn’t answer your calls you claim they are suicidal and police smash their door in(yes I’m exaggerating but only a litttle; look at all the swatting incidents).
In the UK and most of Europe calls like this would be treated as wasting police time and you would get a warning. So why are they common in America?
Is it because , contrary to the marketing , America is actually a nanny state?
About a week and a half ago I published my first article on Substack:
https://samschoenberg.substack.com/p/belief-and-the-truth
It's about my own reflections on the Ideas of Belief and Truth, and about whether It's truly possible to separate the two.
Ha, I'm out in Berkeley this week, but will have to miss the 7/23 meetup... as that's the day I'm returning to New York! So, see you in New York, I guess. :)
Rational Animations recently animated "The Goddess of Everything Else": https://www.youtube.com/watch?v=Bbwp4PbWYzw
(Full disclosure that I am currently doing some writing for RA, but I had no involvement with this video)
Has anyone observed a correlation between niceness/kindness/generosity and lower IQ? I’m not talking about people who are nice enough. I’m thinking of someone who would be first and foremost described as nice or kind by their friends.
On the other hand of the equation, the “term” calculating is associated with being unkind or inauthentic. But from a rationalist pov being calculating seems like a compliment. Why shouldn’t I use my intelligence to make smart decisions about my relationships?
I was thinking this might have to do with a social expectation that relationships are more authentic when they are based on emotions instead of logic.
Why are alleyways (also called laneways) always so ugly? Or are they perhaps not ugly where you are from?
What time Sunday for Berkeley? Afternoon? I might be able to make it.
I've added a fun Friday type entry for my "blog" https://icdstories.substack.com/. Now every Friday I ask GPT4 to make up an ICD-10-CM code and a vignette to go with it!
Whom will Biden appoint to the UFO Disclosure Review Board as economist and sociologist?
(4) QUALIFICATIONS.—Persons nominated to the Review Board— (A) shall be impartial citizens, none of whom shall have had any previous or current involvement with any legacy program or controlling authority relating to the collection, exploitation, or reverse engineering of technologies of unknown origin or the examination of biological evidence of living or deceased nonhuman intelligence; (B) shall be distinguished persons of high national professional reputation in their respective fields who are capable of exercising the independent and objective judgment necessary to the fulfillment of their role in ensuring and facilitating the review, transmission to the public, and public disclosure of records related to the government’s understanding of, and activities associated with unidentified anomalous phenomena, technologies of unknown origin, and non-human intelligence and who possess an appreciation of the value of such material to the public, scholars, and government; and (C) shall include at least — (i) 1 current or former national security official; (ii) 1 current or former foreign service official; (iii) 1 scientist or engineer; (iv) 1 economist; (v) 1 professional historian; and (vi) 1 sociologist.
Last week I published a report on the anonymous salary survey in Berlin, Germany. Tech bias, but maybe still useful for some of you. It also has an interactive dashboard. I am also open to suggestions, collaborations and other improvements.
Check it out:
https://handpickedberlin.com/report-on-berlin-salary-trends-survey-june-2023?utm_source=act&utm_medium=comment&utm_campaign=report_release
Wrote a very short story about what can happen when AI gets too integrated with people:
https://ageofai.substack.com/p/for-the-greater-intelligence
The story could use some improvement, but I haven't found ChatGPT to be that good at writing stories yet...
I only started reading ACX after the move from SSC. I have gone back and read the top posts from SSC but still felt like I was missing out on some of the old classics. I built a website to resurface old content from blogs by sending weekly emails. Let me know if you have any suggestions of other blogs / content you’d like to see, hope you find it helpful!
https://www.evergreenessays.com/
Have you ever felt like you have a song "stuck in your head" while you're doing other things?
If you usually think in words, and the song had lyrics, does that mean you felt like the song lyrics were going through your head at the same time other things were going through your head, ie you had two parallel internal audio streams?
Can you voluntarily give yourself two parallel internal audio streams, ie play two songs with lyrics in your head at the same time?
Can you voluntarily give yourself two parallel internal visual streams, ie visualize a tiger and a sunflower at the same time? If yes, how sure are you that you're not switching back and forth between them very quickly? How many internal visual streams can you maintain at once?
Can you focus on two different areas of your (internal) visual field? IE can you close your eyes and imagine some text in the far top-left corner of your visual field, and some other text in the far bottom-right corner of your visual field, and perceive both clearly, in a way you wouldn't be able to focus on two pieces of text in opposite parts of the visual field in the external world?
What if it's one piece of text, but it goes all the way across your visual field (ie it's in very big font)? Does that make it easier, or not?