2: The left-leaning blogosphere (purple) seems tiny compared to the right (orange, yellow). Am I missing something? Does this reflect bifurcation where the left is captured by traditional media and the right embraces "alternative" media like blogs?
17: There are two important Turing tests plant-based meat has to pass: one where it's indistinguishable to my taste buds, and another where it's indistinguishable to my wallet. We shall see if both are achievable...
I never watched Kony 2012 because I was already really annoyed by liberals slacktivism (I was ahead of the curve!).
BUT - from a purely marketing/tagline standpoint I could never understand why they wanted to ELECT this bad guy, Kony, in 2012. Wouldn't deposing/arresting him be preferrable? STOP KONY 2012. Was printing those extra four letters on red t-shirts really that much more expensive?
19. Are you worried that by posting the explanation of the Zhou Enlai quote you weaken society's ability to be circumspect about predicting the future?
But seriously, thank you for doing this every time I see it referenced earnestly I feel motivated to write an annoying pedantic email.
This is one of those things where I have no idea why people trust the underlying data. I get about 6 hours of sleep a night. If you asked me if it was enough, I’d tell you definitely. Ideal, actually. I get to relax at night with my wife after putting the kids to bed, watch some tv, and still get up early enough in the morning to go for a run while still being able to make the kids breakfast. Perfect!
Then one night I’ll turn it at 8 PM and sleep until 6 and OH MY GOD IS THAT GREAT I HAD NO IDEA.
Basically I have no bead at all on what works best for me. I don’t see how Alexey and Natalia would.
#12 This call for missile defence is very dangerous. The blog's argument is as follows.
* NATO has not been able to intervene in the war to the extent we would like because of Russia's nuclear weapons
* we should build missile defences so that Russia's retaliation threat is not sufficient to deter NATO involvement
The blogpost is literally saying that we should build up these defences so that we can credibly threaten to go to war with Russia without the risk of nuclear response being overwhelming. Given that large residual risk will still be there (as these systems are impressive but highly imperfect), this is inviting strategic error on an X-risk scale.
MAD is a crazy paradigm, but it's going to take a lot to convince me that we know that another alternative is safer especially one that attempts to neuter a nuclear adversary.
2. Is there a way to find particular blogs in the graph?
Blog rolls strike me as a very approximate measure, though probably better than nothing. My impression is that people don't keep track of their blog rolls.
For 8, doing it per capita based on current population makes no sense. The prizes should be divided by average births over the relevant time period, or something. This unduly favors historically large and currently less large states, and handicaps historically small but currently large states.
13. Nazarbayev is by most accounts the real power in Kazakhstan. He is, despite being an awful person, also one of the saner dictators. He tortures and murders people. But usually for rational, if immensely selfish and monstrous, reasons. He's also done a lot of institution building. The rumor is he basically wants to make Kazakhstan into Singapore and he's been making moves in the direction of at least a semi-democracy since 2010 or so. This doesn't seem like a notable break from that trend.
18. They were also common under Communism. There's a lot more attention to how Communist states controlled women's reproduction because that included abortion which is a modern hot button issue. But the Communist state wanted to control everyone's really. Some other dodges: get betrothed to a woman and then break it off. Marry a someone just as they're dying to immediately become a widow(er). In societies with the relevant exemption take minor orders and play at debating becoming a priest...
19. Another famous Chinese misquote. "To get rich is glorious." Deng Xiaopeng never said that. What he said was, "We have to let some people get rich first." It was a reference to Marxist theory not some general statement China was selling out to Mammon.
25. Yes. The issue will be the scale of the cleanup and the loss of key industrial capacity hampering that effort. The idea it will destroy all life is an exaggeration by anti-nuclear activists who tend to spill over into criticizing even peaceful uses. There's a lot of people very invested in making nuclear energy look scary for what I think are bad reasons. Of course, strategically nuclear war would still be so destructive as to be worth it only in the direst circumstances. But you'd end up with something more like post-WW2 Europe and less like Fallout. To be clear, post-WW2 Europe was REALLY BAD.
30. China already does this. Fines are deducted directly from bank accounts as a matter of "convenience." So in Shanghai if a camera sees you outside they just auto-fine you. The thing is that digitization of finance is coming and we as a society have to decide how, not if, we're going to do it. The option of remaining in the past not only hobbles us but won't really work for long. Alas, all the political discussion about finance is dominated by a combination of conservative outrage over corporate liberalism and progressive desires for mostly pretty bad regulatory proposals. This is one thing where the dictatorships have an advantage. There's an easy answer as to who this digitization should serve. The center.
9: I don't blame the thread author for omitting such a complex topic in their brief overview, but I would be interested to hear their thoughts on the status of the "Q source". Last I checked, there was active debate among scholars whether it's (1) a genuine source carrying the tradition forward from M; (2) just that subset of M with the most messianic character; or (3) a spurious invention by detractors of the Trumpist tradition, created to tar the tradition by association with absurd eschatological claims.
I'm more concerned about this bit of a text from the link of [26]:
> Adult males are flower visitors, while females do not feed at all.
How does the females 'not feeding at all' work? They seem way too big to have an extremely short lifespan. Or do they just feed enough as larvae to both transform into adults and then survive just long enough until they can reproduce?
The Wikipedia article on the family of that fly _does_ state:
> They [adults] are infrequently encountered as the adult lifespan can be quite short.
Am I wrong or does the Vitamin D study actually demonstrate the lack of effect of sending people Vitamin D supplements in the mail? Will the full paper maybe describe the follow up finger prick testing to see if serum level actually changed in the supplement groups? Also, were the positive tests for ARIs or Covid random with respect to serum level change or was there a relationship there?
14: "Most neuroimaging studies have samples in the the 2-3 digits, but would need to be in the 4-5 digits to have enough power to detect real effects." This is misleading. The linked article is about neuroimaging studies seeking associations with disorders like schizophrenia and autism. Some neuroimaging effects can be large and highly reliable: you can distinguish the functional neuroimaging responses of left-hand vs. right-hand motor activation with your eyes in real-time, with a single subject and 20 seconds of data. The need for 4-5 digit sample sizes may have a little bit to do with neuroimaging and its acquisition and sampling properties, but it is mostly about the statistical nightmare of identifying group differences in brains associated with heterogeneous or ill-defined psychiatric disorders.
Huh, I always wrote the second person imperative of "salir" as "sale".
People take this as a criticism of Spanish but it's great to have a language that actually has rules that can be applied to deduce the ways words are written or spoken.
6: I think the start of the thread is broadly good, but then it goes off the rails.
The poster includes some factual inaccuracies, especially here (https://twitter.com/ctbeiser/status/1500257579145916418). Jason Russel did have a mental break, but he was never jacking off. It was a one off line in some gawker shit and it caught on. Russel denies jacking off, he isn't jacking off in the film, and the police never claimed he was nor was he charged as such.
The claim that Russel was drugged is also pretty crazy (just apriori), but especially because he did an interview on Oprah (https://youtu.be/Q7MCh19igW4) where he talked about the whole thing and his explanation is total reasonable. His pet project blew up beyond all imagining, and he was doing back-to-back interviews and running on fragments of sleep, while being hounded by random journalists and internet people.
