@scott my browser just killed 10-20% of my smartphones battery while having this thread open. Then it got hot and crashed. Now my browser shows black and white bars. Please look for another, better way for your readers to have seamless discussions. There is plenty of software and services way better then the current substack crap.
Can someone give me an explainer on the Chinese social credit system(s)? Where on the line of "this is a widespread totalitarian thought-policing system" vs "this is just a 21st-century better business bureau and worries about it are just a red scare." What are the good sources to read about this that aren't super partisan?
So what will be the next political craze? Identity politics has been in for the past 5 years but seems to be on the wane. Maybe war with Russia will take its place? Or maybe another 3 years of fighting over vaccines and masks? What else seems likely? As a long shot bet, I'll buy some abortion-war tickets.
Has the present ever felt so absurd -- not a certain era in retrospect, but as it is happening -- or does it always feel this absurd to people as they start to age?
The latest episode of Always Sunny in Philadelphia (which just ran on FXX) is one of the best anti-woke takes from liberal Hollywood. If you think the left can't be anti-woke, I suggest you watch it.
In it, they attack viciously how some of their earlier episodes were erased by woke Netflix because they used black-face, they lean in hard on other racial, classist (white yokels) and other sexual stereotypes acceptable to contemporary moral sensibilities, to point a pointy stick in the eye of hypocritical leftists, and actually make it all very funny.
Wokism is dying and Hollywood is taking the lead in killing it.
I would be very grateful if anyone could help me find an article that I'm 80% sure Scott linked to a few years back: it was about a community mental health worker in India and the general effectiveness of a community-based social work-like approach to mental health, where people with relatively short training can help a lot of others, as opposed to the way that licensed therapists need many years of training and a lot of people can't afford their help. It also mentioned some research done in the 60s and 70s, that showed how these two approaches had comparable results and suggested that the research was kept quiet and then stopped because of lobbying from the licensed professionals.
What's the conclusion on face masks after all the new evidence? This site seems pretty shady but links to several prestigious seeming studies claiming that there is no benefit - https://swprs.org/face-masks-evidence/
On the other hand there are several sites that do claim effectiveness. Scott's old post took source control almost for granted, but that seems to be up for debate now as well.
How much less credible is a doctor's weight loss advice if the doctor himself is still obese? I think it would only be the ad hominem fallacy if one put too much weight (pardon the pun) on that one data point, or applied it where it wasn't applicable.
I've noticed that in the U.S., the GDP deflator and real GDP have been growing at the same pace for a surprisingly long time. This has not been the case in Italy (where the GDP deflator has grown much faster than real GDP) or Japan (where the reverse has been the case). Does anyone have any thoughts why increases in nominal incomes have been equally reflected in real GDP and prices in America for so long?
I take an SSRI and also Adderall. The SSRI prevent reuptake of serotonin, the Adderall prevents the reuptake of dopamine and norepinephrine (if I understand correctly). Could this mean I have low qualities and/or too high uptake of neurotransmitters generally?
I've been reading the dialogues between Eliezer Yudkowsky & MIRI and various members of LessWrong on LessWrong, and my current model of Eliezer's belief is roughly "on current trajectory there is over 90% probability that some form of advanced artificial intelligence will kill all humans and it most likely will happen during this century" (if you think my model of Eliezer's belief is wrong, please tell me).
Since most humans during the history of the world have died, I've made my peace with the fact that I too might die. I think it's most likely that everyone will go to nothingness after they die though there's a nonzero chance that I'm wrong about this, but I think everything we know about neuroscience, cognitive science and physics points towards nonsurvival after death.
I think death is negative in the sense that you are deprived of all the positive possibilities of life, but once you are dead, the condition itself is neutral since there are neither positive or negative things in nothingness, there's just nothing, like dreamless sleep.
What I'm concerned about Yudkowsky's scenario is whether there is going to be suffering, pain and other bad things leading up to it. Do you think there will be such things?
How does one participate in politics without an ideology? I have plenty of strong opinions about policy but no ideology. I voted against Trump because that seemed like a no-brainer, but normally I don't vote because both options seem equally bad. Third parties seem worse.
Some people get really involved with specific issues, but I don't care about specific issues all that much. I just want the center to hold in US politics. I mostly fear the country becoming too radical on either the right or the left, but I consider the danger of radicalism from either side to be roughly equal. Or rather, I consider the greatest danger to democracy in the USA to be radicalism itself. I suspect that the more radical the left becomes, the more radical the right will become and vice versa.
I suppose I can try to play the role of the median voter who votes against whichever side I think is going too far. It's not a clever role to play, but maybe it's the best.
What are some battles where ultra well-built castles were overrun by enemies anyway, and through brute force rather than some clever weapon or tactic that negated the castle's defenses?
I want to know about the most spectacular, high-casualty battles where great castles fell.
We talked a bit about the multiple meanings of the word‘literal’. Here is another one that has multiple meanings.
The verb 'bemuse' and the corresponding participial adjective 'bemused'.
When I first leaned the meanings bemuse meant to make utterly confused or muddle, as with an intoxicating liquor.
Now bemused is often used as adjective meaning something similar to amused.
Okay, fine, I get it. Word meanings change with time. The trouble is that it is not always clear which way the author intended to use it. Context sometimes makes it clear but not alway.
Not clear why it matters what Alexandros thinks about the meta-analysis here.
Let's take someone who might actually have heard of "meta-analysis" before yesterday: Andrew Hill. His original meta-analysis suggested that ivermectin worked, but later when he and his colleagues excluded the fraudulent and problematic studies, he admitted that the effect (if any) is tiny. https://www.medpagetoday.com/special-reports/exclusives/95333
Are there any good studies, analyses, or news articles that show what the cost-performance trend is for virtual reality goggles over time? In short, at what rate are their prices declining each year?
I'd imagine one metric would be shown on a graph whose Y-axis was "Real cost per display pixel" and whose X-axis was "Year".
Every time we have a non-political Open Thread, I think about a political topic or two, and regret being unable to post them. Every time we have a political Open Thread, I forget what I wanted to write about. (So today I decided to write a meta comment instead.)
Did alexandros ever justify his decision to summate multiple outcomes in his meta analyses? I know several people brought this up but I dunno if he answered.
I think it's strange that there have been all these great families brought up but I haven't seen any famous twins yet. The twin rate is about 2% so it's starting to get odd that there hasn't been one example yet.
On the great families subject, I wonder if the Plinys Elder & Younger would qualify? I guess people from ancient Rome are less relevant given the lower population and stronger class effects (eg, most of the population was slaves & plebians who would have had no opportunity to become famous historians)
So within the last couple years we've had two fairly prominent instances of major websites (Tumblr a while ago and Onlyfans more recently) having to scrub themselves (to varying degrees of success) of all pornographic content. From what I can tell, the banks that control those site's loans were breathing down their neck or something like that, basically threatening to back out unless changes were made. My question is, why isn't there a bank that's willing to allow that kind of content on their investments? It seems like an easy way to make a lot of money, since porn is a (quick Google search) $12 billion industry, which seems like a lot of money to be left on the ground. A bank that branded itself as open to those kinds of investments and precommited to never change that policy should have all the major players flocking to it, just for the stability and peace of mind if nothing else.
The Philadelphia ACX Meetup will be hosting a Solstice Celebration on Tuesday, December 21 in Ardmore, PA. You can request to join our Google Group here for full details: https://groups.google.com/g/ACXPhiladelphia
I am getting a new lawnmower. We are considering gas-versus-electric right now. I am willing to take a bit of a hit on price on convenience for environmental concerns, to a limit.
It seems the best reason to go electric is that a lawnmower doesn't have a catalytic converter. My lawnmower might be spewing out all kinds of harmful stuff, so in theory electrifying my mower before my car is a good idea.
But how can I quantify this? And how bad is it? I probably only burn 5 gallons a year in the mower. What exactly am I spitting into the air from a normal mower at that rate, and how much would I expect to pay to remove that stuff otherwise?
Are there any American Expats in Taiwan? I've been considering making a move out there to live and work and was interested in getting an American perspective.
