I figured out the problem. "This is costless in my ontology and expensive in another, making it a good signal of which ontology I believe" is the exact same algorithm that led doctors to go from autopsies to childbirth without washing their hands. It was a way to prove they didn't believe the silly peasant superstitions about disease and demons.
Anyone for whom executive function is a limiting reagent, unless the $10 is exceptionally significant to them. If it's a one-off transaction, it costs more than just the minute to watch the ad- it's the time and energy to evaluate the offer, see if there are any catches, refocus after disrupting your thoughts, fight against any impact the ad would have on you (they wouldn't keep making this deal if it wasn't positive EV for them. Maybe they're mistaken, maybe you're an exception- but that takes thought to figure out)... The fact that it's a pretty good hourly rate doesn't matter if it's not actually running long enough to amortize the evaluation costs.
From research I did on New Zealand: 14 days isn't long enough if people are isolating together- live infections get passed to housemates partway through, who are still contagious when the 14 days are up. So either you have to truly isolate everyone, or extend it by some number of days for each additional person.
Also, I assume you mean "14 days, unless they test positive or show symptoms", but then you have to figure out how to test and verify symptoms for everyone.
A bunch of people have alluded to Decision Tree being "non-representative" in ways I worried people might interpret as "unusually bad" rather than "different by design in ways that predictably made things harder". I also think there are some lessons from the Decision Tree experiment that aren't currently obvious and could save people a lot of heartache. So I got permission from the organizers to share more of the story.
There was a group house in the bay that had a history of taking in strays (more than one in fact, but we're only talking about one). They would get to know people, usually online, with job loss, needing to escape abuse, or just wanting to move to CA and needing some help to get some feet under them and give them a few months couch time and food. It worked in part because the hosts were really good at boundaries and so really could limit themselves to just providing couch, food, and an amount of emotional support they were truly comfortable with. I don't know the whole roster, but I know at least four people this house took in who were much better off for it, and no one who was made worse off.
But they only had one couch, and it took months to get people on their feet, and so many people needed help. So with the help of a benefactor they decided to create an entire house dedicated to giving people this launching pad. That was Decision Tree. The plan was to subsidize rent and provide various RA/RD types to help people get on track.
Unfortunately this meant that you went from a house full of fully functional people + one rescue person, to a house full of rescues (with the people who became functional fastest leaving soonest). There are a million reasons someone might need a place to crash for a few months. An unfortunate number of them are also reasons someone might be a difficult roommate. People who have only ever lived in dysfunctional situations, or are coming off a major trauma, are on average either harder to live or find living with other people harder. The subsidized rent meant that some portion of people genuinely couldn't afford to move, which could force housemates to stay together where richer people would have moved apart.
So that's why I think Decision Tree is not a great data point for group houses as a whole.
John and I had a fantastic offline discussion and I'm currently revising this in light of that. We're also working on a postmortem on the whole thing that I expect to be very informative. I keep mission creeping on my edits and response and it's going to take a while so I'm writing the bare minimum comment to register that this is happening.
I looked into the success of different countries' quarantines. New Zealand had both the best implementation and the best data, so I draw most of my conclusions from them. With a 14 day quarantine (testing on day 3 and 12), New Zealand had a "barely visible on the graph" number of import-adjacent infections. But according to a statistical model, these are caused by infections caught during quarantine (i.e. a couple with one infected member quarantines together, the second member catches it on day 7, leaves after an effective quarantine of only 7 days, then spreads it in the wild), so if people are completely isolated and occasionally tested, 14 days in indeed sufficient
I agree that Decision Tree was non-representative by design (in ways I'm not sure are public), in ways that will make it perform worse on average. I think that should have been noted more explicitly. I also think deluks is being really brave in naming something that made them a worse person, and I'm grateful they provided that data point.
I've been thinking a lot about this comment, and wanted to think more, but it seems useful to have something up as voting starts, so....