> A possible causal pathway is “some unknown cause X led Rome to prosper economically, which in turn led Rome to military success, which in turn caused Roman culture to spread” And one excellent candidate for X is Rome’s location.
I notice a pattern in history: most pre-industrial civilizations put traders as the lowest status class. Some where official about this (eg China) while in other places it was unofficial. Places that didn't do this, and instead treated traders at least somewhat fairly all seem to have ended very wealthy for the pre-industrial period.
I think the Roman Republic was another case. They treated traders fairly and because of that the traders generated wealth for the republic.
>A bachelor tax existed in Argentina around 1900. Men who could prove that they had asked a woman to marry them and had been rebuffed were exempt from the tax. In 1900, this gave rise to the phenomenon of "professional lady rejectors", women who for a fee would swear to the authorities that a man had proposed to them and they had refused.”
That's a bit like the "professional co-respondent of mid twentieth century UK. Divorce was not easily attainable except in cases of adultery, so couples who merely hated each other would arrange fake adultery with a third party , the professional co-repondent and a private detective who would spring out of a wardrobe or other hiding place to.provide photographic evidence.
Professional co-respondents wore a special kind of shoe.
11: I thought we knew this already. No? Same for the Zhou Enlai quote.
26. As long as it doesn't bite.
32: I don't know if I'm in the minority, but losing weight was fairly easy when I was a kid. I had to do so for sports-related reasons on a couple different occasions, and it seemed like calories in/calories out really was the operative dynamic. That all changed when I hit....26, maybe? It got steadily harder afterward, eventually becoming nigh on impossible. Point being that shaming might work on kids, but probably not anyone else. Also, if you're looking to solve our global obesity problems, maybe try to figure out what changes in people's metabolism/endocrine systems between say 16-26.
I'm a bit skeptical on #25. Yes, it's not the first time I've heard that nuclear-winter models in the 80s were wrong - I'll ask a clever friend in applied maths about this. It's the claim that "you will not find peer-reviewed papers on "Nuclear war is not as bad you think" because money and politics go against this." I mean - there are peer-reviewed papers on how some well-publicized estimates on the proportion of women (or people) in the US that will survive rape in their lifetimes may be gross overestimates, or are based on flawed methodology, and surely publishing that is worse for one's professional and personal reputation. What would happen if someone wrote the kind of peer-reviewed revision downwards of the effects of nuclear warfare that #25 seems to think both true and non-existent? Would something start with someone on Twitter saying "Hint hint, I wonder where this American was on August 6, 1945"?
28: I can't claim to understand it a whole lot, but, reading the abstract, it sounds to me like what they did was take a bunch of macaque faces, take parts of them, and put those parts together with a bunch of non-face objects. And then found that the response was more dependent on the non-face objects than the face parts (or something). The idea is that the part of your brain that lights up for faces is atomic, and lights up for faces, and not made up of like, a part that lights up for noses, and a part that lights up for eyes, and a part that lights up for mouths, and then an aggregator of all that.
(At least, I think, I still don't understand this line: "we discovered graded tuning for non-face objects that was more predictive of face preference than was tuning for faces themselves" or the next "The relationship between category-level face selectivity and image-level non-face tuning was not predicted by color and simple shape properties, but by domain-general information encoded in deep neural networks trained on object classification" In any explainable way. But this is the general impression I got from it.)
(Which kinda sounds analogous to me, interestingly.)
You had me all worked up, but the World's Largest Fly appears to be rather friend-shaped! Looks like a funny but harmless cartoon critter to me. Much less freaky than a regular fruit fly blown up to the same size with no changes, really.
I was shocked by the idea that there were 5 Nobel Physicists from DC. With the help of the below tutorial, I produced a query that lists 4 of them. The 5th is David Gross, whose wikipedia metadata is inconsistent with the others.
I'm confused by the meat Turing test one because there are already people who say they can't tell the difference between some plant based burger patties and meat ones.
27 -- this is just a standard case of post-treatment bias, right? You can't run a regression on race and control for something race affects, because the answer you get is impossible to interpret unless you find a way to incorporate all the possible causal channels connecting race, test results, and future income. Notably, this regression *doesn't* measure the "effects of racism beyond those affecting test scores" or anything like that -- the bias in this sort of problem doesn't come with a clear interpretation, magnitude, or direction. (See e.g. this paper by Bohren, Hull, and Imas for a couple of causal channels that would bias this regression in hard-to-interpret directions: https://twitter.com/instrumenthull/status/1503385611050180610 , and this paper by Durlauf and Heckman explaining the error in a similarly flawed regression: https://twitter.com/brendannyhan/status/1287089470152552450)
As someone who did a small-n neuroimaging study (fMRI) and read many more of the same as part of the research, can confirm they are garbage and generally should be considered p-hacked or otherwise wrong unless rigourously proven otherwise.
Results from the national adult literacy survey show that blacks with post-graduate degrees have lower literacy and numeracy skills than college drop-out whites. Black 4-year degree graduates have only slightly better skills than white high school graduates.
It is perfectly rational to have a preference for white applicants over black applicants who are "equally qualified", because "equally qualifed" doesn't actually mean of equal ability. But this is precisely what we should expect considering the scale of affirmative action in this country and the fact that education does not magically transform a person's intellect
24 and 25 on nuclear war seem to be naive and localist in terms of the impacts of such an event. Nearly by definition it means total war and the most vicious of tactics and targeting would be used. Power, fuel, transport, communications, etc would be targeted along with military and political targets. That time a few years ago when Houston or Dallas or wherever had some issue with petrol refining caused months of issues and shortages. And that was a minor disruption which was quickly resolved. If the entire infrastructure of key sites were gone, it’d collapse civilisation for decades if not longer.
Covid showed how anaemic and poor we have become as lean and corporate ideologies of greed destroyed and hollowed out all redundancy and created fragile global supply chains which would cease to function if all or most major ports were knocked out. Even now a small hiccup in supply chains has seen cargo ships lined up for weeks and months unable to deliver their goods due to issues with ports, trucks, and warehouses.
In Christchurch in New Zealand from the 2011 earthquake they were literally unable to rebuild by themselves and many key infrastructure jobs were impossible to compete with local knowledge as no one has built a sewage system from near scratch or electrical grid wholesale in decades. They had to source talent from all over the world in order to try and even over decade later they are not nearly back to where they were. That’s ‘only’ an earthquake and they didn’t have to deal with the other aspects of a MAD catastrophe. Many of the old churches and such were simply impossible to replace with skills to make them being over 100 or 150 years out of date and only a handful of people around the world who could repair them, but couldn’t necessarily build them.
Looking at how bad regional disasters have been and how incredibly reliant on National and intentional responses they were to rebuild themselves over 10 and 20 year periods….if a nation were crippled and unable to draw resources and know how from across an empire to restore a single city, but had to deal with dozens of major cities being knocked out…life would collapse to a very simplified existence for decades and decades with non impacted regions becoming the new global powers. It could take 100 or 200 years to recover…with peaks and valleys of new technology, ie mobile phones and mules for farmers so that 2100 would be both better and worse off than we were in 1800 across various dimensions.