I'd want to move there to take a position as an Engineering project manager, and live there permanently
Some time back, Scott wrote about polygenic screening and the first PGT baby. In the piece, he mentions,
"Screening companies will give you the raw data if you ask for it, so if you want to screen for an embryo with green eyes, all you need to do is find some third party algorithm that can screen genomes to figure out the baby's eye color and plug in your data. Does anything like this exist? I don't think so, but I think it would be trivial for a genetics PhD student to make."
"No, I don't care that you have many studies". This is what it boils down to.
If anyone makes a statement you don't agree with, you're either going to say,
1) You have no proof, or
2) Your proof is a single study; I don't trust single studies claims, and anyway, here is a weakness this study has, or
3) I don't care that you have many studies - parapsychologists *also* have many studies.
Genuinely, I'm not trying to be adversarial, but it's starting to feel like you have *much* different standards of proof for different things.
If you agree with something that has many papers supporting it, you're just going to say something like, "X has been proven over and over and here's links" and not bother to investigate further.
If you disagree with something that has many papers supporting it, you're going to do a deep dive into the papers, discard a majority of them because science nowadays sucks, and then at the end say "You know what, I don't care about how many papers you have anyway".
Does anyone know of data on side effect severity for the covid booster? Especially as it relates to previous side effects? I'd really like scientific data, but happy to hear anecdotes if they're what people have got.
I'm noticing what I'm going to call Lucidity Bias, the evil twin of Occam's Razor. Whereas OR makes sense because it economizes the number of variables. a Lucidity Bias does not economize, it simplifies, for the sake of clarity.
Simplifying is fine if your purpose is communication. It isn't fine if your purpose is understanding.
The Rationalist community has been accused of being full of non-wheel experts trying to invent the wheel, or something like that. I'm not making that accusation, although I'm kinda riffing off of it, because it's at least possible that is a failure mode of those who question all received wisdom.
Non-experts aren't necessarily worse at understanding a field than the experts. For one thing, the experts may suffer from biases present in the field. They may also lack wider knowledge that puts their expert knowledge in perspective. For instance, if experts in a field lack an understanding of mathematical statistics they may come to wrong conclusions because statistics may be essential for analysis of the subject.
It seems to me that rationalists have a thirst for clarity, perhaps too hungry a thirst. Many fields are complicated and aren't easy to explain to the non-expert. After all, many advanced fields require years of study to be merely caught up to the entry level of understanding. In those cases the intellectually hungry among us may prefer an explanation that makes logical sense based on prior known concepts over an explanation that may require more detailed knowledge which uncovers previously unknown concepts, but which requires years of study to uncover and comprehend.
I suggest everyone tweak in the direction of considering that your opinion may change if you had more expert knowledge.
I've started reading Justin E. H. Smith's Hinternet on substack based on a reference to it from here. His latest post https://justinehsmith.substack.com/p/a-surfeit-of-black-bile talks about depression and psychoanalysis - curious to hear what folks here (especially Scott) think about it. Couple of click baity quotes - "Depression makes you a jerk." "psychoanalysis is a pseudoscience".
Triple Interrobang, in relation to your request for comments last OT:
"Why we Cant[sic] Understand the Other Side" oversimplifies the problem. It's more than just outgroup homogeneity on the level of the individual, since a lot of media bakes in "the majority of the Other Side supports X extreme position". Also, a number of the statistics you cite are only weak evidence for your point, due to the potential for differing definitions (obvious examples here are differing definitions of "offensive to minorities" and "God", and differing understanding of what is and is not false information).
Your vid on cancel culture has a few issues:
1) "a movement known as GamerGate engaged in a widespread and extended harassment campaign against those seen as responsible for spreading progressive ideas in the gaming community"
This is more than a little slanted, although I don't entirely fault you for that since this is the line taken by essentially all media coverage*. Firstly, "the movement" didn't harass people; a minority of people in the movement did (that one should probably have occurred to you given your "Other Side" video). Second, the concerns of GamerGate (at least in its final, most wide-scale and impactful phase) were a *little* more nuanced than "spreading progressive ideas in the gaming community"; the objection was largely to progressives attempting to turf "gamers" *out* of games (a couple of the more notable incendiaries: https://dangolding.tumblr.com/post/95985875943/the-end-of-gamershttps://www.gamedeveloper.com/business/-gamers-don-t-have-to-be-your-audience-gamers-are-over- ).
*The thing about GamerGate is that it was basically the watershed rupture between the previously-aligned Grey Tribe (nerdy young mostly-men) and Blue Tribe (the rapidly-nucleating social justice movement). As such there were essentially no media outlets that took the pro-GG position; up until that point "nerdy anti-SJ" wasn't much of a media niche (Scott beat the crowd).
2) You don't really explain why there's the asymmetry in opinions of shame mobs between progressives and conservatives. The obvious explanation is "because there's an asymmetry in the current distribution of such mobs", probably related to things like "progressives use social media more" and "progressives are more politically active". It's noteworthy that the positions on this have basically reversed since the 90s.
3) I think you misattribute fault for some of the effects of "cancellings", in particular people being fired/expelled. A shame mob cannot fire people from their jobs; a shame mob cannot expel someone from a university. Someone with power is taking that action because he or she listened to the shame mob; that's his or her decision and his or her responsibility. I think "stop throwing mobs the heads they demand" is at least as important as "stop forming mobs".
Regarding your more recent "Conspiracy Theorists Aren't Crazy", I think that your title is kind of clickbait and an overstatement. It is a continuum, yes, but the end of that continuum - perfect confirmation bias, in which anybody disagreeing with you is weighted 0 - untethers a person completely from reality. If "crazy" is to have a meaning at all, that almost certainly qualifies. Even the fairly-clickbaity "We Are All Conspiracy Theorists" would be more accurate.
It cites Scott and links to his article: "Recent analysis by Avi Bitterman, a dermatologist in New York, and Scott Alexander, a prominent blogger, suggests that the answer is nuanced."
Nov 27th - print and online editions - The Economist publishes an article on the subject. It does not cite Scott:
Why do people keep saying they have been "humbled" when they win an award or get elected to office? Isn't that the exact opposite of what the word means? Can we stop this? Or is it an ineluctable feature of the mendacity of our times?
Proposed political system- a country uses proportional representation, but also fixes the number of political parties allowed in the constitution- say, somewhere in the range of 5-8. (If I had to pick a number I'd say 6).
This I think solves one of the key problems of PR, which is small fringe parties who earn 1-3% of the vote becoming kingmakers in a coalition. Essentially, the whole political system is held hostage by a very, very small fringe group. More explanation of the problem here (just so I don't write a small essay)- https://en.wikipedia.org/wiki/Proportional_representation#Election_of_minor_parties
This problem is 'solved' in some current PR countries by requiring parties to win a certain % in order to gain representation in parliament. This then introduces the new problem of wasted votes, which can become extreme in some cases- most famously, in 2002 Turkey experienced 46% (!) of its votes being wasted due to a high threshold.
Setting the number of parties allowed in the constitution to my mind fixes this? Fringe parties are impossible- no one's vote is wasted because the 'threshold' rule is not required. We have collectively thousands of years of experience in parliamentary systems, across many different countries on most continents- I think we can make a reasonable judgement as to what the effective limit on the number of parties should be, and at what point beyond which it devolves into ridiculousness & mindless factionalism. I don't see any redeeming societal value to having 10-30+ parties (yes, the latter is possible in some Latin American countries!) Use PR, establish a maximum number of parties, and let candidates & voters sort themselves as they so choose
US military is funding research into a GPS substitute that could work underwater, underground, and at the poles, it is based on detecting the paths of cosmic ray muons. Seems to require multiple detectors intercepting muons but it's not too clear to me how far apart these detectors need to be. Anyway, pretty interesting stuff.
Re China: One of the important things to remember about China is that "China" is almost always an unhelpful unit of analysis. It's so big and diverse that it's more helpful to think about China as a continent rather than a country.
What's happened in China is that several of the coastal cities have upgraded themselves to European levels of prosperity. Then there's still a huge hinterland with very low levels of income.
That's a lot of success! But it's not like China is going to become an overwhelming world dominator very soon.