I think there's A Thing JW both agree is harmful (around assigning people moral responsibility when they're responding to incentives), and that I was trying to fight against. One thing I took from this comment is there's a good chance I had only a partial victory against Harmful Thing, and tried to pull down the master's house with the master's tools. I'd be very interested in exploring that further. (I also think it's possible JW is doing the same thing... it's a hard trap to escape)
I don't think giving up the question "Who should we blame?" entirely is a good idea. Possibly the benefits of the norm would outweigh the costs for LessWrong in particular, but I don't believe such a norm would be a pareto improvement.
Last week we announced a prize for the best example of an evaluation. The winner of the evaluations prize is David Manheim, for his detailed suggestions on quantitative measures in psychology. I selected this answer because, although IAT was already on my list, David provided novel information about multiple tests that saved me a lot of work in evaluating them. David has had involvement with QURI (which funded this work) in the past and may again in the future, so this feels a little awkward, but ultimately it was the best suggestion so it didn’t feel right to take the prize away from him.
EDIT: David has elected to have the prize donated to GiveWell.
Honorable mentions to Orborde on financial stress tests, which was a very relevant suggestion that I was unfortunately already familiar with, and alexrjl on rock climbing route grades, which I would never have thought of in a million years but has less transferability to the kinds of things we want to evaluate.
How useful was this prize? I think running the contest was more useful than $50 of my time, however it was not as useful as it could have been because the target moved after we announced the contest. I went from writing about evaluations as a whole to specifically evaluations that worked, and I’m sure if I’d asked for examples of that they would have been provided. So possibly I should have waited to refine my question before asking for examples. On the other hand, the project was refined in part by looking at a wide array of examples (generated here and elsewhere), and it might have taken longer to hone in on a specific facet without the contest.
Other timeless but year-of-publication restricted anthologies like "Year's Best Dark Fantasy and Horror" and "Year's Best Science Writing" have either "Nth annual" or [year the entries were published] prominently on the title. This is an established convention. The problems of "what the hell book did I read that in?", "Finding the books on Amazon" and "Have I read this already? Who's to say." seem much bigger to me than a fraction of the audience that hasn't picked up that convention AND will be blocked from reading by it.
My risk model treats all the available vaccines as "drug that was developed under political and financial pressure and whose trials ended much sooner than is normally the case".
I think we can do better than that, even in the current information climate. Drugs can be almost anything and have almost any goal. Vaccines are a pretty narrow class of treatment that are attempting to do one thing- give you an immune memory that will trigger if infected by a disease. They do that by exposing you to some part of the disease. The natural cap of badness of that attempt is giving you the disease itself- anything more needs an explanation.
Some examples off the top of my head where a vaccine caused effects you wouldn't get from the disease itself (not all of which rendered the vaccine net negative):
certain adjuvants encouraged cancer (e.g. some pet vaccine)
the vaccine triggered an immune overreaction that left people worse off if actually infected (there was an STD vaccine that did this, and I believe SARS-1)
Allergy to something else in the vaccine (e.g. eggs in the flu vaccine).
Of these, we'd expect #3 to show up nigh immediately upon immunization, and is not dependent on how many people were exposed to actual covid, so we have a fairly large sample size. I vaguely recollect that where #2 was a factor, it was pretty universally true, not a rare reaction- so the sample size is probably large enough for that too.
This is complicated by the fact that at least one of the covid vaccines is using an entirely new mechanism. This could leave us vulnerable to certain problems like the adjuvants, that take a long time and large sample size to catch.
Certainly some people are very sensitive and medically excused from vaccines- but those people are pretty screwed if they catch the actual disease too. The only reason not getting the vaccine is viable for them is herd immunity.
I'm not an expert and I haven't looked into this very long. But there are mechanistic models that can help us predict the risk here, and I think it's a mistake to use the entire collection of FDA-monitored treatments as a reference class.
Eli was in fact checking in while I babbled, in retrospect he was putting out feelers for redirection but I was so happy with how generative I was being I wasn't that responsive, and I think he was reluctant to push back because what if strict separation of babble and prune was the best process? It also might not have been obvious how much transformation my babbled ideas needed to be usable until we did the next step.