This ‘not so bad’ crap for a MAD scenario is naive to the extreme and thinks of only one city at a time or ignores the compounding effect of so many systems being lost at once where people in cities will start murdering each other for food within a week of the truck deliveries stopping.
Even looking at the Marshall plan and such in Europe from the USA, the UK had meat rationing into the 1960s from a war that ended in 1945…and nominally the UK was on the winning side! Over 20 years to restore just cattle and meat production and distribution! How or why would we ever think something orders of magnitude worse, such as a doomsday MAD scenario, would not be as bad? It’d be much worse and we are much more fragile and reliant on brittle paper thin supply chains now vs how the UK was back then when people were closer to the land in terms of knowledge of food growing and non electric preservation techniques. Would we be better or worse in growing victory gardens today vs back then? Worse I’d say, and we would be trying to feed a lot more people.
The message must be that we never ever have a MAD scenario play out. Softening that vital message by simplistically looking at blast radius zones is utter nonsense.
#11: The willingness to expand 'Romanness' to conquered people as the main reason for Roman success is a major proposition in Mary Beard's work, and is the thesis of her popular-audience book 'SPQR'. I was a little surprised that Grunewald didn't reference it, but I enjoyed his take and it is convincing to me.
#30: I saw this twitter manifesto at Zvi's back in February, and made a similar comment there, but: the statement that 'there are no constitutional rights in substance without freedom to transact' is impossible to divorce, in my mind, from 'there are no constitutional rights in substance without having means to transact with.'
I accept that there is a difference between being prevented by a government from using existing resources that are in some sense 'your own' and not having those resources in the first place. But if a transaction must take place for a right to exist in substance, then both the means and the freedom to use them must be present. Else we are conceding that constitutional rights are contingent on wealth.
27 - the reddit post is from October 2021. The author has an updated (as of January of this year) version at his website https://reasonwithoutrestraint.com/race-and-iq/ - many other interesting analyses of 'controversial' topics as well.
I've already had a panic attack eating an Impossible Cheeseburger, "how did I get here this isn't kosher", even though of course it is. It wasn't the first bite, but it was when I was distracted a few bites in talking to a friend.
So I have very high confidence that fake meat will pass the "turing test". MOOring Test?
4. The Vox article on Georgism seems much worse than the writings about Georgism I've seen here. Or maybe I don't fully understand the argument.
This article suggests that there is a close relationship between the land value tax and overly restrictive zoning laws. What is this relationship?
Parts of the article pretend that zoning laws don't exist. "And if someone turns their garage into an apartment, providing an affordable housing option for their community, they pay higher property taxes than similarly situated neighbors who don’t add housing options to their land." The reason this doesn't happen more often isn't because of the taxes. It's because it's illegal.
Parts of the article claim that a land value tax will cause overly restrictive zoning laws to disappear. "It turns NIMBYs into YIMBYs." I want to know how this happens. If I don't want a big apartment building going up next door because it's too noisy, how would reducing their taxes change my opinion? If anything, I would expect that a land value tax makes NIMBYs worse: the new development next door could increase my land value, and so my taxes.
I do think that a land value tax is a useful tool to reduce vacancies & make the tax system more fair. I also think that zoning reform should be implemented in lots of places. But these aren't the same problem. The land value tax seems most useful in places with many vacant lots and urban blight, while zoning reform seems most useful in places with no vacant lots and high rents.
Why has this idea become popular recently? Probably because it's a pretty good idea and people are putting forward good arguments for it. Not because "we've run out of suburbs." We haven't. Before the pandemic, suburbs were growing faster than urban or rural areas [1], and I expect that trend increased with more people working remotely.
To suggest an answer to Naval Gazing's question, I think some of the Western appreciation for Russian equipment is that the designs are often good on paper and well fit to novice users. "Deep Battle", the Soviet WWII doctrine of combined arms, can reportedly be implemented by wargamers with modest skill and a few minutes of prep time. It works even with the minimal coordination you can usually achieve in video games. Soviet strategies are brutally straightforward, and their tanks are designed to match. The fact that the vehicles are often manufactured, maintained, and employed well below the design is often ignored. That they have bad ergonomics and situational awareness is not necessarily understood or experienced by enthusiasts or gamers. By contrast, Western strategies are a lot more complex and subtle. The tanks have tradeoffs that are exploitable if not used by professionals, or when used outside a Western combined arms milieu.
But anti-missile systems are definitely an essential technology for every serious military going forward. A well-supported, well-equipped MBT is scarily survivable, even if poorly deployed and defended tanks get got. As a case study, Israel lost *zero* main tanks or modern APCs to Hamas in Gaza in 2014, despite facing the most advanced Russian missiles and RPGs available at the time. These APS systems are even scarier if those systems can not only shoot down the incoming projectile but simultaneously cue the tank where it was shot from; you might not live long enough to see your ATGM defeated harmlessly in front of the tank.
The missile cruiser Moskva has been confirmed sunk by Russia on April 14 [1]. Ukraine claims they shot it with two cruise missiles on April 13 [2]. On the same day, April 13, Ukraine released the stamp commemorating the small force on Snake Island telling the Moskva to go fuck itself [3].
Coincidence? I. THINK. NOT.
I'm not in general a conspiracy theorist. The only other fringe conspiracy theory I believe is that "smart lights" only exist because lightbulb companies haven't figured out how to make a mediocre LED. But the timing is too funny to not believe.
P.S. there is truly bonkers rumor that there was Christian relic onboard the Moskva [4]. If 2022 was pitched as the next season of Servant of the People, nobody would have bought it.
When I first saw "There are no constitutional rights in substance without freedom to transact" idea, I thought that it's interesting but obviously proves too much. For instance, that there are no constitutional rights without post scarcity communist society.
If we assume that constitutional rights require the ability to spend money, we end up with a situation where some people have more money, therefore more constitutional rights than other people, therefore making these rights not rights but privileges.
Of course the satire in 9 cuts both ways: intentionally or not, it puts forward a case that either Jesus is a composite figure, or the Gospel accounts are based on several oral sources that got garbled in transmission, or, if neither is the case, Jesus was as inconsistent, opportunistic and/or psychologically unstable as Trump.
Re: 20, comments in that link mention the fact that English is the only major language that doesn't have a governing body. I think this is under-reported!
I'm pretty sure English is the only officially-democratic language with more then a million speakers. That's based on my own original research though, I haven't seen anyone else talk about it. I feel like it captures something about the libertarian anglophone philosophy, and I would love to learn more about how this came to be the case. Did no one ever try? When did most linguisitc-authorities come into existence? Has it had any measurable effects on English development? Etc.
14. is misleading. This is specifically in reference to brain-wide association studies (BWAS), e.g., "let's try to correlate, uh, which Pokémon starter people prefer to any possible set of structural and functional indices." These studies work off of questionable hypotheses.
Neuroimaging studies with well-argued hypotheses (i.e., NOT grounded in BWAS and/or generically bad science) and relatively modest sample sizes have a lot to offer.
The Banach story reminds me of the protagonist of Le Guins The Dispossessed. He's a leading expert in theoretical physics, and when he visits the sister planet they insist on calling him "Dr." because a guy like that couldn't possibly not have a credentialed title!