While I'm talking about it, though, I will say what worries me. In the next 20-30 years, China's economy will become significantly larger than then USA's; and it will become a more active member of the international community. At the same time, it is hemmed in by nuclear-capable US military bases in Japan, South Korea, the Philippines, and the US-supported Taiwan. This is the kind of situation that Kennedy triggered the Cuban missile crisis to avoid. I think that China will find it unacceptable to have US weaponry that close to its capital, and will want to force the USA to move a little further away. The only question is whether that negotiation can happen peacefully, or whether it requires armed confrontation. I'm very much hoping that it can be done peacefully.
"You'll note all the countries in question are still below (sometimes vastly) US per capita."
This seems to miss the very obvious fact that China has more than 4x the population as the US. It doesn't need to be anywhere close to US GDP per capita to vastly exceed it in economic, military, and technological prowess. In fact, China is *already* 15% above the US in terms of GDP PPP, which measures how much goods and services the country produces. The situation is even worse than that, because 30% of China's GDP is from manufacturing, compared to only 10% of the US GDP. In terms of how much physical stuff each country produces, the US is at $2.1 trillion per year, while China is at $7.2 trillion per year, or 3.4x America's industrial output. Make no mistake: China is an industrial behemoth that is heads and shoulders above every other country in history in terms of industrial capability, and when it starts dominating the world, shouting "but your GDP per capita is still small!" isn't going to make them stop.
According to a graph linked to on MR a decade ago, college football fans are the most conservative of sports fans (NBA fans are the most liberal). That comports with my experience. Wealthy conservative alumni are also often big doners to non-Ivy League colleges, often for purposes of helping the football program as opposed to helping the academic side of the school.
While watching college football yesterday and thinking about the University of Austin, I got to thinking: "Why can't all these rich conservative donors use their money to influence their universities in a more conservative direction, if that is such a big concern these days?"
Then it hit me. If football is the most conservative institution on campus and college football fans tend to stay conservatives, then the best way to preserve a conservative culture at a university is to support the football program.
So maybe what the University of Austin should focus on is putting a good football program together. Of course, the trouble with that is they picked the wrong town...
When the Omicron variant was announced, the stock market went down a lot, which makes sense since the stock market is based on companies whose profits would go down if there were more lockdowns.
Crypto also crashed. On some level this makes sense because crypto is usually correlated with the stock market. But realistically people can buy just as much crypto during a lockdown as they can at any other time, so I don't see why knowing the stock market crashed because of COVID should do this.
I don’t think that comparing China to Japan or the USSR (or the Nazis etc) is useful. It’s true that the industrial Revolution is a powerful drug, or rather manufacturing capability is. However when competing against the USSR, Japan or Germany the West kept its manufacturing base. And the size of China matters. It doesn’t have to catch up with the US in per capita GDP - though I think it will; it will be twice the GDP of the US at less than half the GDP per capita. Present trends continuing that’s likely in the near future.
I was wondering about a phenomenon I encountered recently (summer of 2020). I was on an astrology summer camp when a few hornets got into our building. Now, these things are mostly harmless according to Wikipedia: won't sting unless you piss them off, if they do sting it's no worse than a wasp.
While there were a few rumors about the then-current news stoey of Asian murder hornets, those were quickly debunked.
Which is why I was surprised to watch a third of my fellow campers go absolutely *hysterical*. After an hour or so of the hornets being around, during which everyone was a bit on edge but otherwise fine, one 15-year-old started screaming at the top of her lungs.
Very soon, two others followed. We quickly moved everyone outside to try and deal with the chaos. Within minutes, we had a group of ten campers, 12 to 17 years old, mostly girls, shuddering and crying on the ground in a group. The occasional terrified scream emanated from the vicinity (I say occasional, but it was more like every 20 seconds).
As I sat outside at 3 AM in the freezing cold playing card games with the others who remained sane, I only wondered what the neighbors might think. The camp leaders ended up sending everyone home early the next day, unsure what was the trigger and therefore unable to prevent any further events.
My questions: is this a thing that has happened often in the past? Personally, I've never seen or heard of anything like it.
Also, what might be the mechanism? These are normally very reasonable people, but one person's irrational panic rendered them all immobile nuisances for over an hour.
Just from an evolutionary perspective, surely that's a huge problem, right? Imagine if a third of your tribe had a complete mental shutdown the moment a lion approached. Seems like a great strategy for getting eaten.
"There are two drugs which both have stellar Phase III trial results indicating they can reduce the COVID-19 death rate dramatically - one from Merck (Molnupiravir) and the other Pfizer (Paxlovid). We have new weapons to slay the dragon, but we aren’t using them yet!"
Maybe not so stellar for molnupiravir, Merck have issued new results that are somewhat less glowing than the originals:
"Nov 26 (Reuters) - Merck & Co (MRK.N) said on Friday updated data from its study of its experimental COVID-19 pill showed the drug was significantly less effective in cutting hospitalizations and deaths than previously reported.
The drugmaker said its pill showed a 30% reduction in hospitalizations and deaths, based on data from 1,433 patients. In October, its data showed a roughly 50% efficacy, based on data from 775 patients. The drug, molnupiravir, was developed with partner Ridgeback Biotherapeutics.
The lower efficacy of Merck's drug could have big implications in terms of whether countries continue to buy the pill. Interim data from 1,200 participants in Pfizer Inc's (PFE.N) trial for its experimental pill, Paxlovid, showed an 89% reduction in hopsitalizations and deaths."
So down to 30% efficacy from 50% efficacy for molnupiravir, while paxlovid so far is still holding. But would further tests show a similar drop? Maybe, maybe not.
"Merck has said data shows molnupiravir is not capable of inducing genetic changes in human cells, but men enrolled in its trials had to abstain from heterosexual intercourse or agree to use contraception. Women of child-bearing age also had to use birth control.
Still, the FDA said in its briefing document that there are safety concerns about potential birth defects from the drug and asked the panel to discuss whether the drug should be available to pregnant women."
Again, part of the hesitance of the FDA is that they are terrified of another thalidomide, and such fears are not assuaged by the drug company itself telling trialists not to get pregnant.
"Yet, the FDA has decided to keep both drugs banned."
I think that is an unfair characterisation of the decision.
"Blood is on their hands."
Yeah, well. I've considered and discarded responses to that, because mostly they were sarcasm and that's not helpful. I would simply like to say that the "paxlovid is only re-packaged ivermectin, Pfizer are doing this in order to sell expensive drugs instead of cheap ivermectin, follow the scientific evidence to save human pain, suffering and death" are making the same emotive appeals - is Mr. Elton going to boycott Paxlovid on these grounds now?
I don't know, Mr. Elton, how many people did *you* murder this morning? (Sorry, I know I said I'd stay off the sarcasm, but some over-the-top emotive sloganeering just breaks the chains on that for me).
Regarding the Omicron variant, I know we don't currently have a ton of info. I have two questions that I'd be curious to hear people's thoughts on:
1. What are the best threads/posts on the subject from the last 48 hours? All the good threads I've seen are from last Thursday or Friday.
2. Is there any argument *against* getting a booster ASAP? It seems like a reasonable precaution to take in case Omicron is quite bad, but is it possible that this would be counterproductive (e.g. if the mutations make it *very* different from existing variants)?
Inspired by Scott's post on testing the effect of CO2 on cognition using WordTwist, I started doing daily chess puzzles while tracking CO2 and other health metrics. I did an interim analysis on the first 88 days of data and found the following:
- Chess puzzles are a low effort (for me), but high variance and streaky measure of cognitive performance
- Note: I didn't test whether performance on chess puzzles generalizes to other cognitive tasks
- No statistically significant effects were observed, but I saw modest effect sizes and p-values for:
- CO2 Levels >600 ppm: R2 = 0.14, p = 0.067
- Coefficient of Variation in blood glucose: R2 = 0.079, p = 0.16
- My current sample size is underpowered to detect the effects I'm looking for. I likely need 3-4x as much data to reliably detect the effect sizes I'm looking for.
- Given how many correlations I looked at, the lack of pre-registration of analyses, and the small number of data points, these effects are likely due to chance/noise in the data, but they're suggestive enough for me to continue tracking.
Questions:
- My variation in rating has long stretches of better or worse than average performance that seem unlikely to be due to chance. Does anyone know of a way to test if this is the case?
