It would really make it a lot more enjoyable to vote on the book reviews if we could vote just by clicking the heart on the review-- at the time we are reading it. Then we could vote on based on our immediate reaction to the review (eg that's a great review!!! vs meh). You'd lose the rank order preference of individuals. But then people could give positive feedback on multiple reveiws if they liked them all a lot. Top winner would be the review the most people liked a lot.
Any tips on the best DNA sequencing services? What do they tend to provide? Do you get your actual genome or just some handwavy stuff like "some of your ancestors were from northern Europe" and "you're more likely to be a gambler"? What percentage of your genome do they sequence?
I tried to read through some FAQs but they tend to feature pretty pictures and quite limited info on what exactly you receive.
"When fasting insulin and the natural logarithm of c-reactive protein were included in the model, an inverse association between BMI and mortality was present"
To carry on the discussion down thread about "Christians needed to invent hospitals because they invented so many wars", here's something I had not previously heard of before: Christians burned down the Library of Antioch!
Or maybe not. But if you want to one-up someone talking about the Tragedy of the Burning of the Great Library of Alexandria, here you go! "Never mind Alexandria, did you know about Antioch?"
Hi Rats, Prats, and -- I can't think of a third list item, drat: If anyone is following the Trump Mar-a-Lago thing, I came across this article -- unsigned but seems to be by an attorney. It goes fairly in-depth about the background and claims presented, with links so you can verify the materials yourself: https://files.catbox.moe/2erd4t.pdf
I am glad to announce the third of a continuing series of Orange County ACX/LW meetups. Meeting this Saturday and most Saturdays. The first meeting was great, and I hope to see many of you at this one.
A) Two conversation starter topics this week will be. (readings at the end)
1) What is open-mindedness
2) Psychedelics.
B) We will also have the card game Predictably Irrational and frisbees. Feel free to bring your own favorite games or distractions. This is a pet-friendly park and meeting.
C) There will be opportunities to go for a walk and talk about an hour after the meeting starts and use some gas barbeques if anyone wants to grill something. There are two easy-access mini-malls nearby with takeout hot food available. Search for Gelson's or Pavilions in the zipcode 92660.
D) Share a surprise! Tell the group about something that happened that was unexpected or changed the way you look at the universe.
E) Make a prediction and give a probability and end condition.
F) Contribute ideas to the future direction of the group. Topics, types of meetings, activities, etc.
Conversation Starter Readings:
Suggested readings for this week are these summaries. These readings are optional, but if you do them, think about what you find interesting, surprising, useful, questionable, vexing, or exciting.
1) Openmindedness.
This week we will try a classic video from the Skeptic/Atheist movement. Questions to think about and discuss? Is this a good description of the reality of open-mindedness? Did it change how you thought about open-mindedness? What do you think are the essential elements of open-mindedness to rationality? Is skepticism necessary for openmindedness and when does it work against it?
Open-mindedness
And/or This SSC essay. “The Control group is out of control”. What are the upsides and downsides of calling paranormal studies the control group for the scientific method. Does this increase our ability to be open to correct ideas, and is it worth shutting the door on exhausted lines of investigation? Why do you think belief in psychic powers affects the results of apparently rigorously replicated experiments? What do you make of the results of the smart rat, dumb rat experiment? Did you realize that double-blind is often thrown around as a claim when the second blinding is poorly done or not done at all and what is the importance of blinding to open-mindedness?
Written The Control Group Is Out Of Control | Slate Star Codex
Audio The Control Group Is Out of Control [Classic]
2) For psychedelics:
We will dip back into some of the best descriptive research done in the 1960s on the phenomenology of psychedelics. Read chapter 1 of “The varieties of psychedelic experience.”This book has many good digests of different people's reactions and experiences with psychedelics. What types of experience most interest you? What did you not know about? What potential applications come to mind after reading these experiences? Do you agree with the taxonomy that the authors create?
the varieties of psychedelic experience.pdf
This is an interview with an experienced PTSD researcher that has been involved with the cutting-edge of PTSD treatment research about MDMA therapy. I generally like what he has to say and find his perspectives useful. I will put one caveat, which is he gets really enthusiastic about a lot of things. He is a generally more optimistic person than me.
Bessel van der Kolk on MDMA assisted therapy for PTSD: More profound than anything we have done
Finally, here are a couple of short videos by a guy that has used a lot of different drugs and ruined his life with them. He at first enthusiastically endorsed MDMA, but it did not stop him from ruining his life with other drugs or properly addressing the deep psychological issues he had to face. Psychedelics are often not enough and can even be a distraction or copium and have their own abuse potential.
If you want a more general introduction to psychedelics, here is a book summary of the recent popular review of psychedelics. “ How to change your mind” by Michael Pollen
The review of WWOTF mentioned that perhaps we should leave some coal lying around in case humanity has to reinvent the industrial revolution generations after some almost-extinction-level event.
The book review contest was fun! I had hoped that a weekly cycle would make me read more and more diverse book reviews, and this is what happened. Great readings and inspirations. Also, while I've always enjoyed Scott's book reviews, I admired his skills even more oftentimes during the contest - it seems so easy to make this or that 'mistake' and annoy the potential reader even in a very good review (and I certainly wouldn't have done better). Great writing is really an art.
I'd be interested to hear about your criteria for selecting your favourite.
This is bad and unacceptable. Telling people to "pick one" will just hand victory to the weirdest instead of the best. I thought Scott of all people would understand this.
When did the existential risk posed by rogue AI come to be perceived as an urgent matter? The first Terminator film put the issue before a large public in 1984, but as far as I know it didn't spark much interest in the real possibility of AI posing a threat to humanity. Yudkowsky and Bostrom were writing about the issue at the turn of the millennium, but how many were taking them up on it at that time? What about in the wake of Bostrom's Superintelligence?
Missed opportunity: having prediction markets for the book contest winner. (Technically, one per review would be easiest, probably.)
Of course, then one would also have to come up with a way to prevent ballot stuffing, but I think this would be desirable in any case. Apart from people with botnets, it might also that some other interest group (e.g. subreddit) focused on the topic of one review (say, fusion energy, ancient china, weird poetry) first lined the book review when it came out (not objectionable), and now links the vote (probably objectionable). Some clear rules, e.g. "You should have made a good faith effort to read at least 2/3 of the contest entries" would probably deter the more honest external readers.
I don't have a good proposition how to do avoid ballot stuffing: restricting voting to the active commentariat (a la wikipedia "only accounts with at least N edits can vote") seems unfair to lurkers. One rather well defined group would be the premium members of ACX. Of course, this is also introduces biases towards the very committed and/or well-of readers, but if the membership of that population was also saved with the vote, it would allow checking for some irregularities. Just having the voting site print out a random string and asking members to post that string as a comment on a special hidden thread would serve in a pitch without any involvement of substack. (And incidentally also break anonymity, but for the book contest that might be acceptable.)
Or you could have the voters first select five reviews they have actually read, then quiz them on the content.
The first replies I've seen disappear like that were somewhat rude (I like to stir shit up, what can I say) so I assumed that was God-Emperor Scott's heavy hand, but the one linked was entirely wholesome and civil.
Ok here's another approach to if not disproving then at least neatly sidestepping the repugnant conclusion. This seems sufficiently obvious that someone must have thought of it already, but to my great annoyance I haven't seen any mention of it on this blog.
The repugnant conclusion is really a constrained optimization problem where you have some fixed amount of resources R and want to find the optimal number of people N and resource allocation function f : [1,N] -> Real assigning an amount of resources to each person in a away that maximizes the total utility over all people while not exceeding the available amount of resources. As I've seen it described, the repugnant conclusion assumes that it is very cheap to add new people with infinitesimal utility. While I don't necessarily disagree with the repugnant conclusion under those assumptions, I don't believe these assumptions hold true in our universe.