Follow up: @elityre asked to discuss doublecrux ideas and I spent a little more than an hour spewing forth topics. There were 74 in the first 30 minutes and its slowed down from there. I don't think I would have done that nearly as easily without the babble challenges.
It turns out these were answers to the slightly wrong question, so we had to do a prune pass. I'm not sure if it would have been better to clarify first, or if the absolute babble freedom ended up leading to better post-pruning ideas.
I can't say it's impossible, but you would have to constantly go back and forth whenever you got new information, while hiding from yourself that you were doing so, and at a certain point it becomes psychologically easier to work hard despite a lack of guaranteed success.
I agree the post strongly frames things as for potential winners who aren't considering other options, but I think that's a mistake for this exact reason. Even if a set that benefits from self-deceit exists, you can't know if you're in it while you're self-deceiving.
Yes, oxytocin and other hormones released during hugs/sex have a similar effect to drugs. Actually, most drugs don’t add anything into your body but will simply increase/activate or affect hormones naturally present, which will give you an altered state of mind
This doesn't seem surprising or compelling to me. People say similar things about morphine and sugar, but of course drugs take advantage of existing systems that evolved (imperfectly) to help you: why would you evolve a neurochemical system set up for the sole purpose of fucking up your life if you took the wrong chemical? The surprising thing would be if drugs used a system that didn't have a normal, survival-aiding purpose.
Regardless of what the truth is, it seems like working hard is the best way forward, rather than getting lazy or seeking side-hustles.
This isn't obvious to me. If you are in, say, the bottom 90% and have no chance of getting an academic job no matter how hard you work, knowing as soon as possible and making exit plans seems like the best thing.
Alternate explanation: people who are capable of believing they are good at a thing will eventually stop trying to improve, while people who are stuck perceiving themselves as unacceptably bad at X will continue working to get better.
I look smarter than people in areas with weaker infrastructure because the infrastructure is taking cognitive load off of me
Invisible hand, bitches!
It's not always obvious when the magic that let you form and maintain something goes away: there's a long lead time before the cracks appear, long enough that you may have lost all the people who could recreate the original magic. Similar to the England/scurvy/lime juice story.
It's easy to misidentify the magic, especially when there are incentives to do so.
Possible that Great Depression led to unusual period of government competence because other jobs were scarce and there was a premium on job stability, so they could attract better applicants.
Anna seems to be skipping over institutional/individual alignment
Certain kinds of coordination have gotten vastly easier or moved from impossible to possible using modern logistics tech.
How are CA's electrical failures related to bad governance? I see bad results but it's not obvious the necessitating circumstances had responses that wouldn't be called bad governance.
Not clear how much worse the US actually did on covid- Europe is having a resurgence now.
Curious about how subcultures are harder to form, seems like it would be way easier than when you were stuck in the same 200 person town for life.
Predicted response: being too easy to form/join is actually bad.
HAM radio culture as described in that autism book was a dead ringer for lesswrong, and that was pre-1975.
Pushing out casuals is hard because sometimes they are nice and your friends.
Gatekeeping is so, so hard/frowned upon in our culture
Unless you call breaking past the gates appropriation, then you're allowed defenses.
Sometimes geeks are really weak on particular skills and bringing in some casuals with those skills is really helpful to the overall subculture.
What are the net effects if it becomes harder to keep out casuals?
Fewer subcultures reach critical mass -> less novel art.
Wasn't there a study showing pop music is changing much slower than it used to? That could be confirmatory if the methodology is good.
How does this interact with science?
Lack of defended factions leads to even more pressure for premature consensus than currently exists.
Something something inter-institution diversity vs intra-institution diversity.
Is it more important that a single campus has many viewpoints represented, or that each viewpoint has a stronghold university focusing on it?
Dad talking about how jazz had got more locally but less nationally diverse.
From the perspective of people living at that time, that's more variety which seems good.
That scene in Treme where the high-status Jazz player has to defend New Orleans jazz.
multiple institutions with very different views are easier to defend than the same institutions with a variety of views.
But any given student is attending at most one school at a time, and will only get a single POV if at at a stronghold institution.