- Any statisticians interested in taking a deeper/more rigorous look at my data or have advice on how I should do so?
- Any suggestions on other quick cognitive assessments that would be less noisy?
Following the discussion in the last post on Helion, a trying-to-get-fusion startup, I've been thinking about how scientific communities develop norms orthogonal to their science.
Fusion has strong collaboration norms. This is most obvious for ITER, which is a collaboration between countries representing more than half the world's population. It is also usually true for private companies. The high temperature superconducting tape designed by Commonwealth Fusion Systems for their tokamak is also being used by Type One Energy to build a stellarator. By being secretive, Helion is betraying the (non-scientific) norms of the community.
I'm thinking that is a kind of founder effect, but I would be interested in hearing other people's thoughts on it.
There are several important examples from the early days of plasma physics, both positive and negative, that helped to establish this norm.
The biggest negative example is the Huemul Project by Ronald Richter. Richter is the origin of the mad-scientist-with-a-secret-base-in-Patagonia trope. In 1951, he and Peron declared that he had GOT FUSION, which caused a lot of excitement and skepticism, until he was arrested for fraud in 1955.
The biggest positive example is from 1968, when a group of Soviet scientists traveled to England to present the results of the T-3 tokamak, which was an order of magnitude hotter than any other experiment. Western scientists were skeptical, so they invited 5 British scientists to come to Russia to do the experiments themselves. The "Culham Five" confirmed the results and published them in Nature the next year. From then on, there was significant collaboration across the Iron Curtain.
Despite being indoctrinated in this field (a la Kuhn), I can see some costs to this attitude. Because we refused to participate in the Cold War, we never had our version of a Space Race. If we ever wanted to get seriously funded, maybe we should have made it something that we can WIN against Russia or China.
Whether or not China IS successful is not a policy question. I think that China's growth so far does provide additional reason to remove obstacles to US growth (NIMBY resistance to investment, restrictions on urban residential and commercial development, structural fiscal deficits, obstacles to freer international trade, active recement of immigrants). But what else?
Question with regard to the intersexual dynamics and relationships:
What are the most interesting papers, findings supported with broad anecdotal evidence, and/or your personal predictions that are very surprising/important and absent in the public discourse/understanding of the topic?
Two new books on AI: Age of AI by Eric Schmidt and Henry Kissinger has gotten the most press, but I found it impossible to finish. Boring and repetitive. Did anyone find it valuable?
Much better is Rule of the Robots: how artificial Intelligence will transform everything by Martin Ford. Lots of good stuff on the path toward AGI, based on interviews with people like Demis Hassabis and Geoff Hinton. Also has the best coverage I've seen on the AI race with China. Maybe these two would be good candidates for reviews here.
Happy to help if you need more folks to help judge ACX applications, develop a rubric etc - I’m off of work for the rest of the calendar year so have plenty of time
Has anybody done a study on the effect of "placebo" in treating COVID? Formally, if you give one group an "experimental" pill (which is just a sugar pill), and the other group the same treatment with no pill, which group does better?
Hopefully you could measure this with a meta-analysis comparing placebo-treated people in published experiments with population-level statistics, controlled for age and calendar date [because of variant shifts].
Writing a series on Brain-Computer Interfaces. The last post talks about emerging technologies - some of you might find it interesting. https://howthehell.substack.com/p/bci-part-2
Thanks for signal boosting my post Scott! It took a while to write!
Everyone who cares about the Paxlovid issue should consider submitting a comment to the FDA. It doesn't have to be very long. Here's the direct link where you can submit your comment to be looked at by the committee that will review Merck's EUA application for Molnupiravir:
Note the deadline for your comment to be considered is tomorrow (November 29th) (the meeting is on the 30th).
The comment should probably address something related to Molnupiravir although strictly speaking it doesn't have to. Either way, you should definitely urge the FDA to do the EUA for Paxlovid ASAP!!
"The FDA still hasn't scheduled a meeting to review Paxlovid. It scheduled its meeting for Molnupiravir 7 weeks out. But it scheduled its review of the vaccines just 3 weeks after the submission. One suggestion, if you submit a comment, is to urge the FDA to schedule its review ASAP and to make the timeline 3 weeks (or less) rather than 7."
Unfortunately it's already been announced/happened, but if there is an ACX Grant Program 2 I think public applications with a prediction market would have been interesting
@scott my browser just killed 10-20% of my smartphones battery while having this thread open. Then it got hot and crashed. Now my browser shows black and white bars. Please look for another, better way for your readers to have seamless discussions. There is plenty of software and services way better then the current substack crap.
The discord link on the front page isn't working for me? Is this happening to anyone else? Does someone have a different one?
Can someone give me an explainer on the Chinese social credit system(s)? Where on the line of "this is a widespread totalitarian thought-policing system" vs "this is just a 21st-century better business bureau and worries about it are just a red scare." What are the good sources to read about this that aren't super partisan?
Is there a way to search comments here that's up-to-date? acxsearch.herokuapp.com is over five months out of date now.
So what will be the next political craze? Identity politics has been in for the past 5 years but seems to be on the wane. Maybe war with Russia will take its place? Or maybe another 3 years of fighting over vaccines and masks? What else seems likely? As a long shot bet, I'll buy some abortion-war tickets.
Has the present ever felt so absurd -- not a certain era in retrospect, but as it is happening -- or does it always feel this absurd to people as they start to age?
I can't "Expand full comment" as of a few hours ago.
The latest episode of Always Sunny in Philadelphia (which just ran on FXX) is one of the best anti-woke takes from liberal Hollywood. If you think the left can't be anti-woke, I suggest you watch it.
In it, they attack viciously how some of their earlier episodes were erased by woke Netflix because they used black-face, they lean in hard on other racial, classist (white yokels) and other sexual stereotypes acceptable to contemporary moral sensibilities, to point a pointy stick in the eye of hypocritical leftists, and actually make it all very funny.
Wokism is dying and Hollywood is taking the lead in killing it.
And what would you say is the international relationship between the banning of homosexuality and the banning of other human pursuits?
I would be very grateful if anyone could help me find an article that I'm 80% sure Scott linked to a few years back: it was about a community mental health worker in India and the general effectiveness of a community-based social work-like approach to mental health, where people with relatively short training can help a lot of others, as opposed to the way that licensed therapists need many years of training and a lot of people can't afford their help. It also mentioned some research done in the 60s and 70s, that showed how these two approaches had comparable results and suggested that the research was kept quiet and then stopped because of lobbying from the licensed professionals.
What's the conclusion on face masks after all the new evidence? This site seems pretty shady but links to several prestigious seeming studies claiming that there is no benefit - https://swprs.org/face-masks-evidence/
On the other hand there are several sites that do claim effectiveness. Scott's old post took source control almost for granted, but that seems to be up for debate now as well.
Is there any use case for a drone autogyro?
So this (what follows) seems like the perfect topic for the rationalist community.
There is an old well established theory (hypothesis) to explain part of the 'world'.
Almost all the experts believe the old theory, and though they have been searching for years,
decades, there is no hard data to support the theory. (No Dark matter has been found... and the
possible un-searched space is small.) https://en.wikipedia.org/wiki/Dark_matter#Detection_of_dark_matter_particles
There is a newer fringe model, which at some level is just an empirical fit to the data.
And for anyone who knows physics, the fringe idea is crazy. When you integrate the potential
energy, you get a logarithmic divergence. So the model is clearly incomplete. And yet the
fit to the data is awesome! https://tritonstation.com/2021/06/28/the-rar-extended-by-weak-lensing/ * So (it's possible) there is some range of 'gravity' for which this MoND
'fit' is right... and though it may look like a crazy idea, it's also crazy to ignore it.
(there is more to MoND than galaxy rotation curves. It also predicts external field effects,
which also look to fit the data. http://astroweb.case.edu/ssm/mond/EFE.html
I've been wanting to post this here, a while (open-thread 200)
How much less credible is a doctor's weight loss advice if the doctor himself is still obese? I think it would only be the ad hominem fallacy if one put too much weight (pardon the pun) on that one data point, or applied it where it wasn't applicable.