Here's another set of (also unrealistic but somewhat better) assumptions: It costs 5 resources to sustain a person. At this level of resources, the person is barely kept alive and suffering greatly. If we increase the number of resources to 10, the person is now at utility 0 and indifferent about being alive or not. From that point on, every additional resource linearly increases it's utility up to a maximum of 10. So now say we have R=1000 resources available. What the conventional conclusion to the repugnant conclusion seems to tell us is that we should maximize the number of people and have 1000 / 5 = 200 people who are all maximally suffering. Far from maximizing utility, this is actually the minimum utility we can achieve. Ok, so say we instead have 100 people with 10 resources each. Still not great, this just gives us 0 utility. If you go through the math, you'll find that under these assumptions, the optimal resource allocation is to give 20 resources each to 50 people, all maximally happy with a total utility of 500.
Now, of course, this model is also flawed. Obviously the numbers are made up and utility should probably have diminishing returns rather than increasing linearly up to the maximum. So with those adjustments, I don't think you would get only maximally happy people, the optimal point would be somewhat less than that (but still well above baseline).
Another aspect to consider is that a large fraction of utility may be derived from common goods that have very low marginal costs for each additional person, which gives a much more favorable tradeoff for adding additional people in a way that does not have to significantly diminish the utility of others. In a hypothetical society of the future that has been optimized for maximum utility, we will have a ginormous numbers of people with not that much resources each, but also mindblowingly good movies and video games, fascinating scientific insights, incredibly advanced medical technology available to everyone, superintelligent AIs that can have deep conversations with you and help you achieve enlightenment, galaxy-scale engineering projects that bring glory to humankind, etc. Doesn't sound so repugnant to me!
What we use nowadays is a line drawn around the city, either reflecting the extent of the settlement, or what its extent was 20/100/800 years ago (The City of St David's, population 1600) or political squabbles over local government fiefdoms (Vegas is not in Vegas). Everything inside the line is counted and everything beyond the pale is not.
This mostly shows up in absolute population counts and densities. A city where the line is drawn tightly around it will have a lower official population and higher density than one which includes a bunch of peripheral farmland and commuter towns.
But there can also be other distortions. The nice bits of the Nottingham urban area are located outside the city proper, even when they are contigious with it. This makes the city look poorer and more deprived than it is.
I'm no geographer but my guess would be to use more fine-grained statistics to create a model of how cities change as distance from the centre increases, and then use the parameters needed to fit a particular city to those curves as a metric of what the city is like. A very cursory google reveals this non-paywalled paper: https://www.researchgate.net/publication/316428846_Scaling_evidence_of_the_homothetic_nature_of_cities as the kind of thing I'd want to build upon.
I'm surprised the Canberra one merits a mention and Melbourne doesn't - is there really such an overconcentration of Rats there that it outweighs the fact Melbourne is >10x Canberra's population?
Disappointed by the lack of a fancy voting system. In my opinion there were four approximately equal reviews. Seems a shame to not be able to distinguish between my second favourite (which I thought was excellent) and my least favourite (which I couldn't even finish).
It also incentivises me to vote tactically which I don't really want to do.
I am a Russian with residency in Turkey, I'm an employee of an American company. I am considering whether to ask them to pay my salary to a Wise USD account (which means using some American bank's account details). When I ask the financial guy in my company, he told me that there's approximately a 5% chance that at some point IRS will ask me who I am and ask me to prove that I don't owe the US any taxes (I don't because I'm not in the US and am not a citizen of the US) and that this would be very difficult and problematic for me. Is that true? Why would it be difficult and problematic?
How do I get Substack to stop sending me emails but NOT stop showing me newly published articles from people I subscribe to when I go to their webpage or use the app?
The whole thing is a mess. There's some sort of difference between Substacks that I pay for and those that I don't, and some difference between the webpage and the app. The result is that my email box is filled up with fucking dozens of substack emails every day. I can't even stay ahead of them to delete them. But when I unclick "send me email" it ALSO stops updating the app. If that makes sense.
I stumbled across the life of William U’Ren recently. I bet Oregonians know of him, but not being from Oregon I had never heard of him, although I was familiar with some of the reforms he promoted, e.g. direct election of senators. I think a few others here might find him intriguing because of his efforts at voting reform and because he was a Georgist.
How did I come across him? I was researching cherry trees for my backyard. Apparently William U’Ren once had an asthma attack while down on his luck, and was taken in by Seth Luelling, the politically progressive horticulturalist who developed the Bing cherry. Seth and his brother Henderson Luelling would rank high on the ACX “Puritan scale”, although they were in fact Quakers. After attempting to run a utopian community elsewhere, they founded the first commercial fruit tree nursery in the Pacific Northwest. The Luellings were staunch abolitionists and might have maintained a stop on the Underground Railroad, though this can’t be proven.
One parent variety of the Bing cherry is the “Black Republican” cherry, which Luelling named in support of Abraham Lincoln. A “Black Republican” back then was one who supported emancipation.
But the famous Bing cherry we see in supermarkets was named for Ah Bing, Seth Luelling’s Chinese orchard foreman. After 35 years working for Luelling, Bing went back to China for a visit in the 1880s and was prohibited to return to the US under the Chinese Exclusion Act.
If anyone finds it helpful, I've made a tier list for the book reviews where I've been keeping track of which ones I've liked over the course of the contest.
How do proponents of converting the US to multimember districts think this is going to, like, work? Globally the average size for multimember districts is 3-7 representatives- right now the US House has 435 members. At 1305 members it would be either the world's largest lower house, or certainly among them- at 3045 members.... we are entering 'patently ridiculous' territory. To be fair, you could still probably keep 1-2 reps for smaller states (Wyoming, etc.) and just use multimember for the bigger ones, but there's absolutely no way that you end up with less 1300-1400 members of the House. I don't see the American people accepting this.
The right way to do multimember, if the US were starting from scratch, would be to simply use larger districts than we have now. I.e. Illinois just to pick a random example has 18 House members right now, this would be reduced to say 6. But that seems politically impossible to force on the country- people like having local representatives for their little fiefdom, they don't want to share representation with 'those people on the other side of the state, ugh'. Voters would be apoplectic if you shrunk the number of state districts.
So, multimember proponents- you're proposing 1500 or so House reps, is that it? How would this work?
In my high school health class, I learned that drinking before your brain is fully developed is bad. I recently learned that the brain matures fully at around 25. Am I losing those sweet, sweet IQ points if I drink before turning 25 (I'm 18)? Or is the effect negligible at my age? A few searches brought me nothing but the usual "drinking bad" and many mentions of alcohol abuse disorder and drunk driving accidents, while I'm more interested in the concrete long-term effects.
Loved reading the reviews - great bang for the time invested. Lately I've started to suspect books are a bit... fat. Padded. Could be that they always were so, and long form blogging just spoiled me? Or I should just find better books.
Also loved the unofficial tradition of adding a personal insight to the reviews. Regardless of whether you agree or not with each particular idea, I just like the personal touch and the guts.
Many years ago, I read about cultures where people routinely changed their names over the course of their lifespan, sometimes several times. Not the vestigial forms we had - nicknames, and, for women (at the time) married names, but something a lot more drastic. And in some cases, the names told a story about the person's life. This seemed utterly weird to me.
But now I realize that this pretty much describes my history on line. DinoNerd is relatively new, and would not have been suitable until I'd been in tech long enough to count as a "dinosaur". I've had at least 3 other nicks I've used all over the net; those seem to have lasted about a decade each. Then there are the single-use nicks, and the few sites - mostly long ago - where I used whatever email ID I happened to have, generally a variant on my legal name. And this feels natural to me, complete with a sense that I'm no longer the person who used the earlier nick, and it's weird to login somewhere I've been long enough to still have it.