Large schools could have a stronghold + tokens from other viewpoints, who will not be as productive as scholars at strongholds for their viewpoints. It's not like there's a shortage of professors.
Chapman leaves out the conversion of casuals -> geeks, or how hard it can be to distinguish future geeks from future casuals.
If it's too hard to join your club, people won't do it and the club will atrophy and ossify. E.g. wikipedia editor culture.
Ray's melting gold essay
Where the hell did the 6:1 ratio in Chapman's essay come from?
"Under [the Geeks, MOPs, and Sociopaths] model [...] it has somehow become easier and more common for people to successfully ape the appearances of an institutional culture, while not truly being true to it "
This has been happening in television. It used to be tactical competence (acting, dialogue, continuity), which is easy to assess early, was strongly associated with strategic competence (which often can't be assessed until the show is over) (Sopranos, Mad Men). Or tactical incompetence made it obvious who had strategic competence (Babylon 5). But lately they've figured out how to have good tactical competence without deep strategy, leading to shows that are compelling until a deeply disappointing ending (Game of Thrones).
The strongest form of Effective Altruism parasitizes first-world institutions by benefiting from the intangible resources like trust and prosociality, while not contributing to them, and moving resources from the populations with those institutions to the (much, much needier) populations outside of them. This can be very moral on an individual level (because those outside the institutions are so much needier than those inside, for the exact reason of being outside), but disastrous if widespread because the institutions do actually need resources to run.
This suggests that bringing people into the institutional ecosystem (e.g. immigration) is better/more sustainable than moving resources from one ecosystem to another.
Maybe taking away all the rewards of power/declaring them immoral was a terrible idea, because it drove all the good-but-not-100%-self-sacrificing people out of power.
Amazon seems way more competent than Google but I'd rather work for Google because they're nicer.
Eh, is that true? I hated working for Google specifically because they didn't care about my actual productivity, and Amazon has supposedly softened from when everyone was wrung out in 18 months.
Door desks are still stupid though.
See also: not buying developers the best laptops.
Some institutions (PTA) run on invisible slack and we ate up all the slack.
Two income trap.
I'm learning to draw and have all the resources I could possibly use at my fingertips. Art school is less necessary than it used to be. Grad school used to be the only way to keep up with academic research: now I have sci-hub and twitter. This can simultaneously be awesome for me, outside of art school and academia, but if art school/academia had an off-the-books goal it will suffer as people/resources leave.
Existing institutions that have invisibly lost the magic can inhibit the formation/expansion of newer institutions that have it.
WEIRD cultures' fairness towards strangers is in fact really weird and would be bad to lose.
Destroying minority-owned institutions that made them less dependent on the majority had a worse impact than the majority excluding the minority from their institutions (e.g. Tulsa massacre, "Urban Renewal" destroying black business districts)
Coding bootcamps were originally aimed at people who knew how to code but needed a little polish before they could excel at software eng. They ran through that backlog and are now their taking people from 0->coder. This greatly lowers the value of a bootcamp grad and the bubble is due to pop soon.
[Redacted] described the same thing with plumbing. Master plumbers are replaced with people who have been taught how to execute a few things but can't manage the system as a whole.
Treating public education as a thing done for the students as opposed to the country (which benefits from educated citizens) has huge implications for how you focus, especially what you do with the tails of the IQ curve.
Treating public education as for the students rather than for the country is a huge luxury. As a country declines, it can either give it up or collapse.
The concept of nepotism being bad is fairly new and precious.
Many political institutions are dead inside but received lip service anyway. Some of Trump's inflammatory statements weren't different from what other politicians thought, people just objected to or enjoyed him saying it out loud.
George Washington was an unusually good country founder.
Cults are typically providing a (semblance of a) real need and much of the stigmatization of cults stigmatizes meeting that need, which just pushes people towards unhealthy ways of meeting it.
Institutions (and people) taking charge without responsibility do a lot of damage.
Say I have the option to purchase a lottery ticket with a 1/10m chance of winning, a $1m guaranteed jackpot, and a $1 ticket price. The EV of a ticket is $1m/10m - $1 = -$0.9, a bad deal. What is the value of information that narrows the odds?