I've noticed that in the U.S., the GDP deflator and real GDP have been growing at the same pace for a surprisingly long time. This has not been the case in Italy (where the GDP deflator has grown much faster than real GDP) or Japan (where the reverse has been the case). Does anyone have any thoughts why increases in nominal incomes have been equally reflected in real GDP and prices in America for so long?
https://twitter.com/Enopoletus/status/1465723015740760068
I take an SSRI and also Adderall. The SSRI prevent reuptake of serotonin, the Adderall prevents the reuptake of dopamine and norepinephrine (if I understand correctly). Could this mean I have low qualities and/or too high uptake of neurotransmitters generally?
Anyone have any recommendations on reading guides/companions for Nietzsche?
I've been reading the dialogues between Eliezer Yudkowsky & MIRI and various members of LessWrong on LessWrong, and my current model of Eliezer's belief is roughly "on current trajectory there is over 90% probability that some form of advanced artificial intelligence will kill all humans and it most likely will happen during this century" (if you think my model of Eliezer's belief is wrong, please tell me).
https://www.lesswrong.com/posts/CpvyhFy9WvCNsifkY/discussion-with-eliezer-yudkowsky-on-agi-interventions
https://www.lesswrong.com/posts/7im8at9PmhbT4JHsW/ngo-and-yudkowsky-on-alignment-difficulty
https://www.lesswrong.com/posts/hwxj4gieR7FWNwYfa/ngo-and-yudkowsky-on-ai-capability-gains-1
https://www.lesswrong.com/posts/vwLxd6hhFvPbvKmBH/yudkowsky-and-christiano-discuss-takeoff-speeds
Since most humans during the history of the world have died, I've made my peace with the fact that I too might die. I think it's most likely that everyone will go to nothingness after they die though there's a nonzero chance that I'm wrong about this, but I think everything we know about neuroscience, cognitive science and physics points towards nonsurvival after death.
I think death is negative in the sense that you are deprived of all the positive possibilities of life, but once you are dead, the condition itself is neutral since there are neither positive or negative things in nothingness, there's just nothing, like dreamless sleep.
What I'm concerned about Yudkowsky's scenario is whether there is going to be suffering, pain and other bad things leading up to it. Do you think there will be such things?
How does one participate in politics without an ideology? I have plenty of strong opinions about policy but no ideology. I voted against Trump because that seemed like a no-brainer, but normally I don't vote because both options seem equally bad. Third parties seem worse.
Some people get really involved with specific issues, but I don't care about specific issues all that much. I just want the center to hold in US politics. I mostly fear the country becoming too radical on either the right or the left, but I consider the danger of radicalism from either side to be roughly equal. Or rather, I consider the greatest danger to democracy in the USA to be radicalism itself. I suspect that the more radical the left becomes, the more radical the right will become and vice versa.
I suppose I can try to play the role of the median voter who votes against whichever side I think is going too far. It's not a clever role to play, but maybe it's the best.
What are some battles where ultra well-built castles were overrun by enemies anyway, and through brute force rather than some clever weapon or tactic that negated the castle's defenses?
I want to know about the most spectacular, high-casualty battles where great castles fell.
We talked a bit about the multiple meanings of the word‘literal’. Here is another one that has multiple meanings.
The verb 'bemuse' and the corresponding participial adjective 'bemused'.
When I first leaned the meanings bemuse meant to make utterly confused or muddle, as with an intoxicating liquor.
Now bemused is often used as adjective meaning something similar to amused.
Okay, fine, I get it. Word meanings change with time. The trouble is that it is not always clear which way the author intended to use it. Context sometimes makes it clear but not alway.
How do the rest of you use it and interpret it?
Not clear why it matters what Alexandros thinks about the meta-analysis here.
Let's take someone who might actually have heard of "meta-analysis" before yesterday: Andrew Hill. His original meta-analysis suggested that ivermectin worked, but later when he and his colleagues excluded the fraudulent and problematic studies, he admitted that the effect (if any) is tiny. https://www.medpagetoday.com/special-reports/exclusives/95333
Are there any good studies, analyses, or news articles that show what the cost-performance trend is for virtual reality goggles over time? In short, at what rate are their prices declining each year?
I'd imagine one metric would be shown on a graph whose Y-axis was "Real cost per display pixel" and whose X-axis was "Year".
Re: the parapsychologist analogy. I wonder if you would hold Pfizer to the same standard?
We know that not too long ago they made false claims and had to pay large fines: https://abcnews.go.com/Business/pfizer-fined-23-billion-illegal-marketing-off-label/story?id=8477617
Even our government "experts" seem to continually make important mistakes.
I think skepticism is warranted no matter who the expert is.
I don't trust any of them. I trust those with large financial incentives even less.
Every time we have a non-political Open Thread, I think about a political topic or two, and regret being unable to post them. Every time we have a political Open Thread, I forget what I wanted to write about. (So today I decided to write a meta comment instead.)
Did alexandros ever justify his decision to summate multiple outcomes in his meta analyses? I know several people brought this up but I dunno if he answered.
Question: Does Scott have facial hair? His writing reads like a clean-shaven fellow's, with nary a hint of beard, moustache or mutton-chop.
Is my description accurate?
I think it's strange that there have been all these great families brought up but I haven't seen any famous twins yet. The twin rate is about 2% so it's starting to get odd that there hasn't been one example yet.
On the great families subject, I wonder if the Plinys Elder & Younger would qualify? I guess people from ancient Rome are less relevant given the lower population and stronger class effects (eg, most of the population was slaves & plebians who would have had no opportunity to become famous historians)
So within the last couple years we've had two fairly prominent instances of major websites (Tumblr a while ago and Onlyfans more recently) having to scrub themselves (to varying degrees of success) of all pornographic content. From what I can tell, the banks that control those site's loans were breathing down their neck or something like that, basically threatening to back out unless changes were made. My question is, why isn't there a bank that's willing to allow that kind of content on their investments? It seems like an easy way to make a lot of money, since porn is a (quick Google search) $12 billion industry, which seems like a lot of money to be left on the ground. A bank that branded itself as open to those kinds of investments and precommited to never change that policy should have all the major players flocking to it, just for the stability and peace of mind if nothing else.
The Philadelphia ACX Meetup will be hosting a Solstice Celebration on Tuesday, December 21 in Ardmore, PA. You can request to join our Google Group here for full details: https://groups.google.com/g/ACXPhiladelphia
I am getting a new lawnmower. We are considering gas-versus-electric right now. I am willing to take a bit of a hit on price on convenience for environmental concerns, to a limit.
It seems the best reason to go electric is that a lawnmower doesn't have a catalytic converter. My lawnmower might be spewing out all kinds of harmful stuff, so in theory electrifying my mower before my car is a good idea.
But how can I quantify this? And how bad is it? I probably only burn 5 gallons a year in the mower. What exactly am I spitting into the air from a normal mower at that rate, and how much would I expect to pay to remove that stuff otherwise?
Can anyone recommend a good password manager? I use an iPhone, a Kindle tablet, an iMac, a Windows PC and a Linux box.
Anyone know of any good rationalist communities in London? I'm looking to meet new people into Scott, Tyler Cowen, etc.
Are there any American Expats in Taiwan? I've been considering making a move out there to live and work and was interested in getting an American perspective.
I'd want to move there to take a position as an Engineering project manager, and live there permanently
Some time back, Scott wrote about polygenic screening and the first PGT baby. In the piece, he mentions,
"Screening companies will give you the raw data if you ask for it, so if you want to screen for an embryo with green eyes, all you need to do is find some third party algorithm that can screen genomes to figure out the baby's eye color and plug in your data. Does anything like this exist? I don't think so, but I think it would be trivial for a genetics PhD student to make."
Well, it's here. The Fertility Institute's Eye Color Selection Program: https://www.fertility-docs.com/programs-and-services/pgd-screening/choose-your-babys-eye-color.php
"Beware the man of a single study".
But also,
"No, I don't care that you have many studies". This is what it boils down to.
If anyone makes a statement you don't agree with, you're either going to say,
1) You have no proof, or
2) Your proof is a single study; I don't trust single studies claims, and anyway, here is a weakness this study has, or
3) I don't care that you have many studies - parapsychologists *also* have many studies.