I'm guessing many others have similar experiences, and wondering what thoughts people have about the cultural and psychological effects, based on their personal experiences.
> [...] research may — inadvertently — stigmatize individuals or human groups. It may be discriminatory, racist, sexist, ableist or homophobic. It may provide justification for undermining the human rights of specific groups,
> Race and ethnicity are sociopolitical constructs.
>In some cases, however, potential harms to the populations studied may outweigh the benefit of publication
I wonder if now that we realized doctors were giving our opioids like candy, there will be a major movement to change the way people - especially kids - are medicated. I wonder what Scott thinks about whether this is a problem and if it’s gotten worse over time.
Taxpayers contribute to college education, and in the next California budget around 13% is earmarked for higher ed. I guess we assume that college grads contribute more to society, so it is worthwhile for all of us to pitch in. But the student loan crisis is a free market indicator that way to many people signed up for the college dream. What metrics should America use in deciding whom and how many should be sent to university? I have this vague idea that taxpayers should only fund excellent students. That seems more fair.
I myself don't think that any of the alternative voting schemes are better than FPTP. In particular, the complexity of alternatives raises doubt about the integrity of the process that isn't there in a properly regulated simple polling system.
On edit: I am strongly biased against these alternative systems because every instance that I have seen them used/proposed (Hugo Awards, post-Trump voting) it has been to keep the "weirdos and wreckers" on my side out. (If there are other uses, I would appreciate hearing about them - for instance, did anyone propose using ranked alternative to keep AOC from being elected?)
I feel like the book review entries are overall weaker than last year- is that just me or do people share that impression? Off the top of my head the Galen or Henry George reviews both feel like I'd pick them over any of the ones in this round.
You have 15 candidates, of which exactly one is a work of fiction, and you're using FPTP. Are you trying to split the vote and hand victory to God Emperor of Dune? And, in a world in which God Emperor of Dune actually is the most popular (as would be measured by a reasonable voting method), how could its victory seem legitimate when it could so easily be explained by vote splitting?
What should you use? I recommend STAR Voting (https://www.starvoting.us/star); it's simple, expressive, doesn't make voters agonize over which of two similar options is slightly better, and there's an easy tool for making online STAR elections (https://star.vote/). Other commenters have mentioned Approval Voting and Minimax as other options, and both of them are reasonable; Approval is good if you want the system to be as simple as possible (though voters need to consider strategy more than they do under more complicated methods), and Minimax, like all Condorcet methods, delivers excellent results (though I prefer the scoring ballot to a ranked ballot here so that I'm not focusing on minor differences).
What is not reasonable is using FPTP. It ignores the majority of voters' preferences. It makes being different from other candidates more important than being better than other candidates. It fails to give authors a good sense of how well-received their book review was since it doesn't distinguish between broadly-popular reviews that excite few people and polarizing reviews that excite a few people but which most readers aren't interested in. And, of course, the real problem (https://forum.effectivealtruism.org/posts/QCHLkgudfgbovgoan/cause-exploration-prizes-voting-methods) is with real-world elections, not online polls - but online polls like this one are excellent opportunities to bolster the profile of little-known but high-quality voting methods, and the data can be legitimately useful in understanding voter behavior. Using FPTP here doesn't just risk unrepresentative outcomes - it's a missed opportunity for building awareness and generating data.
There's been a lot of discussion here about EA and weird seagull-eye-pecking causes but, in terms of currently-existing-human type charities, what do we, the ACX hive mind, support? The GiveWell recommended list? (Malaria consortium, against malaria foundation, Helen Keller int'l, and New Incentives)
Bonus question: What are your thoughts on the recent removal from the recommended list of GiveDirectly and the deworming charities?
Ever wondered how Australia ended up with its voting system (known in Australia as "preferential voting")? It was caused by the equivalent of a GOP split.
Wikipedia: "The preferential system was introduced for federal elections in 1918, in response to the rise of the Country Party, a party representing small farmers. The Country Party split the anti-Labor vote in conservative country areas, allowing Labor candidates to win on a minority vote. The conservative federal government of Billy Hughes introduced preferential voting as a means of allowing competition between the two conservative parties without putting seats at risk."
Dr. Haidt is a thoughtful psychologist and has written some great books, but The Righteous Mind was pretty nonsensical, in my view. So, I liked your excellent review, that happened to agree with my opinion!
This review gets my vote. Is there a specific place I go to vote?
I'm a bit surprised this is a plain first-past-the-post vote, given all the discussion we've had about alternatives to current voting systems to select our Elected Representatives. At least we'll get a clear winner this time (unless the vote gets split so there are three 'first places').
I actually would have read another ten reviews, these slipped by incredibly easily and quickly. So if Scott feels that he needs some 'filler' content in future, posting another contest entrant that didn't make the shortlist would be fine by me.
We go deep into the weeds of their polygenic prediction algorithms, what kind of consumer adoption and regulation they expect, why the genome is relatively easy to improve, why natural selection hasn't already optimized these outcomes, differences in ancestral populations, and much more.
In the comment thread for the Exhaustion review, someone going by Fred mentioned that cranial traction sometimes helps with CFS. I did a little research, and it seemed safe and easy enough to try, so I did… and, within 24 hours, my fatigue level had gone from "debilitating and persistent" to something like normal.
Background: I had some good days, but mostly bad days. On a bad day, things as simple as a doctor appointment or a grocery trip would leave me exhausted for days afterward. Riding in a car could be a painful experience, sometimes requiring concentration just to keep from crying out in pain at every little bump. Walking my dog to the park and back was a major expedition, and I generally returned home staggering with fatigue… or, simply collapsed on the way home and had to be picked up. When I was lucky, post-walk fatigue would only lay me out for the rest of the day.
Working a single hour in a day was a major achievement, and I flaked out on my students almost as often as not. Stimulants, like adderall and modafinil, tended to make my fatigue *worse*, since they made me much more likely to overdo things. Caffeine was a devil's bargain. Best of all: the good days were becoming fewer and fewer, and the bad days worse and worse. (All five times I collapsed while walking the dog were in the last year.)
To my great surprise, self-traction had immediate and dramatic results. All I did was grip my skull and lift upward under my own power for 5-10 seconds… and it led to an incredible feeling of relief. I have tried to find words for the sensation I experienced, but nothing fits well. The best description I can manage is a strained metaphor: As long as I can remember, certainly all of my adult life, I have been buried up to my eyes in quicksand; during each self-traction "session", I felt like I was being actively lifted another 20-30cm clear. Bear in mind that an entire "session" of manual cranial traction is less than ten seconds long.
Eightish "sessions" and less than 24 hours later, I took my dog on her vet-recommended 3km walk. "Night and day" barely scratches the surface. Walking was easy, natural, automatic. Holding myself up didn't take constant effort. Keeping up with the dog was easy. About a kilometer in, I felt like breaking into a sprint—so I did. I ran for nearly a kilometer before slowing to catch my breath. Something that hadn't happened to me in decades happened—I *did* catch my breath. After resting for a minute or two, I felt up to running again. The dog vetoed this. (I realized afterward that her unwillingness to exert was probably because it was 37°C outside.)
Days later, I saw my physical therapist, and I asked her about this. She confirmed that my particular brand of chronic fatigue can, indeed, be caused by neck "instability", and (among other things) recommended an over-the-door traction device. My primary care provider agreed with her on all counts. So, on the advice of two capital-D Doctors, I got one of those devices. Now I use it, following my PT's directions, about once per two days. This is less portable, but much more repeatable and reliable than manual traction, especially on myself.