If it narrowed the odds all the way to 1/1, the value would be $999,999, obviously a very good deal. I should be willing to pay up to almost $999,999 for it.
But if the odds are doubled? The EV of a purchase is now $1/5-$1 = -$0.80. In one sense I’ve gained $0.1. But since buying a lottery ticket is still EV < 0, I won’t purchase one, and so the information nets me no money. The information is worth $0 to me, as long as I have the option not to buy a ticket.
But… maybe there are multiple interventions that double my odds? Under the above rule, the first three are worth nothing and the fourth is worth $2^4/10 - $1 = $0.6. It seems wrong that 4 identical interventions would have such different values. I could just divide .6 by 4, but what if I’m not certain how many doublings will be available?
Can form treaties with ants to stay out of my kitchen in exchange for sugar cubes
Ants develop anti-poison technology
Ants identify humans as source of poison and begin formulating counter attacks
Ants identify humans as source of food and begin worshipping them
Discworld/Children of Time style computers -> less dependence on rare earth metals
Ants begin taking out poison factories
Ants begin raiding poison factories to use on other bugs
Massive decrease in ant predators, as ants organize to take them down -> increase in other insects eaten by those predators. Spider in SimAnt taken down instantly
Ants used as spies in war -> new and better pesticides, ant proofing, and ant detection
Necrocolony in chernobyl begins biting other colonies and turning them into zombies
That species of ant in Europe that's all one giant supercolony learns to distinguish between colonies and thus war is returned.
Giant increase in funding for studying ants -> I regret not going to grad school to study eusocial insects now that it's both interesting and profitable
Ants can be hired to do microcleaning, Disney-style.
Microconstruction takes off as ants are used as construction labor
Ants weather proof their shelter
Ants get good at hiding their colonies, but you can hire ants to find other ant colonies.
Ant chemical detectors, similar to the bee detectors we have now.
Huge debate about the morality of ant farms
Ants can coordinate to have a single item carried by many at once.
Ants develop heating and cooling systems for their colonies, expanding their possible range.
Ants discover global warming and develop carbon sink technology, we are all saved.
Ants develop eugenics program leading to super ants.
Ants learn to conglomerate into large shapes, can act as humanoid -> competing with humans for jobs. They are treated like illegal immigrants.
Ants recognize nuclear war as threat, organize worldwide campaign to disable all nukes simultaneously
Ants recognize nuclear war as awesome for them because it kills their smartest predators, organize worldwide simultaneous launch.
Process of uplifting teaches humans about the development of intelligence -> brings about or prevents UFAI.
Process of uplifting teaches humans about the development of intelligence -> terrorists and governments uplift other animals to be used as spies. Every animal becomes suspect, biome is nuked and paved.
Ant art becomes possible and trendy, creating an art bubble that eventually pops. Art colonies then left to starve to death as their work is no longer valuable.
Ants as watch dogs: learn to identify family members and trip alarm if someone comes in without one of them. Fire ants used as full guard dogs.
Ants used to clean infected wounds and do microsurgeries.
Jains and other sects that honor insect life become smug and righteous, leverage into better treatment for all insects just in case.
No construction can take place without an Ant Impact Statement. Ants become new construction chokepoint, get rich off of NIMBY bribes.
Ant digging abilities used to find treasure.
Ants replace mine rats.
Ants employed to keep public areas clean. Parks, beaches, etc become way less disgusting.
Ants used for simple home automation, like pet feeders.
Beautiful ant farms placed in museums, later looked back on as akin to when we displayed humans in zoos.
Ant colonies develop mental illnesses. They are made hopelessly worse by human psychologists. Ant psychologists become a thing.
Ratatouille but with ants
Ants take over contagious-disease medical care, as they can't possibly get sick from a human disease. Human doctors and nurses very upset about this, attempt to block with protests and strikes but that just accelerate ant dominance.
Movie about child with ant colony best friend.
Ant-facing businesses become the hot new thing, billions invested in tiny furniture, nicer colonies, etc. Ants prove to mostly not care and the bubble pops.
