Except for the adopted ones, the ART-conceived (assisted reproductive technology) kids wouldn’t exist otherwise. And besides, the whole point of IVG (in-vitro gametogenesis) is to reduce the need for donor gametes.
How is being conceived in a different way to the norm being 'commodified'? I want to have kids with my partner, and right now that's not possible for us. If the Ivy Natal Fertility group succeed, it would be. The ability to have biological children is no more than what straight couples have, with the added bonus that all children born of this method would be intentional and desperately wanted.
Scott has some interesting thoughts outside the mainstream such that I am tentatively excited for some of these grants. If 20%+ end up paying off at least somewhat that would considerably raise the prestige and interest in later ACX grants I'd think.
I assume Scott has already done a forecasting success thing that he will reveal some years from now about which ones he feels most confident about. Of course like start ups having a 1 or even 2 in 10 hit rate is actually quite good.
I love the beetle one. Can we make that an ACX meme somehow?
(I also like that a nuclear war one showed up. EAers have always rated nuclear war and pandemic/bioterrorism as two of the 3 worst existential risks, but generally had an attitude that those two big problems were mostly better left to establishment government & science. Covd-19 suggests that this may have been misplaced confidence when it came to public health; perhaps we have misplaced confidence in the other one too.)
Of course we wouldn't eat the beetles, except to fend off starvation in desperate🪲circumstances. Obviously.
But then, there's some elasticity around what counts as "desperate circumstances," depending on how the beetles taste.🪲And we can't very well claim to know the exact hedonic exchange rate between eating the beetles versus other🪲food sources, if we haven't tried them even once...
🪲Of course we wouldn't eat the beetles, except for experimental purposes, to determine under what circumstances🪲we should reconsider our anti-beetle-eating🪲stance. Obviously.
Also we should take the relative supply🪲🪲of beetles and other goods into consideration.🪲We do produce a lot of plastic waste,🪲after all. Look at them. Just crawling around,🪲not🪲being put to any better use...🪲🪲🪲
🪲🪲Of course🪲we should🪲put🪲🪲the beetles🪲to the🪲most🪲🪲🪲economically efficient🪲use.🪲Obviously.🪲And if that🪲🪲means eating them,🪲well,🪲it's not like🪲they🪲were so🪲bad that🪲one🪲🪲time...🪲🪲🪲🪲
Half of me loves the beetle one and half of me is thinking, if 20 years from now we're looking back on this and one of these has gone horribly wrong it's going to be that one.
There are already many species of bacteria that consume crude oil. They're all over the floor of the Gulf of Mexico, for example, where oil seeps from the sea floor and has for umpty millions of years. They're sometimes used to hasten the breakdown of oil when it's spilled accidentally.
So about three months equivalent of fossil fuel burning - not really a material consideration for a one-off even assuming a full plastic to CO2 conversion.
Thank you. (Hmm I think there is a math mistake here. One ton of CO2 is 12/(12+2*16) tons of carbon, call it 1/3, so total carbon in plastic is equal to carbon (w/ hydrogen) that we burn every year. We should just burn all plastic and get whatever energy we can out of it.
Yes, you're right, I didn't take into account the combustion chemistry (albeit it's not an order of magnitude difference).
You joke about burning plastic (I assume) but some forms are reasonably good feedstock for energy from waste power plants (assuming you've got all your filters up to date to clean out the toxic emissions resulting from it). That said, it's not a particularly efficient fuel by any means, but it's just a way to get some use out of what would otherwise go to landfill.
I specifically came here to say the same thing - these are generally great, but the plastic beetle one seems the most perfect fit for this kind of grant. Some smart biologist just needs a bit of funding for some equipment to keep beetles in; low probability of success, potentially hugely impactful.
To address a comment below, I'm not too concerned about catastrophic risk on this one as the plastic itself is unlikely to be nourishing enough to sustain explosive population growth. Perhaps the ideal outcome is that there is some kind of cheap spray that makes the plastic much more digestible to the beetles, thereby controlling spread (but that's a problem that only happens if this project is *wildly* successful).