It has now been three weeks. I have been comparatively free of fatigue. I have had occasional bouts; when I notice what's happening, I try some quick self-traction or a brief stint in the traction harness. This often improves it, and earns a lesser version of that same initial feeling of abject relief. Whether or not they respond to traction, these bouts no longer come anywhere near what used to be my normal "baseline bad day".
Grocery stores are a breeze. Housework is manageable. I've only missed one tutoring session, and in fact I have reinstated a lapsed student *and* scheduled more weekly sessions, all while *also* managing to "work" 4-8 hours without having to spend days recovering. Not only am I walking the dog regularly again, she's now getting between five and twenty kilometers of walking *every day*. In fact, sometimes I achieve the holy grail of dog walking: the dog *wants* to go home and end the walk. (And not because of 37°C temperatures, either. I learned my lesson.)
There's still the matter of my mental health, which is a monumental problem of its own, and the genetically-inevitable joint problems that are flaring up in response to my skyrocketing activity level, and of course the habits of inactivity that must individually be broken… but, for the first time in a very long time, I feel like there's a way forward for me. No longer a mentally ill robot in a crumbling, useless body, I'm now a mentally ill robot in a mildly out-of-shape body. As I devour the lowest-hanging fruits of fitness, my overall physical health improves noticeably every day. And it's all because Professor Anna Schaffner wrote a book about chronic exhaustion, some yet-anonymous ACX reader wrote a review of it, Scott posted it as a finalist, and, deep in the comment thread, somebody going by Fred offhandedly mentioned that cranial traction sometimes helps.
Coleman Hughes interviews Rafael Mangual-- 2 hours, but fairly effiecient
Mangual's premise is that violent crime does serious damage which goes well beyond the injury to the immediate victims (well sourced), and that this is getting ignored by people from safe neighborhoods who focus on abuses by the police and justice system.
It's a rough listen for me-- I haven't been saying BLM and ACAB (all cops are bastards) or advocating for defunding the police, but I've been focusing on the likes of Radley Balko (journalist who focuses on justice system atrocities) and The Innocence Project (gets falsely convicted people out of prison).
It seems plausible to me that many police are abusive and protected by the system *and* that police might, on the average, be doing useful work.
Once upon a time, an engineer built a novel AI. In order to deal with AI Safety, he air-gapped it. Some time passed, and the engineer noticed that the AI was trying to break through the air gap. To prevent this, he crippled the AI permanently.
If you follow this link, you can get to the original of this comment where there are working links to the actual comments from Richard Gadsden, Brinkwater, and jumpingjacksplash:
I love intellectual debates and Astral Codex Ten has some of the best. But even more important to me is feeling like I could possibly make a difference in our world. For that reason I need the debate to lead somewhere.
If I come to believe that the debates are just a sophisticated game where we get points for sounding intelligent and persuasive, but they have no significance in the real world… Well, that's actually a downer for me.
In the Model Monday thread of Aug 1, a lot of people had a great time whomping on the plan for Neon. And, indeed, who wants to live in a narrow dark manmade canyon?
But I saw a glimmer of something good in this proposed boondoggle. The fact is that our current approach to civilization (and way too many humans) is destroying nature, and the compact form of The Line would be good for nature.
I don't know your gut feelings about nature, but I sincerely advise you: destroying it is not a good plan. So when I see a proposal that claims to be good for nature, my ears prick up. That there is not a lot of living things in Saudi Arabia is irrelevant. (As is the fact that MBS is a monster.)
So I spent a good chunk of August trying to see if I could fix Neom. With invaluable help from
How do you prevent the voting from being gamed? E.g from people voting multiple times or from people voting to help the person they know (without actually having read the reviews)
I'm torn between the reviews for "The Dawn of Everything" and "The Righteous Mind". I felt both were competent take downs of overrated books, which is the kind of thing I find gratifying, but I'm leaning towards "The Righteous Mind" because I didn't find the reviewers' hypothesis about gossip traps very compelling in the "The Dawn of Everything", although it was interesting.
Dall-E and the like can do art in various styles based on verbal prompts. Any thoughts about when or if an AI can produce a style (visual or musical) that catches on for people?
If non-existant potential future people have moral mass, why don't non-existant potential past people? What about potential future people of the counterfactual potential past people?
(edit: and how will we count their book-review votes?)
It would really make it a lot more enjoyable to vote on the book reviews if we could vote just by clicking the heart on the review-- at the time we are reading it. Then we could vote on based on our immediate reaction to the review (eg that's a great review!!! vs meh). You'd lose the rank order preference of individuals. But then people could give positive feedback on multiple reveiws if they liked them all a lot. Top winner would be the review the most people liked a lot.
Well this was interesting. Bill Mahr argues with an actual physical “Straw Man” to make a point.
https://m.youtube.com/watch?v=lTFnj-9EY4M
Perhaps next episode he’ll illustrate the Motte and Bailey idea with a real castle.
Any tips on the best DNA sequencing services? What do they tend to provide? Do you get your actual genome or just some handwavy stuff like "some of your ancestors were from northern Europe" and "you're more likely to be a gambler"? What percentage of your genome do they sequence?
I tried to read through some FAQs but they tend to feature pretty pictures and quite limited info on what exactly you receive.
I loved almost all the book reviews, and they are all very well written. I thank all the authors!
I have ranked the reviews according to the number of times they contain the word "focus" and its derivatives:
0 Consciousness And The Brain
0 The Castrato
0 Viral (was COVID a lab leak?)
0 Kora In Hell (William Carlos Williams poetry)
1 The Dawn Of Everything (ancient hunter-gatherers)
1 The Internationalists (treaty to make war illegal)
1 1587: A Year Of No Significance (Ming China)
3 The Future Of Fusion Energy
3 The Outlier (biography of Jimmy Carter)
3 The Society Of The Spectacle
3 Exhaustion (chronic fatigue)
3 God Emperor Of Dune
6 The Anti-Politics Machine (how development aid goes wrong)
7 Making Nature (history of the scientific journal Nature)
8 The Righteous Mind
Special congrats to the first four reviewers for sparing us the f**us word!
Does anyone know what the graph looks like for:
X - axis: days since booster shot
Y - added (or subtracted if negative at any point?) immunity from booster shot
Trying to advise parents on optimal booster scheduling
After a while of homelessness, some people have trouble sleeping indoors.
https://www.metafilter.com/196388/The-Trauma-Of-Homelessness-Doesnt-End-Under-A-Roof#8289499
There's one person in the comments who spent a year camping, and then didn't want to sleep indoors.
The short version is that homelessness is an even harder problem than it looks like.
https://www.nature.com/articles/s41366-022-01211-2
"When fasting insulin and the natural logarithm of c-reactive protein were included in the model, an inverse association between BMI and mortality was present"
To carry on the discussion down thread about "Christians needed to invent hospitals because they invented so many wars", here's something I had not previously heard of before: Christians burned down the Library of Antioch!
Or maybe not. But if you want to one-up someone talking about the Tragedy of the Burning of the Great Library of Alexandria, here you go! "Never mind Alexandria, did you know about Antioch?"
https://historyforatheists.com/2022/08/burning-the-library-of-antioch/
Hi Rats, Prats, and -- I can't think of a third list item, drat: If anyone is following the Trump Mar-a-Lago thing, I came across this article -- unsigned but seems to be by an attorney. It goes fairly in-depth about the background and claims presented, with links so you can verify the materials yourself: https://files.catbox.moe/2erd4t.pdf
"Ergo, an effective altruist should want to replace our present inefficient governments with a startup regime—an accountable monarchy."
https://graymirror.substack.com/p/is-effective-altruism-effective
Discuss. /ducks
Hello folks!