I haven't gone and read the methodology yet, but the blogosphere map seems very out-of-date. It's got the old version of Pharyngula on there, which I don't think has been used for 5 years. And Worthwhile, which barely functions as a blog any more. It's a weird combination of the blogosphere now and what it looked like 5-10 years ago.
Maybe that's just because things have weird half-lives on the internet when they're not very active any more, but they're still available. But it makes the map a bit difficult to parse.
Isn't the fourth one a subset of the third? Trump may not be clinically insane (but then again he may be) - but that just drives home that the trilemma, as stated, is a false trichotomy; Jesus may have been a moral teacher who, like a lot of other young men, also had what we now call schizotypal traits. (It is difficult to see how someone could get into the prophet/messiah track, common at the time, and *not* have schizotypal traits.)
2: Cool graph, but it can't be complete. Fox is a tiny yellow dot. Where's the tiny purple dot for NYT? Do humans so rarely link to these old corporate media outlets from their blogs that ACX is 30 times bigger than fox in the link graph and NYT doesn't even show up? [Edit: Krugman is a red dot a quarter the size of ACX but the rest of NYT is absent] Does almost all their traffic come from deliberate subsidies by big tech instead of humans linking to them? It approximately seems that way. Prior to Youtube's public decision to subsidize old corporate media by bumping them waaaaay up in all search results, they were getting almost no traffic on youtube.
Categories seem to be
purple: rationalist blogs
red: left/center econ blogs
yellow: right econ blogs
orange: mostly libertarian blogs
yellow: righty blogs (single-issue free speech promoting blogs like thefire.org and volokh get miscategorized as yellow or orange)
purple: lefty blogs
green: wall street (mix of apolitical trading blogs and hyperpartisan blogs that mention wall street)
26: Why is the size-scaling limit for avians >1 OOM larger than the size-scaling limit for flying arthropods?
32: Of course people respond to incentives and fat shaming is an incentive. But you're fighting an uphill battle against hunger unless you categorically change which foods are available to eat. Some foods are vastly more satiating per calorie than others. Anything that spikes insulin is bad because it makes the brain less sensitive to leptin and causes more hunger on the rebound. Meats and vegetables are great. (I have succeeded in reducing my bodyfat percentage from 20% to 16% over the last two months using the rules described here: https://www.lesswrong.com/posts/PPeSqatTRsuFw3ug9/food-manufacturers-are-out-to-get-you)
Singapore probably could have got much more than that 33% reduction in childhood obesity without fat-shaming anyone if only they banned all caloric beverages other than milk.
"New finding suggest the cult of Trump has much wider geographical extent than previously thought:
According to the new manuscript uncovered at archaeological site on the Pontic steppe, sometime in the second decade of 21 century the settlement of Novgorodskoye was renamed to New York after the mythological birthplace of Trump. The locals believed the renaming will help to earn the favour of the powerful deity and protect the settlement from the destruction by Ork Horde."
28: One of my favorite findings in cognitive science! The paper that started it all is Dehaene (2007). One of the more interesting examples - why are there such strong (complete?) overlap in hippocampal processing of episodic memory & egocentric navigation? How about semantic memory & allocentric (landmark) navigation? Arguably because of deep similarities between the two functions. Buzsaki & Tingley (2018) argue that HF performs sequence processing, leveraged across diverse applications.
Other examples of cultural recycling discussed in Dehaene (2007) include reading & face recognition, arithmetic vs eye movement, music vs speech. These are all construed as needing similar cost functions.
Marblestone et al (2016) argue that DL is moving towards a similar design pattern, which will move us much closer towards AGI.
> We hypothesize that (1) the brain optimizes cost functions, (2) the cost functions are diverse and differ across brain locations and over development, and (3) optimization operates within a pre-structured architecture matched to the computational problems posed by behavior. In support of these hypotheses, we argue that a range of implementations of credit assignment through multiple layers of neurons are compatible with our current knowledge of neural circuitry, and that the brain's specialized systems can be interpreted as enabling efficient optimization for specific problem classes. Such a heterogeneously optimized system, enabled by a series of interacting cost functions, serves to make learning data-efficient and precisely targeted to the needs of the organism. We suggest directions by which neuroscience could seek to refine and test these hypotheses.
Buzsaki & Tingley (2018). Space and Time: The Hippocampus as a Sequence Generator
Dehaene (2007). Cultural recycling of cortical maps
Marblestone et al (2016). Toward an Integration of Deep Learning and Neuroscience
“ 1: History of the belief that garlic and magnets are natural enemies.” Pliny falls victim to the replication crisis.
2: The left-leaning blogosphere (purple) seems tiny compared to the right (orange, yellow). Am I missing something? Does this reflect bifurcation where the left is captured by traditional media and the right embraces "alternative" media like blogs?
17: There are two important Turing tests plant-based meat has to pass: one where it's indistinguishable to my taste buds, and another where it's indistinguishable to my wallet. We shall see if both are achievable...
Almost certainly unoriginal but I have to say it: #33 is clearly wrong because that doesn't look like Elon Musk.
I never watched Kony 2012 because I was already really annoyed by liberals slacktivism (I was ahead of the curve!).
BUT - from a purely marketing/tagline standpoint I could never understand why they wanted to ELECT this bad guy, Kony, in 2012. Wouldn't deposing/arresting him be preferrable? STOP KONY 2012. Was printing those extra four letters on red t-shirts really that much more expensive?
19. Are you worried that by posting the explanation of the Zhou Enlai quote you weaken society's ability to be circumspect about predicting the future?
But seriously, thank you for doing this every time I see it referenced earnestly I feel motivated to write an annoying pedantic email.
Bret Devereaux has a great series of posts touching on #11 with much the same thesis. https://acoup.blog/2021/06/25/collections-the-queens-latin-or-who-were-the-romans-part-ii-citizens-and-allies/
This is also a good nuclear discussion w Luisa Rodríguez and Rob Wiblin
https://80000hours.org/podcast/episodes/luisa-rodriguez-why-global-catastrophes-seem-unlikely-to-kill-us-all/
#23 (the debate on how much sleep you need)
This is one of those things where I have no idea why people trust the underlying data. I get about 6 hours of sleep a night. If you asked me if it was enough, I’d tell you definitely. Ideal, actually. I get to relax at night with my wife after putting the kids to bed, watch some tv, and still get up early enough in the morning to go for a run while still being able to make the kids breakfast. Perfect!
Then one night I’ll turn it at 8 PM and sleep until 6 and OH MY GOD IS THAT GREAT I HAD NO IDEA.
Basically I have no bead at all on what works best for me. I don’t see how Alexey and Natalia would.
Is it fair to say that the French Revolution never really ended?
I propose we call the plant meat Turing test the Moo-ring test. Because it's to see if you can accurately determine if your meal ever moo'd.
#12 This call for missile defence is very dangerous. The blog's argument is as follows.
* NATO has not been able to intervene in the war to the extent we would like because of Russia's nuclear weapons
* we should build missile defences so that Russia's retaliation threat is not sufficient to deter NATO involvement
The blogpost is literally saying that we should build up these defences so that we can credibly threaten to go to war with Russia without the risk of nuclear response being overwhelming. Given that large residual risk will still be there (as these systems are impressive but highly imperfect), this is inviting strategic error on an X-risk scale.