Humans decide aliens caused the uplift, begin praying to aliens for their own uplift
Humans wage war to destroy all ants, resulting destruction plus lack of ants leads to human extinction as well.
Ants used to cross air-gap in unnetworked computers. Eventually hackers discover a way to hack ants and thousands of highly secured computers are compromised.
Ant exercise program advertised. It is terrible and the progenitor is eventually sued out of existence.
Ant mafia develops.
New genius-ant-proof containers developed to store food.
I’m Elizabeth.You may remember me from such series as Epistemic Spot Checks and the LessWrong Covid Effort, but for the last year my main focus has been developing a method for Knowledge Bootstrapping- going from 0 to 1 in an unfamiliar field without undue deference to credentialism. I’m at the stage where I have a system that works well for me, and I’ve gotten feedback from a few other people about what works and doesn’t work for them, but there’s a long way to go. A lot of my knowledge is implicit and not explained on the page, plus I am only one person; what works for me will not translate perfectly for every human. So I’m looking for test subjects.
One particular part of my method is breaking down one large question into many smaller questions. This has several purposes: it forces you to clarify what you actually care about, and makes it more obvious what information is relevant. I describe this process and the reasoning behind it here, but not very well. I’m looking for test subjects that have a research question, and would like to practice breaking it down into smaller questions, with the goal of refining the technique and my teaching of it.
What This Looks Like
Come up with a question you might like to research.
You book a phone call with me via calendly, or email me at elizabeth -at- acesounderglass.com to set up a time.
We discuss your question in an attempt to break it down into smaller parts.
I sure hope some people actually go off and research the new questions but there’s no commitment required to do so.
What Are the Expected Outcomes?
You will have a better understanding of what you actually want to know and will be better positioned to find answers.
You will be better able to break down your next research question, without me.
I will make some of my metis on breaking down questions more explicit.
I will become better at teaching the technique of breaking down questions.
I learn techniques from you I couldn’t have learned on my own.
Yes, but not as part of this workflow, because Roam doesn't support offline yet and not being online is pretty critical. I will sometimes move things I write as part of this workflow into Roam after the fact.
Do you read literature sources on it as well?
Not sure what "it" refers to or what you mean by "literature sources". If you mean "do I do the kind of research I frequently talk about elsewhere using this workflow?" the answer is no.
Wikipedia defines the dodo bird verdict as "the claim that all empirically validated psychotherapies, regardless of their specific components, produce equivalent outcomes", which is noticeably different than "talking to a friend produces equivalent outcomes to therapy". They can both be true, but I don't think they're the same thing.
Additional factor: availability of compatible people who didn't move there for this group. This is important for several reasons, including:
Inflow of new ideas
Ease of joining and leaving. If rationalists take over a small town, the only thing there for them is other rationalists. That makes joining and leaving into very binary decisions. It doesn't let people slowly notice incompatibilities and amoeba into another social group.
This sure seems like it should work. My experience is that there's either nothing, or whatever quality analyses exist are drowned out by pap reviews (it is possible I should tolerate reading more pap reviews in order to find the gems). However I think you're right that for issues that have an academic presence, google scholar will return good results.
My experience is that readability doesn't translate much to quality and might even be negatively correlated, because reality is messy and simplifications are easier to read. I do think works that make themselves easy to double check are probably higher quality on average, but haven't rigorously tested this.
I originally named the types of knowledge "Type 1", "Type 2", and "Type 3", but was encouraged by early reviewers to actually name them. In light of the conversation here, I think doing that was a mistake. Unless I was sure my names were out of the park correct (and maybe not even then), I should have left it generic so I could get input on more people for what the names, and for that matter definitions, should be.
A thing I really structured to capture was that "i did actual research and had actual models for why masks would help against covid, but it's still not type-3", which is why "know why" doesn't feel right to me.
I tentatively think that some of what you're calling engineering knowledge would fit into what I call scientific (which is a strike agains the names), and/or that I didn't do a good enough job explaining why engineering knowledge is useful.