I am glad to announce the third of a continuing series of Orange County ACX/LW meetups. Meeting this Saturday and most Saturdays. The first meeting was great, and I hope to see many of you at this one.
Saturday, 9/3/22, 2 pm
1900 Port Carlow Place, Newport Beach, 92660
The Picnic tables outside the community clubhouse
33.6173166789459, -117.85885652037152
https://goo.gl/maps/WmzxQhBM2vdpJvz39
Plus code 8554J48R+WFJ
Contact me, Michael, at michaelmichalchik+acxlw@gmail.com with questions or requests.
Activities (all activities are optional)
A) Two conversation starter topics this week will be. (readings at the end)
1) What is open-mindedness
2) Psychedelics.
B) We will also have the card game Predictably Irrational and frisbees. Feel free to bring your own favorite games or distractions. This is a pet-friendly park and meeting.
C) There will be opportunities to go for a walk and talk about an hour after the meeting starts and use some gas barbeques if anyone wants to grill something. There are two easy-access mini-malls nearby with takeout hot food available. Search for Gelson's or Pavilions in the zipcode 92660.
D) Share a surprise! Tell the group about something that happened that was unexpected or changed the way you look at the universe.
E) Make a prediction and give a probability and end condition.
F) Contribute ideas to the future direction of the group. Topics, types of meetings, activities, etc.
Conversation Starter Readings:
Suggested readings for this week are these summaries. These readings are optional, but if you do them, think about what you find interesting, surprising, useful, questionable, vexing, or exciting.
1) Openmindedness.
This week we will try a classic video from the Skeptic/Atheist movement. Questions to think about and discuss? Is this a good description of the reality of open-mindedness? Did it change how you thought about open-mindedness? What do you think are the essential elements of open-mindedness to rationality? Is skepticism necessary for openmindedness and when does it work against it?
Open-mindedness
And/or This SSC essay. “The Control group is out of control”. What are the upsides and downsides of calling paranormal studies the control group for the scientific method. Does this increase our ability to be open to correct ideas, and is it worth shutting the door on exhausted lines of investigation? Why do you think belief in psychic powers affects the results of apparently rigorously replicated experiments? What do you make of the results of the smart rat, dumb rat experiment? Did you realize that double-blind is often thrown around as a claim when the second blinding is poorly done or not done at all and what is the importance of blinding to open-mindedness?
Written The Control Group Is Out Of Control | Slate Star Codex
Audio The Control Group Is Out of Control [Classic]
2) For psychedelics:
We will dip back into some of the best descriptive research done in the 1960s on the phenomenology of psychedelics. Read chapter 1 of “The varieties of psychedelic experience.”This book has many good digests of different people's reactions and experiences with psychedelics. What types of experience most interest you? What did you not know about? What potential applications come to mind after reading these experiences? Do you agree with the taxonomy that the authors create?
the varieties of psychedelic experience.pdf
This is an interview with an experienced PTSD researcher that has been involved with the cutting-edge of PTSD treatment research about MDMA therapy. I generally like what he has to say and find his perspectives useful. I will put one caveat, which is he gets really enthusiastic about a lot of things. He is a generally more optimistic person than me.
Bessel van der Kolk on MDMA assisted therapy for PTSD: More profound than anything we have done
Finally, here are a couple of short videos by a guy that has used a lot of different drugs and ruined his life with them. He at first enthusiastically endorsed MDMA, but it did not stop him from ruining his life with other drugs or properly addressing the deep psychological issues he had to face. Psychedelics are often not enough and can even be a distraction or copium and have their own abuse potential.
What MDMA Feels Like
“I was wrong about psychedelics”
https://www.youtube.com/watch?v=QRpXtHR0o5o
If you want a more general introduction to psychedelics, here is a book summary of the recent popular review of psychedelics. “ How to change your mind” by Michael Pollen
https://www.hustleescape.com/book-summary-how-to-change-your-mind-by-michael-pollan/
Terra Ignota hives, could they be implemented in our world? Could ACX fans/rationalist/EA become a hive?
I voted for the book review that made me change the way I viewed the world the most.
The review of WWOTF mentioned that perhaps we should leave some coal lying around in case humanity has to reinvent the industrial revolution generations after some almost-extinction-level event.
Brett Deveraux has conveniently just published a post explaining all the other things except coal that you need to start an industrial revolution - for example, spinning jennies help a lot: https://acoup.blog/2022/08/26/collections-why-no-roman-industrial-revolution/
The book review contest was fun! I had hoped that a weekly cycle would make me read more and more diverse book reviews, and this is what happened. Great readings and inspirations. Also, while I've always enjoyed Scott's book reviews, I admired his skills even more oftentimes during the contest - it seems so easy to make this or that 'mistake' and annoy the potential reader even in a very good review (and I certainly wouldn't have done better). Great writing is really an art.
I'd be interested to hear about your criteria for selecting your favourite.
Pitching in a meta-vote for approval voting in future book contests.
I almost ended up not voting. It took me 10 minutes to settle on 1, when in 5 seconds I already had my top 3.
This is bad and unacceptable. Telling people to "pick one" will just hand victory to the weirdest instead of the best. I thought Scott of all people would understand this.
Who is with me? How can we make Scott reconsider?
We are having a meetup in Mexico City in the context of ACX Meetups Everywhere. Find more information in our LW page:
https://www.lesswrong.com/events/bejXvxGjQ7rYudF88/acx-cdmx-meetups-everywhere-1
When you travel overseas and need cash, do you go to a traditional currency exchange kiosk, or do you just use an ATM?
Have you ever compared the exchange rates?
When did the existential risk posed by rogue AI come to be perceived as an urgent matter? The first Terminator film put the issue before a large public in 1984, but as far as I know it didn't spark much interest in the real possibility of AI posing a threat to humanity. Yudkowsky and Bostrom were writing about the issue at the turn of the millennium, but how many were taking them up on it at that time? What about in the wake of Bostrom's Superintelligence?
Is there a Savannah Georgia meetup group? if not, would anyone be interested in one?
Missed opportunity: having prediction markets for the book contest winner. (Technically, one per review would be easiest, probably.)
Of course, then one would also have to come up with a way to prevent ballot stuffing, but I think this would be desirable in any case. Apart from people with botnets, it might also that some other interest group (e.g. subreddit) focused on the topic of one review (say, fusion energy, ancient china, weird poetry) first lined the book review when it came out (not objectionable), and now links the vote (probably objectionable). Some clear rules, e.g. "You should have made a good faith effort to read at least 2/3 of the contest entries" would probably deter the more honest external readers.
I don't have a good proposition how to do avoid ballot stuffing: restricting voting to the active commentariat (a la wikipedia "only accounts with at least N edits can vote") seems unfair to lurkers. One rather well defined group would be the premium members of ACX. Of course, this is also introduces biases towards the very committed and/or well-of readers, but if the membership of that population was also saved with the vote, it would allow checking for some irregularities. Just having the voting site print out a random string and asking members to post that string as a comment on a special hidden thread would serve in a pitch without any involvement of substack. (And incidentally also break anonymity, but for the book contest that might be acceptable.)
Or you could have the voters first select five reviews they have actually read, then quiz them on the content.
Also, voting systems.
What's up with disappearing comments?
I've been getting replies in the mail that are no longer showing on the thread, e.g. https://astralcodexten.substack.com/p/open-thread-239/comment/8692502 by Gunflint in reply to my comment https://astralcodexten.substack.com/p/open-thread-239/comment/8689688 .
The first replies I've seen disappear like that were somewhat rude (I like to stir shit up, what can I say) so I assumed that was God-Emperor Scott's heavy hand, but the one linked was entirely wholesome and civil.