MAD is a crazy paradigm, but it's going to take a lot to convince me that we know that another alternative is safer especially one that attempts to neuter a nuclear adversary.
2. Is there a way to find particular blogs in the graph?
Blog rolls strike me as a very approximate measure, though probably better than nothing. My impression is that people don't keep track of their blog rolls.
For 8, doing it per capita based on current population makes no sense. The prizes should be divided by average births over the relevant time period, or something. This unduly favors historically large and currently less large states, and handicaps historically small but currently large states.
13. Nazarbayev is by most accounts the real power in Kazakhstan. He is, despite being an awful person, also one of the saner dictators. He tortures and murders people. But usually for rational, if immensely selfish and monstrous, reasons. He's also done a lot of institution building. The rumor is he basically wants to make Kazakhstan into Singapore and he's been making moves in the direction of at least a semi-democracy since 2010 or so. This doesn't seem like a notable break from that trend.
18. They were also common under Communism. There's a lot more attention to how Communist states controlled women's reproduction because that included abortion which is a modern hot button issue. But the Communist state wanted to control everyone's really. Some other dodges: get betrothed to a woman and then break it off. Marry a someone just as they're dying to immediately become a widow(er). In societies with the relevant exemption take minor orders and play at debating becoming a priest...
19. Another famous Chinese misquote. "To get rich is glorious." Deng Xiaopeng never said that. What he said was, "We have to let some people get rich first." It was a reference to Marxist theory not some general statement China was selling out to Mammon.
25. Yes. The issue will be the scale of the cleanup and the loss of key industrial capacity hampering that effort. The idea it will destroy all life is an exaggeration by anti-nuclear activists who tend to spill over into criticizing even peaceful uses. There's a lot of people very invested in making nuclear energy look scary for what I think are bad reasons. Of course, strategically nuclear war would still be so destructive as to be worth it only in the direst circumstances. But you'd end up with something more like post-WW2 Europe and less like Fallout. To be clear, post-WW2 Europe was REALLY BAD.
30. China already does this. Fines are deducted directly from bank accounts as a matter of "convenience." So in Shanghai if a camera sees you outside they just auto-fine you. The thing is that digitization of finance is coming and we as a society have to decide how, not if, we're going to do it. The option of remaining in the past not only hobbles us but won't really work for long. Alas, all the political discussion about finance is dominated by a combination of conservative outrage over corporate liberalism and progressive desires for mostly pretty bad regulatory proposals. This is one thing where the dictatorships have an advantage. There's an easy answer as to who this digitization should serve. The center.
14. Best paper of all time. http://prefrontal.org/files/posters/Bennett-Salmon-2009.pdf.
9: I don't blame the thread author for omitting such a complex topic in their brief overview, but I would be interested to hear their thoughts on the status of the "Q source". Last I checked, there was active debate among scholars whether it's (1) a genuine source carrying the tradition forward from M; (2) just that subset of M with the most messianic character; or (3) a spurious invention by detractors of the Trumpist tradition, created to tar the tradition by association with absurd eschatological claims.
I'm more concerned about this bit of a text from the link of [26]:
> Adult males are flower visitors, while females do not feed at all.
How does the females 'not feeding at all' work? They seem way too big to have an extremely short lifespan. Or do they just feed enough as larvae to both transform into adults and then survive just long enough until they can reproduce?
The Wikipedia article on the family of that fly _does_ state:
> They [adults] are infrequently encountered as the adult lifespan can be quite short.
Wild
Am I wrong or does the Vitamin D study actually demonstrate the lack of effect of sending people Vitamin D supplements in the mail? Will the full paper maybe describe the follow up finger prick testing to see if serum level actually changed in the supplement groups? Also, were the positive tests for ARIs or Covid random with respect to serum level change or was there a relationship there?
I liked the “historical Trump” thread. A good satire on Jesus studies.
That Metaculus link is *conditional on such a test being conducted* which dramatically changes the interpretation.
14: "Most neuroimaging studies have samples in the the 2-3 digits, but would need to be in the 4-5 digits to have enough power to detect real effects." This is misleading. The linked article is about neuroimaging studies seeking associations with disorders like schizophrenia and autism. Some neuroimaging effects can be large and highly reliable: you can distinguish the functional neuroimaging responses of left-hand vs. right-hand motor activation with your eyes in real-time, with a single subject and 20 seconds of data. The need for 4-5 digit sample sizes may have a little bit to do with neuroimaging and its acquisition and sampling properties, but it is mostly about the statistical nightmare of identifying group differences in brains associated with heterogeneous or ill-defined psychiatric disorders.
Didn't Hellenistic people settle on the north coast of the Black Sea in antiquity? How many cities derive their names from those settlements?
Huh, I always wrote the second person imperative of "salir" as "sale".
People take this as a criticism of Spanish but it's great to have a language that actually has rules that can be applied to deduce the ways words are written or spoken.
> 2: Jacob Wood’s Graph Of The Blogosphere.
Maybe you could use RSS to make something with this UI that lets you select blogs and shows their latest posts.
6: I think the start of the thread is broadly good, but then it goes off the rails.
The poster includes some factual inaccuracies, especially here (https://twitter.com/ctbeiser/status/1500257579145916418). Jason Russel did have a mental break, but he was never jacking off. It was a one off line in some gawker shit and it caught on. Russel denies jacking off, he isn't jacking off in the film, and the police never claimed he was nor was he charged as such.
The claim that Russel was drugged is also pretty crazy (just apriori), but especially because he did an interview on Oprah (https://youtu.be/Q7MCh19igW4) where he talked about the whole thing and his explanation is total reasonable. His pet project blew up beyond all imagining, and he was doing back-to-back interviews and running on fragments of sleep, while being hounded by random journalists and internet people.
11 (Rome)
> A possible causal pathway is “some unknown cause X led Rome to prosper economically, which in turn led Rome to military success, which in turn caused Roman culture to spread” And one excellent candidate for X is Rome’s location.
I notice a pattern in history: most pre-industrial civilizations put traders as the lowest status class. Some where official about this (eg China) while in other places it was unofficial. Places that didn't do this, and instead treated traders at least somewhat fairly all seem to have ended very wealthy for the pre-industrial period.
I think the Roman Republic was another case. They treated traders fairly and because of that the traders generated wealth for the republic.
>A bachelor tax existed in Argentina around 1900. Men who could prove that they had asked a woman to marry them and had been rebuffed were exempt from the tax. In 1900, this gave rise to the phenomenon of "professional lady rejectors", women who for a fee would swear to the authorities that a man had proposed to them and they had refused.”
That's a bit like the "professional co-respondent of mid twentieth century UK. Divorce was not easily attainable except in cases of adultery, so couples who merely hated each other would arrange fake adultery with a third party , the professional co-repondent and a private detective who would spring out of a wardrobe or other hiding place to.provide photographic evidence.
Professional co-respondents wore a special kind of shoe.
Does Alexey Guzey have kids?
11: I thought we knew this already. No? Same for the Zhou Enlai quote.