Genuinely, I'm not trying to be adversarial, but it's starting to feel like you have *much* different standards of proof for different things.
If you agree with something that has many papers supporting it, you're just going to say something like, "X has been proven over and over and here's links" and not bother to investigate further.
If you disagree with something that has many papers supporting it, you're going to do a deep dive into the papers, discard a majority of them because science nowadays sucks, and then at the end say "You know what, I don't care about how many papers you have anyway".
Does anyone know of data on side effect severity for the covid booster? Especially as it relates to previous side effects? I'd really like scientific data, but happy to hear anecdotes if they're what people have got.
I'm noticing what I'm going to call Lucidity Bias, the evil twin of Occam's Razor. Whereas OR makes sense because it economizes the number of variables. a Lucidity Bias does not economize, it simplifies, for the sake of clarity.
Simplifying is fine if your purpose is communication. It isn't fine if your purpose is understanding.
The Rationalist community has been accused of being full of non-wheel experts trying to invent the wheel, or something like that. I'm not making that accusation, although I'm kinda riffing off of it, because it's at least possible that is a failure mode of those who question all received wisdom.
Non-experts aren't necessarily worse at understanding a field than the experts. For one thing, the experts may suffer from biases present in the field. They may also lack wider knowledge that puts their expert knowledge in perspective. For instance, if experts in a field lack an understanding of mathematical statistics they may come to wrong conclusions because statistics may be essential for analysis of the subject.
It seems to me that rationalists have a thirst for clarity, perhaps too hungry a thirst. Many fields are complicated and aren't easy to explain to the non-expert. After all, many advanced fields require years of study to be merely caught up to the entry level of understanding. In those cases the intellectually hungry among us may prefer an explanation that makes logical sense based on prior known concepts over an explanation that may require more detailed knowledge which uncovers previously unknown concepts, but which requires years of study to uncover and comprehend.
I suggest everyone tweak in the direction of considering that your opinion may change if you had more expert knowledge.
It's been four months; are there any results from the ACX surveys?
I've started reading Justin E. H. Smith's Hinternet on substack based on a reference to it from here. His latest post https://justinehsmith.substack.com/p/a-surfeit-of-black-bile talks about depression and psychoanalysis - curious to hear what folks here (especially Scott) think about it. Couple of click baity quotes - "Depression makes you a jerk." "psychoanalysis is a pseudoscience".
Triple Interrobang, in relation to your request for comments last OT:
"Why we Cant[sic] Understand the Other Side" oversimplifies the problem. It's more than just outgroup homogeneity on the level of the individual, since a lot of media bakes in "the majority of the Other Side supports X extreme position". Also, a number of the statistics you cite are only weak evidence for your point, due to the potential for differing definitions (obvious examples here are differing definitions of "offensive to minorities" and "God", and differing understanding of what is and is not false information).
Your vid on cancel culture has a few issues:
1) "a movement known as GamerGate engaged in a widespread and extended harassment campaign against those seen as responsible for spreading progressive ideas in the gaming community"
This is more than a little slanted, although I don't entirely fault you for that since this is the line taken by essentially all media coverage*. Firstly, "the movement" didn't harass people; a minority of people in the movement did (that one should probably have occurred to you given your "Other Side" video). Second, the concerns of GamerGate (at least in its final, most wide-scale and impactful phase) were a *little* more nuanced than "spreading progressive ideas in the gaming community"; the objection was largely to progressives attempting to turf "gamers" *out* of games (a couple of the more notable incendiaries: https://dangolding.tumblr.com/post/95985875943/the-end-of-gamers https://www.gamedeveloper.com/business/-gamers-don-t-have-to-be-your-audience-gamers-are-over- ).
*The thing about GamerGate is that it was basically the watershed rupture between the previously-aligned Grey Tribe (nerdy young mostly-men) and Blue Tribe (the rapidly-nucleating social justice movement). As such there were essentially no media outlets that took the pro-GG position; up until that point "nerdy anti-SJ" wasn't much of a media niche (Scott beat the crowd).
2) You don't really explain why there's the asymmetry in opinions of shame mobs between progressives and conservatives. The obvious explanation is "because there's an asymmetry in the current distribution of such mobs", probably related to things like "progressives use social media more" and "progressives are more politically active". It's noteworthy that the positions on this have basically reversed since the 90s.
3) I think you misattribute fault for some of the effects of "cancellings", in particular people being fired/expelled. A shame mob cannot fire people from their jobs; a shame mob cannot expel someone from a university. Someone with power is taking that action because he or she listened to the shame mob; that's his or her decision and his or her responsibility. I think "stop throwing mobs the heads they demand" is at least as important as "stop forming mobs".
Regarding your more recent "Conspiracy Theorists Aren't Crazy", I think that your title is kind of clickbait and an overstatement. It is a continuum, yes, but the end of that continuum - perfect confirmation bias, in which anybody disagreeing with you is weighted 0 - untethers a person completely from reality. If "crazy" is to have a meaning at all, that almost certainly qualifies. Even the fairly-clickbaity "We Are All Conspiracy Theorists" would be more accurate.
Congrats to Scott for being cited in The Economist! Timeline:
Nov 18th - Scott publishes his epic Ivermectin post.
Nov 19th - online edition only - The Economist publishes a detailed graphic on the "Ivermectin only helps COVID-19 patients with worms" hypothesis:
https://www.economist.com/graphic-detail/2021/11/18/ivermectin-may-help-covid-19-patients-but-only-those-with-worms
It cites Scott and links to his article: "Recent analysis by Avi Bitterman, a dermatologist in New York, and Scott Alexander, a prominent blogger, suggests that the answer is nuanced."
Nov 27th - print and online editions - The Economist publishes an article on the subject. It does not cite Scott:
https://www.economist.com/united-states/2021/11/27/ivermectin-may-help-covid-19-patients-but-only-those-with-worms
Why do people keep saying they have been "humbled" when they win an award or get elected to office? Isn't that the exact opposite of what the word means? Can we stop this? Or is it an ineluctable feature of the mendacity of our times?
More "great family" trivia: Judith Love Cohen (https://en.wikipedia.org/wiki/Judith_Love_Cohen), who did a bunch of work on the Apollo program, was the mother of Neil Siegel (https://en.wikipedia.org/wiki/Neil_Siegel), who did a lot of work on navigation systems (for the military but this work has since found its way into consumer devices)... and she was also the mother of Jack Black. (Including link for consistency, https://en.wikipedia.org/wiki/Jack_Black :P )
(No idea if other members of this family are particularly noteworthy.)
Proposed political system- a country uses proportional representation, but also fixes the number of political parties allowed in the constitution- say, somewhere in the range of 5-8. (If I had to pick a number I'd say 6).
This I think solves one of the key problems of PR, which is small fringe parties who earn 1-3% of the vote becoming kingmakers in a coalition. Essentially, the whole political system is held hostage by a very, very small fringe group. More explanation of the problem here (just so I don't write a small essay)- https://en.wikipedia.org/wiki/Proportional_representation#Election_of_minor_parties
This problem is 'solved' in some current PR countries by requiring parties to win a certain % in order to gain representation in parliament. This then introduces the new problem of wasted votes, which can become extreme in some cases- most famously, in 2002 Turkey experienced 46% (!) of its votes being wasted due to a high threshold.
Setting the number of parties allowed in the constitution to my mind fixes this? Fringe parties are impossible- no one's vote is wasted because the 'threshold' rule is not required. We have collectively thousands of years of experience in parliamentary systems, across many different countries on most continents- I think we can make a reasonable judgement as to what the effective limit on the number of parties should be, and at what point beyond which it devolves into ridiculousness & mindless factionalism. I don't see any redeeming societal value to having 10-30+ parties (yes, the latter is possible in some Latin American countries!) Use PR, establish a maximum number of parties, and let candidates & voters sort themselves as they so choose
US military is funding research into a GPS substitute that could work underwater, underground, and at the poles, it is based on detecting the paths of cosmic ray muons. Seems to require multiple detectors intercepting muons but it's not too clear to me how far apart these detectors need to be. Anyway, pretty interesting stuff.