I believe without evidence that we're getting AI from the race between malware and anti-malware.
It probably won't resemble human consciousness and won't be able to communicate well with us.
Risk of eating the world? Doesn't seem likely. Eat all the computer capacity? Maybe.
Ok here's another approach to if not disproving then at least neatly sidestepping the repugnant conclusion. This seems sufficiently obvious that someone must have thought of it already, but to my great annoyance I haven't seen any mention of it on this blog.
The repugnant conclusion is really a constrained optimization problem where you have some fixed amount of resources R and want to find the optimal number of people N and resource allocation function f : [1,N] -> Real assigning an amount of resources to each person in a away that maximizes the total utility over all people while not exceeding the available amount of resources. As I've seen it described, the repugnant conclusion assumes that it is very cheap to add new people with infinitesimal utility. While I don't necessarily disagree with the repugnant conclusion under those assumptions, I don't believe these assumptions hold true in our universe.
Here's another set of (also unrealistic but somewhat better) assumptions: It costs 5 resources to sustain a person. At this level of resources, the person is barely kept alive and suffering greatly. If we increase the number of resources to 10, the person is now at utility 0 and indifferent about being alive or not. From that point on, every additional resource linearly increases it's utility up to a maximum of 10. So now say we have R=1000 resources available. What the conventional conclusion to the repugnant conclusion seems to tell us is that we should maximize the number of people and have 1000 / 5 = 200 people who are all maximally suffering. Far from maximizing utility, this is actually the minimum utility we can achieve. Ok, so say we instead have 100 people with 10 resources each. Still not great, this just gives us 0 utility. If you go through the math, you'll find that under these assumptions, the optimal resource allocation is to give 20 resources each to 50 people, all maximally happy with a total utility of 500.
Now, of course, this model is also flawed. Obviously the numbers are made up and utility should probably have diminishing returns rather than increasing linearly up to the maximum. So with those adjustments, I don't think you would get only maximally happy people, the optimal point would be somewhat less than that (but still well above baseline).
Another aspect to consider is that a large fraction of utility may be derived from common goods that have very low marginal costs for each additional person, which gives a much more favorable tradeoff for adding additional people in a way that does not have to significantly diminish the utility of others. In a hypothetical society of the future that has been optimized for maximum utility, we will have a ginormous numbers of people with not that much resources each, but also mindblowingly good movies and video games, fascinating scientific insights, incredibly advanced medical technology available to everyone, superintelligent AIs that can have deep conversations with you and help you achieve enlightenment, galaxy-scale engineering projects that bring glory to humankind, etc. Doesn't sound so repugnant to me!
Is there a better way to do city statistics?
What we use nowadays is a line drawn around the city, either reflecting the extent of the settlement, or what its extent was 20/100/800 years ago (The City of St David's, population 1600) or political squabbles over local government fiefdoms (Vegas is not in Vegas). Everything inside the line is counted and everything beyond the pale is not.
This mostly shows up in absolute population counts and densities. A city where the line is drawn tightly around it will have a lower official population and higher density than one which includes a bunch of peripheral farmland and commuter towns.
But there can also be other distortions. The nice bits of the Nottingham urban area are located outside the city proper, even when they are contigious with it. This makes the city look poorer and more deprived than it is.
I'm no geographer but my guess would be to use more fine-grained statistics to create a model of how cities change as distance from the centre increases, and then use the parameters needed to fit a particular city to those curves as a metric of what the city is like. A very cursory google reveals this non-paywalled paper: https://www.researchgate.net/publication/316428846_Scaling_evidence_of_the_homothetic_nature_of_cities as the kind of thing I'd want to build upon.
I'm surprised the Canberra one merits a mention and Melbourne doesn't - is there really such an overconcentration of Rats there that it outweighs the fact Melbourne is >10x Canberra's population?
Disappointed by the lack of a fancy voting system. In my opinion there were four approximately equal reviews. Seems a shame to not be able to distinguish between my second favourite (which I thought was excellent) and my least favourite (which I couldn't even finish).
It also incentivises me to vote tactically which I don't really want to do.
Were we supposed to vote for our favorite book review, or for the review that we think most plausibly imitated the writing style of Astral Codex Ten?
I am a Russian with residency in Turkey, I'm an employee of an American company. I am considering whether to ask them to pay my salary to a Wise USD account (which means using some American bank's account details). When I ask the financial guy in my company, he told me that there's approximately a 5% chance that at some point IRS will ask me who I am and ask me to prove that I don't owe the US any taxes (I don't because I'm not in the US and am not a citizen of the US) and that this would be very difficult and problematic for me. Is that true? Why would it be difficult and problematic?
How do I get Substack to stop sending me emails but NOT stop showing me newly published articles from people I subscribe to when I go to their webpage or use the app?
The whole thing is a mess. There's some sort of difference between Substacks that I pay for and those that I don't, and some difference between the webpage and the app. The result is that my email box is filled up with fucking dozens of substack emails every day. I can't even stay ahead of them to delete them. But when I unclick "send me email" it ALSO stops updating the app. If that makes sense.
I stumbled across the life of William U’Ren recently. I bet Oregonians know of him, but not being from Oregon I had never heard of him, although I was familiar with some of the reforms he promoted, e.g. direct election of senators. I think a few others here might find him intriguing because of his efforts at voting reform and because he was a Georgist.
https://en.m.wikipedia.org/wiki/William_Simon_U'Ren
How did I come across him? I was researching cherry trees for my backyard. Apparently William U’Ren once had an asthma attack while down on his luck, and was taken in by Seth Luelling, the politically progressive horticulturalist who developed the Bing cherry. Seth and his brother Henderson Luelling would rank high on the ACX “Puritan scale”, although they were in fact Quakers. After attempting to run a utopian community elsewhere, they founded the first commercial fruit tree nursery in the Pacific Northwest. The Luellings were staunch abolitionists and might have maintained a stop on the Underground Railroad, though this can’t be proven.
One parent variety of the Bing cherry is the “Black Republican” cherry, which Luelling named in support of Abraham Lincoln. A “Black Republican” back then was one who supported emancipation.
But the famous Bing cherry we see in supermarkets was named for Ah Bing, Seth Luelling’s Chinese orchard foreman. After 35 years working for Luelling, Bing went back to China for a visit in the 1880s and was prohibited to return to the US under the Chinese Exclusion Act.
If anyone finds it helpful, I've made a tier list for the book reviews where I've been keeping track of which ones I've liked over the course of the contest.
https://tiermaker.com/create/acx-book-review-contest-entries-15209116
How do proponents of converting the US to multimember districts think this is going to, like, work? Globally the average size for multimember districts is 3-7 representatives- right now the US House has 435 members. At 1305 members it would be either the world's largest lower house, or certainly among them- at 3045 members.... we are entering 'patently ridiculous' territory. To be fair, you could still probably keep 1-2 reps for smaller states (Wyoming, etc.) and just use multimember for the bigger ones, but there's absolutely no way that you end up with less 1300-1400 members of the House. I don't see the American people accepting this.
The right way to do multimember, if the US were starting from scratch, would be to simply use larger districts than we have now. I.e. Illinois just to pick a random example has 18 House members right now, this would be reduced to say 6. But that seems politically impossible to force on the country- people like having local representatives for their little fiefdom, they don't want to share representation with 'those people on the other side of the state, ugh'. Voters would be apoplectic if you shrunk the number of state districts.
So, multimember proponents- you're proposing 1500 or so House reps, is that it? How would this work?
https://youtu.be/wPMum9Slz7M
An animator sets out to animate each letter of the alphabet and ends up with a 15 minute action movie.