26. As long as it doesn't bite.
32: I don't know if I'm in the minority, but losing weight was fairly easy when I was a kid. I had to do so for sports-related reasons on a couple different occasions, and it seemed like calories in/calories out really was the operative dynamic. That all changed when I hit....26, maybe? It got steadily harder afterward, eventually becoming nigh on impossible. Point being that shaming might work on kids, but probably not anyone else. Also, if you're looking to solve our global obesity problems, maybe try to figure out what changes in people's metabolism/endocrine systems between say 16-26.
I'm a bit skeptical on #25. Yes, it's not the first time I've heard that nuclear-winter models in the 80s were wrong - I'll ask a clever friend in applied maths about this. It's the claim that "you will not find peer-reviewed papers on "Nuclear war is not as bad you think" because money and politics go against this." I mean - there are peer-reviewed papers on how some well-publicized estimates on the proportion of women (or people) in the US that will survive rape in their lifetimes may be gross overestimates, or are based on flawed methodology, and surely publishing that is worse for one's professional and personal reputation. What would happen if someone wrote the kind of peer-reviewed revision downwards of the effects of nuclear warfare that #25 seems to think both true and non-existent? Would something start with someone on Twitter saying "Hint hint, I wonder where this American was on August 6, 1945"?
28: I can't claim to understand it a whole lot, but, reading the abstract, it sounds to me like what they did was take a bunch of macaque faces, take parts of them, and put those parts together with a bunch of non-face objects. And then found that the response was more dependent on the non-face objects than the face parts (or something). The idea is that the part of your brain that lights up for faces is atomic, and lights up for faces, and not made up of like, a part that lights up for noses, and a part that lights up for eyes, and a part that lights up for mouths, and then an aggregator of all that.
(At least, I think, I still don't understand this line: "we discovered graded tuning for non-face objects that was more predictive of face preference than was tuning for faces themselves" or the next "The relationship between category-level face selectivity and image-level non-face tuning was not predicted by color and simple shape properties, but by domain-general information encoded in deep neural networks trained on object classification" In any explainable way. But this is the general impression I got from it.)
(Which kinda sounds analogous to me, interestingly.)
Link 9 feels like successor of very good quality 19th century French satire The Napoleon Myth (https://ia600504.us.archive.org/34/items/napoleonmyth00evan/napoleonmyth00evan.pdf" or "https://gallica.bnf.fr/ark:/12148/bpt6k5657559b/f2.item.texteImage" in the original French). This small pamphlet "demonstrates" through ~~coincidences~~ that Napoleon as just a reinterpretation of the Sun-God Apollo, in a very clever parody of gotchatheists.
You had me all worked up, but the World's Largest Fly appears to be rather friend-shaped! Looks like a funny but harmless cartoon critter to me. Much less freaky than a regular fruit fly blown up to the same size with no changes, really.
I was shocked by the idea that there were 5 Nobel Physicists from DC. With the help of the below tutorial, I produced a query that lists 4 of them. The 5th is David Gross, whose wikipedia metadata is inconsistent with the others.
tutorial:
https://tinysubversions.com/notes/how-to-query-wikipedia/
https://dbpedia.org/snorql/?query=SELECT+%3Fwinner+WHERE+%7B+%0D%0A+%3Fwinner+%3Chttp%3A%2F%2Fpurl.org%2Fdc%2Fterms%2Fsubject%3E+%3Chttp%3A%2F%2Fdbpedia.org%2Fresource%2FCategory%3ANobel_laureates_in_Physics%3E%0D%0A+.%0D%0A+%3Fwinner+%3Chttp%3A%2F%2Fdbpedia.org%2Fontology%2FbirthPlace%3E+%3Chttp%3A%2F%2Fdbpedia.org%2Fresource%2FWashington%2C_D.C.%3E%0D%0A%7D%0D%0A
I'm confused by the meat Turing test one because there are already people who say they can't tell the difference between some plant based burger patties and meat ones.
27 -- this is just a standard case of post-treatment bias, right? You can't run a regression on race and control for something race affects, because the answer you get is impossible to interpret unless you find a way to incorporate all the possible causal channels connecting race, test results, and future income. Notably, this regression *doesn't* measure the "effects of racism beyond those affecting test scores" or anything like that -- the bias in this sort of problem doesn't come with a clear interpretation, magnitude, or direction. (See e.g. this paper by Bohren, Hull, and Imas for a couple of causal channels that would bias this regression in hard-to-interpret directions: https://twitter.com/instrumenthull/status/1503385611050180610 , and this paper by Durlauf and Heckman explaining the error in a similarly flawed regression: https://twitter.com/brendannyhan/status/1287089470152552450)
33. Can.. uh.. can anybody remember me why we are doing this instead of banning neural networks?
The absence of the usual "misleading links" banner implies that all the linked opinions are now true in the past and in the future.
As someone who did a small-n neuroimaging study (fMRI) and read many more of the same as part of the research, can confirm they are garbage and generally should be considered p-hacked or otherwise wrong unless rigourously proven otherwise.
27
Results from the national adult literacy survey show that blacks with post-graduate degrees have lower literacy and numeracy skills than college drop-out whites. Black 4-year degree graduates have only slightly better skills than white high school graduates.
https://web.archive.org/web/20170501225350/https://nces.ed.gov/pubs93/93275.pdf
It is perfectly rational to have a preference for white applicants over black applicants who are "equally qualified", because "equally qualifed" doesn't actually mean of equal ability. But this is precisely what we should expect considering the scale of affirmative action in this country and the fact that education does not magically transform a person's intellect
24 and 25 on nuclear war seem to be naive and localist in terms of the impacts of such an event. Nearly by definition it means total war and the most vicious of tactics and targeting would be used. Power, fuel, transport, communications, etc would be targeted along with military and political targets. That time a few years ago when Houston or Dallas or wherever had some issue with petrol refining caused months of issues and shortages. And that was a minor disruption which was quickly resolved. If the entire infrastructure of key sites were gone, it’d collapse civilisation for decades if not longer.
Covid showed how anaemic and poor we have become as lean and corporate ideologies of greed destroyed and hollowed out all redundancy and created fragile global supply chains which would cease to function if all or most major ports were knocked out. Even now a small hiccup in supply chains has seen cargo ships lined up for weeks and months unable to deliver their goods due to issues with ports, trucks, and warehouses.
In Christchurch in New Zealand from the 2011 earthquake they were literally unable to rebuild by themselves and many key infrastructure jobs were impossible to compete with local knowledge as no one has built a sewage system from near scratch or electrical grid wholesale in decades. They had to source talent from all over the world in order to try and even over decade later they are not nearly back to where they were. That’s ‘only’ an earthquake and they didn’t have to deal with the other aspects of a MAD catastrophe. Many of the old churches and such were simply impossible to replace with skills to make them being over 100 or 150 years out of date and only a handful of people around the world who could repair them, but couldn’t necessarily build them.
Looking at how bad regional disasters have been and how incredibly reliant on National and intentional responses they were to rebuild themselves over 10 and 20 year periods….if a nation were crippled and unable to draw resources and know how from across an empire to restore a single city, but had to deal with dozens of major cities being knocked out…life would collapse to a very simplified existence for decades and decades with non impacted regions becoming the new global powers. It could take 100 or 200 years to recover…with peaks and valleys of new technology, ie mobile phones and mules for farmers so that 2100 would be both better and worse off than we were in 1800 across various dimensions.