Popular article: https://www.shephardmedia.com/news/defence-notes/muon-based-research-could-facilitate-arctic-operat/
A related paper: https://www.nature.com/articles/s41598-020-75843-7
Re China: One of the important things to remember about China is that "China" is almost always an unhelpful unit of analysis. It's so big and diverse that it's more helpful to think about China as a continent rather than a country.
What's happened in China is that several of the coastal cities have upgraded themselves to European levels of prosperity. Then there's still a huge hinterland with very low levels of income.
That's a lot of success! But it's not like China is going to become an overwhelming world dominator very soon.
While I'm talking about it, though, I will say what worries me. In the next 20-30 years, China's economy will become significantly larger than then USA's; and it will become a more active member of the international community. At the same time, it is hemmed in by nuclear-capable US military bases in Japan, South Korea, the Philippines, and the US-supported Taiwan. This is the kind of situation that Kennedy triggered the Cuban missile crisis to avoid. I think that China will find it unacceptable to have US weaponry that close to its capital, and will want to force the USA to move a little further away. The only question is whether that negotiation can happen peacefully, or whether it requires armed confrontation. I'm very much hoping that it can be done peacefully.
"You'll note all the countries in question are still below (sometimes vastly) US per capita."
This seems to miss the very obvious fact that China has more than 4x the population as the US. It doesn't need to be anywhere close to US GDP per capita to vastly exceed it in economic, military, and technological prowess. In fact, China is *already* 15% above the US in terms of GDP PPP, which measures how much goods and services the country produces. The situation is even worse than that, because 30% of China's GDP is from manufacturing, compared to only 10% of the US GDP. In terms of how much physical stuff each country produces, the US is at $2.1 trillion per year, while China is at $7.2 trillion per year, or 3.4x America's industrial output. Make no mistake: China is an industrial behemoth that is heads and shoulders above every other country in history in terms of industrial capability, and when it starts dominating the world, shouting "but your GDP per capita is still small!" isn't going to make them stop.
Okay, a cranky pedant question. What is the function of the word ‘literally’ in a sentence like
“I literally never saw him before.”
It could just mean
“I never saw him before.”
But it seems to mean something like
“I never saw him before. Why would you even suggest that I had?”
Then again, it could just be a verbal tic that is making the rounds.
According to a graph linked to on MR a decade ago, college football fans are the most conservative of sports fans (NBA fans are the most liberal). That comports with my experience. Wealthy conservative alumni are also often big doners to non-Ivy League colleges, often for purposes of helping the football program as opposed to helping the academic side of the school.
While watching college football yesterday and thinking about the University of Austin, I got to thinking: "Why can't all these rich conservative donors use their money to influence their universities in a more conservative direction, if that is such a big concern these days?"
Then it hit me. If football is the most conservative institution on campus and college football fans tend to stay conservatives, then the best way to preserve a conservative culture at a university is to support the football program.
So maybe what the University of Austin should focus on is putting a good football program together. Of course, the trouble with that is they picked the wrong town...
When the Omicron variant was announced, the stock market went down a lot, which makes sense since the stock market is based on companies whose profits would go down if there were more lockdowns.
Crypto also crashed. On some level this makes sense because crypto is usually correlated with the stock market. But realistically people can buy just as much crypto during a lockdown as they can at any other time, so I don't see why knowing the stock market crashed because of COVID should do this.
Does anyone have a good explanation for this?
I don’t think that comparing China to Japan or the USSR (or the Nazis etc) is useful. It’s true that the industrial Revolution is a powerful drug, or rather manufacturing capability is. However when competing against the USSR, Japan or Germany the West kept its manufacturing base. And the size of China matters. It doesn’t have to catch up with the US in per capita GDP - though I think it will; it will be twice the GDP of the US at less than half the GDP per capita. Present trends continuing that’s likely in the near future.
I was wondering about a phenomenon I encountered recently (summer of 2020). I was on an astrology summer camp when a few hornets got into our building. Now, these things are mostly harmless according to Wikipedia: won't sting unless you piss them off, if they do sting it's no worse than a wasp.
While there were a few rumors about the then-current news stoey of Asian murder hornets, those were quickly debunked.
Which is why I was surprised to watch a third of my fellow campers go absolutely *hysterical*. After an hour or so of the hornets being around, during which everyone was a bit on edge but otherwise fine, one 15-year-old started screaming at the top of her lungs.
Very soon, two others followed. We quickly moved everyone outside to try and deal with the chaos. Within minutes, we had a group of ten campers, 12 to 17 years old, mostly girls, shuddering and crying on the ground in a group. The occasional terrified scream emanated from the vicinity (I say occasional, but it was more like every 20 seconds).
As I sat outside at 3 AM in the freezing cold playing card games with the others who remained sane, I only wondered what the neighbors might think. The camp leaders ended up sending everyone home early the next day, unsure what was the trigger and therefore unable to prevent any further events.
My questions: is this a thing that has happened often in the past? Personally, I've never seen or heard of anything like it.
Also, what might be the mechanism? These are normally very reasonable people, but one person's irrational panic rendered them all immobile nuisances for over an hour.
Just from an evolutionary perspective, surely that's a huge problem, right? Imagine if a third of your tribe had a complete mental shutdown the moment a lion approached. Seems like a great strategy for getting eaten.
The second edition of A student's introduction to English grammar by Huddleston, Pullum, and Reynolds is now out.
"There are two drugs which both have stellar Phase III trial results indicating they can reduce the COVID-19 death rate dramatically - one from Merck (Molnupiravir) and the other Pfizer (Paxlovid). We have new weapons to slay the dragon, but we aren’t using them yet!"
Maybe not so stellar for molnupiravir, Merck have issued new results that are somewhat less glowing than the originals:
https://www.reuters.com/business/healthcare-pharmaceuticals/merck-says-covid-19-pill-cuts-hospitalization-death-risk-by-30-2021-11-26/
"Nov 26 (Reuters) - Merck & Co (MRK.N) said on Friday updated data from its study of its experimental COVID-19 pill showed the drug was significantly less effective in cutting hospitalizations and deaths than previously reported.
The drugmaker said its pill showed a 30% reduction in hospitalizations and deaths, based on data from 1,433 patients. In October, its data showed a roughly 50% efficacy, based on data from 775 patients. The drug, molnupiravir, was developed with partner Ridgeback Biotherapeutics.
The lower efficacy of Merck's drug could have big implications in terms of whether countries continue to buy the pill. Interim data from 1,200 participants in Pfizer Inc's (PFE.N) trial for its experimental pill, Paxlovid, showed an 89% reduction in hopsitalizations and deaths."
So down to 30% efficacy from 50% efficacy for molnupiravir, while paxlovid so far is still holding. But would further tests show a similar drop? Maybe, maybe not.
"Merck has said data shows molnupiravir is not capable of inducing genetic changes in human cells, but men enrolled in its trials had to abstain from heterosexual intercourse or agree to use contraception. Women of child-bearing age also had to use birth control.
Still, the FDA said in its briefing document that there are safety concerns about potential birth defects from the drug and asked the panel to discuss whether the drug should be available to pregnant women."
Again, part of the hesitance of the FDA is that they are terrified of another thalidomide, and such fears are not assuaged by the drug company itself telling trialists not to get pregnant.
"Yet, the FDA has decided to keep both drugs banned."
I think that is an unfair characterisation of the decision.
"Blood is on their hands."
Yeah, well. I've considered and discarded responses to that, because mostly they were sarcasm and that's not helpful. I would simply like to say that the "paxlovid is only re-packaged ivermectin, Pfizer are doing this in order to sell expensive drugs instead of cheap ivermectin, follow the scientific evidence to save human pain, suffering and death" are making the same emotive appeals - is Mr. Elton going to boycott Paxlovid on these grounds now?
https://www.youtube.com/watch?v=ufy2AweXRkc
"How many people is the FDA murdering?"
I don't know, Mr. Elton, how many people did *you* murder this morning? (Sorry, I know I said I'd stay off the sarcasm, but some over-the-top emotive sloganeering just breaks the chains on that for me).
Hopefully nominative determinism won't check out with Dr. Bitterman.
The French Church endorses effective altruism !
So this year the French traduction of the catholic missal (= the book with the text for the mass) changed.