Analysis of the end: https://www.youtube.com/watch?v=icN229DNC4w&lc=UgygmaXxNuwlilQwA4J4AaABAg
There's probably more to it than that.
Discussion: https://www.metafilter.com/196364/The-alphabet-animated-one-letter-at-a-time
Scott + anyone else: what do you think of routine use of seroquel/quetiapine being used semi routinely in ICU setting?
In my high school health class, I learned that drinking before your brain is fully developed is bad. I recently learned that the brain matures fully at around 25. Am I losing those sweet, sweet IQ points if I drink before turning 25 (I'm 18)? Or is the effect negligible at my age? A few searches brought me nothing but the usual "drinking bad" and many mentions of alcohol abuse disorder and drunk driving accidents, while I'm more interested in the concrete long-term effects.
I wish we could have used a ranked voting system of some sort instead. Having to just pick one out of 15 feels a bit frustrating.
Loved reading the reviews - great bang for the time invested. Lately I've started to suspect books are a bit... fat. Padded. Could be that they always were so, and long form blogging just spoiled me? Or I should just find better books.
Also loved the unofficial tradition of adding a personal insight to the reviews. Regardless of whether you agree or not with each particular idea, I just like the personal touch and the guts.
Names:
Many years ago, I read about cultures where people routinely changed their names over the course of their lifespan, sometimes several times. Not the vestigial forms we had - nicknames, and, for women (at the time) married names, but something a lot more drastic. And in some cases, the names told a story about the person's life. This seemed utterly weird to me.
But now I realize that this pretty much describes my history on line. DinoNerd is relatively new, and would not have been suitable until I'd been in tech long enough to count as a "dinosaur". I've had at least 3 other nicks I've used all over the net; those seem to have lasted about a decade each. Then there are the single-use nicks, and the few sites - mostly long ago - where I used whatever email ID I happened to have, generally a variant on my legal name. And this feels natural to me, complete with a sense that I'm no longer the person who used the earlier nick, and it's weird to login somewhere I've been long enough to still have it.
I'm guessing many others have similar experiences, and wondering what thoughts people have about the cultural and psychological effects, based on their personal experiences.
How bad is the European energy crisis going to get?
I may have missed it, but do you have plans to write up anything more about Lorien? Curious to hear how it's going!
Nature Human Behaviour comes out as woke gender :
https://www.nature.com/articles/s41562-022-01443-2
> [...] research may — inadvertently — stigmatize individuals or human groups. It may be discriminatory, racist, sexist, ableist or homophobic. It may provide justification for undermining the human rights of specific groups,
> Race and ethnicity are sociopolitical constructs.
>In some cases, however, potential harms to the populations studied may outweigh the benefit of publication
Your facts are transphobic, sweety.
https://www.nytimes.com/2022/08/27/health/teens-psychiatric-drugs.html
I wonder if now that we realized doctors were giving our opioids like candy, there will be a major movement to change the way people - especially kids - are medicated. I wonder what Scott thinks about whether this is a problem and if it’s gotten worse over time.
'pick one'.... 'fancy'...??? what blog am i on? or is this also a joke like 'ten more book reviews'!?
Taxpayers contribute to college education, and in the next California budget around 13% is earmarked for higher ed. I guess we assume that college grads contribute more to society, so it is worthwhile for all of us to pitch in. But the student loan crisis is a free market indicator that way to many people signed up for the college dream. What metrics should America use in deciding whom and how many should be sent to university? I have this vague idea that taxpayers should only fund excellent students. That seems more fair.
I myself don't think that any of the alternative voting schemes are better than FPTP. In particular, the complexity of alternatives raises doubt about the integrity of the process that isn't there in a properly regulated simple polling system.
On edit: I am strongly biased against these alternative systems because every instance that I have seen them used/proposed (Hugo Awards, post-Trump voting) it has been to keep the "weirdos and wreckers" on my side out. (If there are other uses, I would appreciate hearing about them - for instance, did anyone propose using ranked alternative to keep AOC from being elected?)
I feel like the book review entries are overall weaker than last year- is that just me or do people share that impression? Off the top of my head the Galen or Henry George reviews both feel like I'd pick them over any of the ones in this round.
You have 15 candidates, of which exactly one is a work of fiction, and you're using FPTP. Are you trying to split the vote and hand victory to God Emperor of Dune? And, in a world in which God Emperor of Dune actually is the most popular (as would be measured by a reasonable voting method), how could its victory seem legitimate when it could so easily be explained by vote splitting?
What should you use? I recommend STAR Voting (https://www.starvoting.us/star); it's simple, expressive, doesn't make voters agonize over which of two similar options is slightly better, and there's an easy tool for making online STAR elections (https://star.vote/). Other commenters have mentioned Approval Voting and Minimax as other options, and both of them are reasonable; Approval is good if you want the system to be as simple as possible (though voters need to consider strategy more than they do under more complicated methods), and Minimax, like all Condorcet methods, delivers excellent results (though I prefer the scoring ballot to a ranked ballot here so that I'm not focusing on minor differences).
What is not reasonable is using FPTP. It ignores the majority of voters' preferences. It makes being different from other candidates more important than being better than other candidates. It fails to give authors a good sense of how well-received their book review was since it doesn't distinguish between broadly-popular reviews that excite few people and polarizing reviews that excite a few people but which most readers aren't interested in. And, of course, the real problem (https://forum.effectivealtruism.org/posts/QCHLkgudfgbovgoan/cause-exploration-prizes-voting-methods) is with real-world elections, not online polls - but online polls like this one are excellent opportunities to bolster the profile of little-known but high-quality voting methods, and the data can be legitimately useful in understanding voter behavior. Using FPTP here doesn't just risk unrepresentative outcomes - it's a missed opportunity for building awareness and generating data.
There's been a lot of discussion here about EA and weird seagull-eye-pecking causes but, in terms of currently-existing-human type charities, what do we, the ACX hive mind, support? The GiveWell recommended list? (Malaria consortium, against malaria foundation, Helen Keller int'l, and New Incentives)
Bonus question: What are your thoughts on the recent removal from the recommended list of GiveDirectly and the deworming charities?
Ever wondered how Australia ended up with its voting system (known in Australia as "preferential voting")? It was caused by the equivalent of a GOP split.
Wikipedia: "The preferential system was introduced for federal elections in 1918, in response to the rise of the Country Party, a party representing small farmers. The Country Party split the anti-Labor vote in conservative country areas, allowing Labor candidates to win on a minority vote. The conservative federal government of Billy Hughes introduced preferential voting as a means of allowing competition between the two conservative parties without putting seats at risk."
Dr. Haidt is a thoughtful psychologist and has written some great books, but The Righteous Mind was pretty nonsensical, in my view. So, I liked your excellent review, that happened to agree with my opinion!
This review gets my vote. Is there a specific place I go to vote?
Where can I find other work by "the ACX podcast team"?
I'm a bit surprised this is a plain first-past-the-post vote, given all the discussion we've had about alternatives to current voting systems to select our Elected Representatives. At least we'll get a clear winner this time (unless the vote gets split so there are three 'first places').
I actually would have read another ten reviews, these slipped by incredibly easily and quickly. So if Scott feels that he needs some 'filler' content in future, posting another contest entrant that didn't make the shortlist would be fine by me.
I think you all will enjoy my interview of Steve Hsu about embryo selection. https://www.dwarkeshpatel.com/p/steve-hsu
We go deep into the weeds of their polygenic prediction algorithms, what kind of consumer adoption and regulation they expect, why the genome is relatively easy to improve, why natural selection hasn't already optimized these outcomes, differences in ancestral populations, and much more.
In the comment thread for the Exhaustion review, someone going by Fred mentioned that cranial traction sometimes helps with CFS. I did a little research, and it seemed safe and easy enough to try, so I did… and, within 24 hours, my fatigue level had gone from "debilitating and persistent" to something like normal.