This ‘not so bad’ crap for a MAD scenario is naive to the extreme and thinks of only one city at a time or ignores the compounding effect of so many systems being lost at once where people in cities will start murdering each other for food within a week of the truck deliveries stopping.
Even looking at the Marshall plan and such in Europe from the USA, the UK had meat rationing into the 1960s from a war that ended in 1945…and nominally the UK was on the winning side! Over 20 years to restore just cattle and meat production and distribution! How or why would we ever think something orders of magnitude worse, such as a doomsday MAD scenario, would not be as bad? It’d be much worse and we are much more fragile and reliant on brittle paper thin supply chains now vs how the UK was back then when people were closer to the land in terms of knowledge of food growing and non electric preservation techniques. Would we be better or worse in growing victory gardens today vs back then? Worse I’d say, and we would be trying to feed a lot more people.
The message must be that we never ever have a MAD scenario play out. Softening that vital message by simplistically looking at blast radius zones is utter nonsense.
#11: The willingness to expand 'Romanness' to conquered people as the main reason for Roman success is a major proposition in Mary Beard's work, and is the thesis of her popular-audience book 'SPQR'. I was a little surprised that Grunewald didn't reference it, but I enjoyed his take and it is convincing to me.
#30: I saw this twitter manifesto at Zvi's back in February, and made a similar comment there, but: the statement that 'there are no constitutional rights in substance without freedom to transact' is impossible to divorce, in my mind, from 'there are no constitutional rights in substance without having means to transact with.'
I accept that there is a difference between being prevented by a government from using existing resources that are in some sense 'your own' and not having those resources in the first place. But if a transaction must take place for a right to exist in substance, then both the means and the freedom to use them must be present. Else we are conceding that constitutional rights are contingent on wealth.
27 - the reddit post is from October 2021. The author has an updated (as of January of this year) version at his website https://reasonwithoutrestraint.com/race-and-iq/ - many other interesting analyses of 'controversial' topics as well.
I've already had a panic attack eating an Impossible Cheeseburger, "how did I get here this isn't kosher", even though of course it is. It wasn't the first bite, but it was when I was distracted a few bites in talking to a friend.
So I have very high confidence that fake meat will pass the "turing test". MOOring Test?
4. The Vox article on Georgism seems much worse than the writings about Georgism I've seen here. Or maybe I don't fully understand the argument.
This article suggests that there is a close relationship between the land value tax and overly restrictive zoning laws. What is this relationship?
Parts of the article pretend that zoning laws don't exist. "And if someone turns their garage into an apartment, providing an affordable housing option for their community, they pay higher property taxes than similarly situated neighbors who don’t add housing options to their land." The reason this doesn't happen more often isn't because of the taxes. It's because it's illegal.
Parts of the article claim that a land value tax will cause overly restrictive zoning laws to disappear. "It turns NIMBYs into YIMBYs." I want to know how this happens. If I don't want a big apartment building going up next door because it's too noisy, how would reducing their taxes change my opinion? If anything, I would expect that a land value tax makes NIMBYs worse: the new development next door could increase my land value, and so my taxes.
I do think that a land value tax is a useful tool to reduce vacancies & make the tax system more fair. I also think that zoning reform should be implemented in lots of places. But these aren't the same problem. The land value tax seems most useful in places with many vacant lots and urban blight, while zoning reform seems most useful in places with no vacant lots and high rents.
Why has this idea become popular recently? Probably because it's a pretty good idea and people are putting forward good arguments for it. Not because "we've run out of suburbs." We haven't. Before the pandemic, suburbs were growing faster than urban or rural areas [1], and I expect that trend increased with more people working remotely.
[1] https://www.pewresearch.org/social-trends/2018/05/22/demographic-and-economic-trends-in-urban-suburban-and-rural-communities/
To suggest an answer to Naval Gazing's question, I think some of the Western appreciation for Russian equipment is that the designs are often good on paper and well fit to novice users. "Deep Battle", the Soviet WWII doctrine of combined arms, can reportedly be implemented by wargamers with modest skill and a few minutes of prep time. It works even with the minimal coordination you can usually achieve in video games. Soviet strategies are brutally straightforward, and their tanks are designed to match. The fact that the vehicles are often manufactured, maintained, and employed well below the design is often ignored. That they have bad ergonomics and situational awareness is not necessarily understood or experienced by enthusiasts or gamers. By contrast, Western strategies are a lot more complex and subtle. The tanks have tradeoffs that are exploitable if not used by professionals, or when used outside a Western combined arms milieu.
But anti-missile systems are definitely an essential technology for every serious military going forward. A well-supported, well-equipped MBT is scarily survivable, even if poorly deployed and defended tanks get got. As a case study, Israel lost *zero* main tanks or modern APCs to Hamas in Gaza in 2014, despite facing the most advanced Russian missiles and RPGs available at the time. These APS systems are even scarier if those systems can not only shoot down the incoming projectile but simultaneously cue the tank where it was shot from; you might not live long enough to see your ATGM defeated harmlessly in front of the tank.
The missile cruiser Moskva has been confirmed sunk by Russia on April 14 [1]. Ukraine claims they shot it with two cruise missiles on April 13 [2]. On the same day, April 13, Ukraine released the stamp commemorating the small force on Snake Island telling the Moskva to go fuck itself [3].
Coincidence? I. THINK. NOT.
I'm not in general a conspiracy theorist. The only other fringe conspiracy theory I believe is that "smart lights" only exist because lightbulb companies haven't figured out how to make a mediocre LED. But the timing is too funny to not believe.
P.S. there is truly bonkers rumor that there was Christian relic onboard the Moskva [4]. If 2022 was pitched as the next season of Servant of the People, nobody would have bought it.
[1] https://www.bbc.com/news/world-europe-61114843
[2] https://www.washingtonpost.com/world/2022/04/14/ukraine-russian-missile-cruiser-moskva-warship/; paywall
[3] https://www.jpost.com/international/article-704096
[4] https://twitter.com/Popehat/status/1514763143892791297 (epistemic status: literally a rumor from Twitter)
When I first saw "There are no constitutional rights in substance without freedom to transact" idea, I thought that it's interesting but obviously proves too much. For instance, that there are no constitutional rights without post scarcity communist society.
If we assume that constitutional rights require the ability to spend money, we end up with a situation where some people have more money, therefore more constitutional rights than other people, therefore making these rights not rights but privileges.
Of course the satire in 9 cuts both ways: intentionally or not, it puts forward a case that either Jesus is a composite figure, or the Gospel accounts are based on several oral sources that got garbled in transmission, or, if neither is the case, Jesus was as inconsistent, opportunistic and/or psychologically unstable as Trump.
Today is Good Friday and also Passover, so to those who commemorate these observances, the blessings of the holy seasons to you.
To those who don't, enjoy your chocolate Easter eggs on Sunday and your Satanic hot cross buns!