The old translation of the solemn benediction for the Advent read
"May he make your faith firm, your hope joyful and your 𝗰𝗵𝗮𝗿𝗶𝘁𝘆 𝗰𝗼𝗻𝘀𝘁𝗮𝗻𝘁"
The new translation read
"May he make your faith firm, your hope joyful, your 𝗰𝗵𝗮𝗿𝗶𝘁𝘆 𝗲𝗳𝗳𝗲𝗰𝘁𝗶𝘃𝗲".
From constant charity to effective charity, isn't this the whole point of EA ?
Regarding the Omicron variant, I know we don't currently have a ton of info. I have two questions that I'd be curious to hear people's thoughts on:
1. What are the best threads/posts on the subject from the last 48 hours? All the good threads I've seen are from last Thursday or Friday.
2. Is there any argument *against* getting a booster ASAP? It seems like a reasonable precaution to take in case Omicron is quite bad, but is it possible that this would be counterproductive (e.g. if the mutations make it *very* different from existing variants)?
Inspired by Scott's post on testing the effect of CO2 on cognition using WordTwist, I started doing daily chess puzzles while tracking CO2 and other health metrics. I did an interim analysis on the first 88 days of data and found the following:
- Chess puzzles are a low effort (for me), but high variance and streaky measure of cognitive performance
- Note: I didn't test whether performance on chess puzzles generalizes to other cognitive tasks
- No statistically significant effects were observed, but I saw modest effect sizes and p-values for:
- CO2 Levels >600 ppm: R2 = 0.14, p = 0.067
- Coefficient of Variation in blood glucose: R2 = 0.079, p = 0.16
- My current sample size is underpowered to detect the effects I'm looking for. I likely need 3-4x as much data to reliably detect the effect sizes I'm looking for.
- Given how many correlations I looked at, the lack of pre-registration of analyses, and the small number of data points, these effects are likely due to chance/noise in the data, but they're suggestive enough for me to continue tracking.
Questions:
- My variation in rating has long stretches of better or worse than average performance that seem unlikely to be due to chance. Does anyone know of a way to test if this is the case?
- Any statisticians interested in taking a deeper/more rigorous look at my data or have advice on how I should do so?
- Any suggestions on other quick cognitive assessments that would be less noisy?
If you're interested in the detailed write-up, you can find it at: https://www.quantifieddiabetes.com/2021/11/using-chess-puzzles-to-assess-cognition-exploratory-analysis-of-co2-and-other-mediators-shows-suggestive-but-not-conclusive-effects.html
Original post by Scott: https://astralcodexten.substack.com/p/eight-hundred-slightly-poisoned-word
Following the discussion in the last post on Helion, a trying-to-get-fusion startup, I've been thinking about how scientific communities develop norms orthogonal to their science.
Fusion has strong collaboration norms. This is most obvious for ITER, which is a collaboration between countries representing more than half the world's population. It is also usually true for private companies. The high temperature superconducting tape designed by Commonwealth Fusion Systems for their tokamak is also being used by Type One Energy to build a stellarator. By being secretive, Helion is betraying the (non-scientific) norms of the community.
I'm thinking that is a kind of founder effect, but I would be interested in hearing other people's thoughts on it.
There are several important examples from the early days of plasma physics, both positive and negative, that helped to establish this norm.
The biggest negative example is the Huemul Project by Ronald Richter. Richter is the origin of the mad-scientist-with-a-secret-base-in-Patagonia trope. In 1951, he and Peron declared that he had GOT FUSION, which caused a lot of excitement and skepticism, until he was arrested for fraud in 1955.
The biggest positive example is from 1968, when a group of Soviet scientists traveled to England to present the results of the T-3 tokamak, which was an order of magnitude hotter than any other experiment. Western scientists were skeptical, so they invited 5 British scientists to come to Russia to do the experiments themselves. The "Culham Five" confirmed the results and published them in Nature the next year. From then on, there was significant collaboration across the Iron Curtain.
Despite being indoctrinated in this field (a la Kuhn), I can see some costs to this attitude. Because we refused to participate in the Cold War, we never had our version of a Space Race. If we ever wanted to get seriously funded, maybe we should have made it something that we can WIN against Russia or China.
I am considering using https://newsfounded.com/polandeng/most-of-those-convicted-of-rape-in-sweden-are-immigrants-but-the-problem-is-the-scientists-who-write-about-it/ as an example of something in an upcoming post. Does anyone have any evidence that it isn't true, or is exaggerated, or any other reason I shouldn't?
Whether or not China IS successful is not a policy question. I think that China's growth so far does provide additional reason to remove obstacles to US growth (NIMBY resistance to investment, restrictions on urban residential and commercial development, structural fiscal deficits, obstacles to freer international trade, active recement of immigrants). But what else?
How much should I care about Omnicron?
> Steve Jobs is the biological sister (adopted and raised apart) of award-winning novelist Mona Simpson
Maybe that should be "biological brother".
Question with regard to the intersexual dynamics and relationships:
What are the most interesting papers, findings supported with broad anecdotal evidence, and/or your personal predictions that are very surprising/important and absent in the public discourse/understanding of the topic?
Two new books on AI: Age of AI by Eric Schmidt and Henry Kissinger has gotten the most press, but I found it impossible to finish. Boring and repetitive. Did anyone find it valuable?
Much better is Rule of the Robots: how artificial Intelligence will transform everything by Martin Ford. Lots of good stuff on the path toward AGI, based on interviews with people like Demis Hassabis and Geoff Hinton. Also has the best coverage I've seen on the AI race with China. Maybe these two would be good candidates for reviews here.
Happy to help if you need more folks to help judge ACX applications, develop a rubric etc - I’m off of work for the rest of the calendar year so have plenty of time
Has anybody done a study on the effect of "placebo" in treating COVID? Formally, if you give one group an "experimental" pill (which is just a sugar pill), and the other group the same treatment with no pill, which group does better?
Hopefully you could measure this with a meta-analysis comparing placebo-treated people in published experiments with population-level statistics, controlled for age and calendar date [because of variant shifts].
Writing a series on Brain-Computer Interfaces. The last post talks about emerging technologies - some of you might find it interesting. https://howthehell.substack.com/p/bci-part-2
Thanks for signal boosting my post Scott! It took a while to write!
Everyone who cares about the Paxlovid issue should consider submitting a comment to the FDA. It doesn't have to be very long. Here's the direct link where you can submit your comment to be looked at by the committee that will review Merck's EUA application for Molnupiravir:
https://www.regulations.gov/commenton/FDA-2021-N-0758-0012
Note the deadline for your comment to be considered is tomorrow (November 29th) (the meeting is on the 30th).
The comment should probably address something related to Molnupiravir although strictly speaking it doesn't have to. Either way, you should definitely urge the FDA to do the EUA for Paxlovid ASAP!!
See AllAmericanBreakfast's writeup:
https://www.lesswrong.com/posts/5gzkTjPrrNxCDDARc/submit-comments-on-paxlovid-to-the-fda-deadline-nov-29th
(also on the EA Forum : https://forum.effectivealtruism.org/posts/N46uFYAtmrYgneFrC/submit-comments-on-paxlovid-to-the-fda-deadline-nov-29th?fbclid=IwAR31pfEhH5QyV7QiQjJVTAfE696Wxb1rrehdCVnUut9ux1UiTA6lbMFOlyM)
They make this excellent point:
"The FDA still hasn't scheduled a meeting to review Paxlovid. It scheduled its meeting for Molnupiravir 7 weeks out. But it scheduled its review of the vaccines just 3 weeks after the submission. One suggestion, if you submit a comment, is to urge the FDA to schedule its review ASAP and to make the timeline 3 weeks (or less) rather than 7."
Do the rapid tests and RT-PCR tests currently offered in U.S pharmacies such as CVS, work well in detecting the Omicron variant?
I’m a lowly medical student, and I’m creating a medical Education program. Would any of you like to help me develop it or to learn more about it?
I’m coding it myself and I believe you will be intrigued by even its current capabilities : )
Ask for contact information if interested.
Unfortunately it's already been announced/happened, but if there is an ACX Grant Program 2 I think public applications with a prediction market would have been interesting