Background: I had some good days, but mostly bad days. On a bad day, things as simple as a doctor appointment or a grocery trip would leave me exhausted for days afterward. Riding in a car could be a painful experience, sometimes requiring concentration just to keep from crying out in pain at every little bump. Walking my dog to the park and back was a major expedition, and I generally returned home staggering with fatigue… or, simply collapsed on the way home and had to be picked up. When I was lucky, post-walk fatigue would only lay me out for the rest of the day.
Working a single hour in a day was a major achievement, and I flaked out on my students almost as often as not. Stimulants, like adderall and modafinil, tended to make my fatigue *worse*, since they made me much more likely to overdo things. Caffeine was a devil's bargain. Best of all: the good days were becoming fewer and fewer, and the bad days worse and worse. (All five times I collapsed while walking the dog were in the last year.)
To my great surprise, self-traction had immediate and dramatic results. All I did was grip my skull and lift upward under my own power for 5-10 seconds… and it led to an incredible feeling of relief. I have tried to find words for the sensation I experienced, but nothing fits well. The best description I can manage is a strained metaphor: As long as I can remember, certainly all of my adult life, I have been buried up to my eyes in quicksand; during each self-traction "session", I felt like I was being actively lifted another 20-30cm clear. Bear in mind that an entire "session" of manual cranial traction is less than ten seconds long.
Eightish "sessions" and less than 24 hours later, I took my dog on her vet-recommended 3km walk. "Night and day" barely scratches the surface. Walking was easy, natural, automatic. Holding myself up didn't take constant effort. Keeping up with the dog was easy. About a kilometer in, I felt like breaking into a sprint—so I did. I ran for nearly a kilometer before slowing to catch my breath. Something that hadn't happened to me in decades happened—I *did* catch my breath. After resting for a minute or two, I felt up to running again. The dog vetoed this. (I realized afterward that her unwillingness to exert was probably because it was 37°C outside.)
Days later, I saw my physical therapist, and I asked her about this. She confirmed that my particular brand of chronic fatigue can, indeed, be caused by neck "instability", and (among other things) recommended an over-the-door traction device. My primary care provider agreed with her on all counts. So, on the advice of two capital-D Doctors, I got one of those devices. Now I use it, following my PT's directions, about once per two days. This is less portable, but much more repeatable and reliable than manual traction, especially on myself.
It has now been three weeks. I have been comparatively free of fatigue. I have had occasional bouts; when I notice what's happening, I try some quick self-traction or a brief stint in the traction harness. This often improves it, and earns a lesser version of that same initial feeling of abject relief. Whether or not they respond to traction, these bouts no longer come anywhere near what used to be my normal "baseline bad day".
Grocery stores are a breeze. Housework is manageable. I've only missed one tutoring session, and in fact I have reinstated a lapsed student *and* scheduled more weekly sessions, all while *also* managing to "work" 4-8 hours without having to spend days recovering. Not only am I walking the dog regularly again, she's now getting between five and twenty kilometers of walking *every day*. In fact, sometimes I achieve the holy grail of dog walking: the dog *wants* to go home and end the walk. (And not because of 37°C temperatures, either. I learned my lesson.)
There's still the matter of my mental health, which is a monumental problem of its own, and the genetically-inevitable joint problems that are flaring up in response to my skyrocketing activity level, and of course the habits of inactivity that must individually be broken… but, for the first time in a very long time, I feel like there's a way forward for me. No longer a mentally ill robot in a crumbling, useless body, I'm now a mentally ill robot in a mildly out-of-shape body. As I devour the lowest-hanging fruits of fitness, my overall physical health improves noticeably every day. And it's all because Professor Anna Schaffner wrote a book about chronic exhaustion, some yet-anonymous ACX reader wrote a review of it, Scott posted it as a finalist, and, deep in the comment thread, somebody going by Fred offhandedly mentioned that cranial traction sometimes helps.
Thanks Fred.
https://www.youtube.com/watch?v=iQTpRBkcQes
Coleman Hughes interviews Rafael Mangual-- 2 hours, but fairly effiecient
Mangual's premise is that violent crime does serious damage which goes well beyond the injury to the immediate victims (well sourced), and that this is getting ignored by people from safe neighborhoods who focus on abuses by the police and justice system.
It's a rough listen for me-- I haven't been saying BLM and ACAB (all cops are bastards) or advocating for defunding the police, but I've been focusing on the likes of Radley Balko (journalist who focuses on justice system atrocities) and The Innocence Project (gets falsely convicted people out of prison).
It seems plausible to me that many police are abusive and protected by the system *and* that police might, on the average, be doing useful work.
Once upon a time, an engineer built a novel AI. In order to deal with AI Safety, he air-gapped it. Some time passed, and the engineer noticed that the AI was trying to break through the air gap. To prevent this, he crippled the AI permanently.
- A retelling of The Tower of Babel
If you follow this link, you can get to the original of this comment where there are working links to the actual comments from Richard Gadsden, Brinkwater, and jumpingjacksplash:
https://docs.google.com/document/d/1Qz7_UxvIdVCeUw_xoJNBF35lgph-CUN_hCBgLsvJyTA/
I love intellectual debates and Astral Codex Ten has some of the best. But even more important to me is feeling like I could possibly make a difference in our world. For that reason I need the debate to lead somewhere.
If I come to believe that the debates are just a sophisticated game where we get points for sounding intelligent and persuasive, but they have no significance in the real world… Well, that's actually a downer for me.
In the Model Monday thread of Aug 1, a lot of people had a great time whomping on the plan for Neon. And, indeed, who wants to live in a narrow dark manmade canyon?
But I saw a glimmer of something good in this proposed boondoggle. The fact is that our current approach to civilization (and way too many humans) is destroying nature, and the compact form of The Line would be good for nature.
I don't know your gut feelings about nature, but I sincerely advise you: destroying it is not a good plan. So when I see a proposal that claims to be good for nature, my ears prick up. That there is not a lot of living things in Saudi Arabia is irrelevant. (As is the fact that MBS is a monster.)
So I spent a good chunk of August trying to see if I could fix Neom. With invaluable help from
Richard Gadsden,
Brinkwater, and
jumpingjacksplash,
I believe I have done that.
Rodes.pub/LineLoop
You enjoyed picking apart the pathetic Neom proposal, why don't you test your intellectual prowess on something more rugged?
But perhaps the next sparkling debate is more interesting to you than saving our planet.
You can also reach me at PRX555@gmail.com.
How do you prevent the voting from being gamed? E.g from people voting multiple times or from people voting to help the person they know (without actually having read the reviews)
What long-form reads does everyone have online? Whether subscriptions (substacks, newspapers, etc.), blogs, anything else?
Basically I'm looking to replace Twitter use in my downtime and keen for any recommendations!
I'm torn between the reviews for "The Dawn of Everything" and "The Righteous Mind". I felt both were competent take downs of overrated books, which is the kind of thing I find gratifying, but I'm leaning towards "The Righteous Mind" because I didn't find the reviewers' hypothesis about gossip traps very compelling in the "The Dawn of Everything", although it was interesting.
I want to vote for more than one book review. Can I vote multiple times? if I use different names?
Dall-E and the like can do art in various styles based on verbal prompts. Any thoughts about when or if an AI can produce a style (visual or musical) that catches on for people?
If non-existant potential future people have moral mass, why don't non-existant potential past people? What about potential future people of the counterfactual potential past people?
(edit: and how will we count their book-review votes?)
The Dawn Of Everything is the runaway winner, if you count publicly visible likes.
No approval voting? You're not worried about vote-splitting?