Satanic, you ask? Which leads into the seasonal video episode of History For Atheists:
https://www.youtube.com/watch?v=YJq70tf0AsY
That's the preaching done! 😀
Re: 20, comments in that link mention the fact that English is the only major language that doesn't have a governing body. I think this is under-reported!
I'm pretty sure English is the only officially-democratic language with more then a million speakers. That's based on my own original research though, I haven't seen anyone else talk about it. I feel like it captures something about the libertarian anglophone philosophy, and I would love to learn more about how this came to be the case. Did no one ever try? When did most linguisitc-authorities come into existence? Has it had any measurable effects on English development? Etc.
https://www.vanityfair.com/news/2022/03/the-virus-hunting-nonprofit-at-the-center-of-the-lab-leak-controversy
(Published March 31, likely to have slipped through)
14. is misleading. This is specifically in reference to brain-wide association studies (BWAS), e.g., "let's try to correlate, uh, which Pokémon starter people prefer to any possible set of structural and functional indices." These studies work off of questionable hypotheses.
Neuroimaging studies with well-argued hypotheses (i.e., NOT grounded in BWAS and/or generically bad science) and relatively modest sample sizes have a lot to offer.
The Banach story reminds me of the protagonist of Le Guins The Dispossessed. He's a leading expert in theoretical physics, and when he visits the sister planet they insist on calling him "Dr." because a guy like that couldn't possibly not have a credentialed title!
I haven't gone and read the methodology yet, but the blogosphere map seems very out-of-date. It's got the old version of Pharyngula on there, which I don't think has been used for 5 years. And Worthwhile, which barely functions as a blog any more. It's a weird combination of the blogosphere now and what it looked like 5-10 years ago.
Maybe that's just because things have weird half-lives on the internet when they're not very active any more, but they're still available. But it makes the map a bit difficult to parse.
RE: the giant fly
It is categorized as "Insects discovered in 1833." What is the significance?
https://en.wikipedia.org/wiki/Category:Insects_described_in_1833
Musk: "To get a self-driving car, we need to solve real-world intelligence"
https://youtu.be/cdZZpaB2kDM
Isn't the fourth one a subset of the third? Trump may not be clinically insane (but then again he may be) - but that just drives home that the trilemma, as stated, is a false trichotomy; Jesus may have been a moral teacher who, like a lot of other young men, also had what we now call schizotypal traits. (It is difficult to see how someone could get into the prophet/messiah track, common at the time, and *not* have schizotypal traits.)
What a difficult job moderating is-- a dreamwidth moderator takes a look at twitter and thinks they're doing a difficult job pretty well.
https://twitter.com/rahaeli/status/1515061592416788481?t=8b4AhKiq27e0SDqE0A32Yw&s=09
What it takes to train a tank crew
https://threadreaderapp.com/thread/1513858825027305482.html
Russians might be trying to get by with 2 soldiers per tank. That would explain a lot.
https://twitter.com/The_Tech_Son/status/1513754333384527877
2: Cool graph, but it can't be complete. Fox is a tiny yellow dot. Where's the tiny purple dot for NYT? Do humans so rarely link to these old corporate media outlets from their blogs that ACX is 30 times bigger than fox in the link graph and NYT doesn't even show up? [Edit: Krugman is a red dot a quarter the size of ACX but the rest of NYT is absent] Does almost all their traffic come from deliberate subsidies by big tech instead of humans linking to them? It approximately seems that way. Prior to Youtube's public decision to subsidize old corporate media by bumping them waaaaay up in all search results, they were getting almost no traffic on youtube.
Categories seem to be
purple: rationalist blogs
red: left/center econ blogs
yellow: right econ blogs
orange: mostly libertarian blogs
yellow: righty blogs (single-issue free speech promoting blogs like thefire.org and volokh get miscategorized as yellow or orange)
purple: lefty blogs
green: wall street (mix of apolitical trading blogs and hyperpartisan blogs that mention wall street)
26: Why is the size-scaling limit for avians >1 OOM larger than the size-scaling limit for flying arthropods?
32: Of course people respond to incentives and fat shaming is an incentive. But you're fighting an uphill battle against hunger unless you categorically change which foods are available to eat. Some foods are vastly more satiating per calorie than others. Anything that spikes insulin is bad because it makes the brain less sensitive to leptin and causes more hunger on the rebound. Meats and vegetables are great. (I have succeeded in reducing my bodyfat percentage from 20% to 16% over the last two months using the rules described here: https://www.lesswrong.com/posts/PPeSqatTRsuFw3ug9/food-manufacturers-are-out-to-get-you)
Singapore probably could have got much more than that 33% reduction in childhood obesity without fat-shaming anyone if only they banned all caloric beverages other than milk.
The Nobel figure is missing a lot of international Nobels. There are at least half a dozen Australian born winners. And France only one?
Commemoration of another anniversary today - the Easter Rising of 1916.
We were asked in a previous post "Who Gets Self-Determination?" and this is the answer of the Irish revolutionaries back then:
Proclamation of the Irish Republic and independence (text)
https://www.gov.ie/en/publication/bfa965-proclamation-of-independence/#
Reading of the Proclamation outside the GPO today
https://www.youtube.com/watch?v=nvZwbK0E_i0&t=147s
https://www.rte.ie/news/ireland/2022/0417/1292796-easter/
#3 and #9 is great combo:
"New finding suggest the cult of Trump has much wider geographical extent than previously thought:
According to the new manuscript uncovered at archaeological site on the Pontic steppe, sometime in the second decade of 21 century the settlement of Novgorodskoye was renamed to New York after the mythological birthplace of Trump. The locals believed the renaming will help to earn the favour of the powerful deity and protect the settlement from the destruction by Ork Horde."
https://en.wikipedia.org/wiki/New_York_(Ukraine)
28: One of my favorite findings in cognitive science! The paper that started it all is Dehaene (2007). One of the more interesting examples - why are there such strong (complete?) overlap in hippocampal processing of episodic memory & egocentric navigation? How about semantic memory & allocentric (landmark) navigation? Arguably because of deep similarities between the two functions. Buzsaki & Tingley (2018) argue that HF performs sequence processing, leveraged across diverse applications.
Other examples of cultural recycling discussed in Dehaene (2007) include reading & face recognition, arithmetic vs eye movement, music vs speech. These are all construed as needing similar cost functions.
Marblestone et al (2016) argue that DL is moving towards a similar design pattern, which will move us much closer towards AGI.
> We hypothesize that (1) the brain optimizes cost functions, (2) the cost functions are diverse and differ across brain locations and over development, and (3) optimization operates within a pre-structured architecture matched to the computational problems posed by behavior. In support of these hypotheses, we argue that a range of implementations of credit assignment through multiple layers of neurons are compatible with our current knowledge of neural circuitry, and that the brain's specialized systems can be interpreted as enabling efficient optimization for specific problem classes. Such a heterogeneously optimized system, enabled by a series of interacting cost functions, serves to make learning data-efficient and precisely targeted to the needs of the organism. We suggest directions by which neuroscience could seek to refine and test these hypotheses.
Buzsaki & Tingley (2018). Space and Time: The Hippocampus as a Sequence Generator
Dehaene (2007). Cultural recycling of cortical maps
Marblestone et al (2016). Toward an Integration of Deep Learning and Neuroscience