Covid-19 has boosted my skepticism of establishment government & science so much as to make me fear for the entire EA project - I feel like if we don't find a way to address rampant demosclerosis nothing else matters, nothing else will work. I'd feel a lot happier if any of Scott's grants or Zvi's grants explicitly addressed the issue. Or if I had an idea of my OWN to address it - I might then have applied for a grant myself!
I just keep noticing that in areas of concern for which people successfully "raise awareness" the government's response is to build a bureaucracy to "address" the issue...and that bureaucracy near-inevitably makes the problem WORSE.
The TSA, created to make travel statistically safer, makes airports SO "safe" as to become so inconvenient/expensive/annoying it causes more people to drive instead of fly. The increased driving means more car accidents; net effect (at least on the current margin) is to make travel statistically less safe.
Nuclear power regulations, created to make power generation safer, require SO MUCH "safety" that nuclear power plants are unaffordable and take too long to develop, so states end up relying more on coal/oil/gas for baseline power. Which is more polluting and kills more people (mining accidents and air pollution); net effect is to make power generation less safe.
The FDA/CDC/WHO, created to make us healthier, have killed over 100,000 Americans in the last few years (and a great many foreigners) by postponing vaccine approval, by preventing cheap instant covid tests or challenge tests, by not letting firms update vaccines to new variants or letting them optimize dosage levels to produce more doses or letting them optimize vaccine timing to improve effectiveness etcetera; net effect is to make us less healthy. (and fewer of us alive).
If we can't fix this dynamic I have to expect the result of *successfully* bringing more attention to nuclear war prevention or asteroid strike prevention or unfriendly AI prevention or what-have-you is that once we show the way - once we demonstrate progress along a promising path toward improving things - our government will make these issues a priority and form an Official Department to be in charge of each issue whereupon said departments will end up making THESE problems worse too.
We might be better off trying to fix these kind of problems IN SECRET to avoid regulatory attention! A private EA cabal with secret symbols, handshakes, meetings. Like the FreeMasons.
I hadn't thought about the net effects of nuclear power regulations and TSA hassles. But you have to think about the public reaction to those: how would the public feel about having a nuclear reactor built if the government wasn't endorsing it as safe to an extreme extent? And would people be willing to fly in a post-9/11 world without the security theater of TSA? It's true that the regulatory agencies may have swung the pendulum too far toward safety, but the public probably wouldn't accept is as far the other way as you might like.
Without the TSA (especially once we privatize the airports and air control) it would be between airlines, their insurers, and their customers what level of security is chosen. Security measures that aren’t cost-effective (eg taking off shoes because one crazy guy once tried a thing involving shoes that failed and couldn’t possibly have worked) would tend to fall by the wayside. Measures which *actually matter* (eg, hardening the cockpit doors) would tend to stay. The reason the dumb measures would go away is that companies could offer lower prices by getting rid of that stuff.
Personally I would pay EXTRA for a flight with NO SECURITY AT ALL at the customer level. Let people walk straight to the gate without lining up anywhere at all. Let ‘em even have open-carry or concealed-carry guns if they like!
The funny thing is that this level of security *already exists* in the form of small-airport private planes. Very rich people CAN get on a plane with no security; it’s just us plebes in the cattle-car planes that can’t!
Allow companies to COMPETE on security and some may pick more, others less. If everybody LIKES getting felt up by wand-wielding morons after a Rapiscan pass then that will remain the standard but maybe one or two lines experiment with less to drop prices; if it works then more follow.
The main argument FOR federalizing this stuff is externality-based - if a low-security plane gets hijacked and hits a building that affects the people in the building too. But that specific kind of attack became impossible moments after 9/11 and remains so; it’s just not worth worrying about.
9/11 revealed the massive externalities possible from unsecured plane flight, and not just hijacking - also bombing, or bargaining with passenger lives. That security theater wasn't just for passengers' peace of mind, but for everyone in a building that might be struck by a runaway plane. I don't think the mass public would tolerate a full libertarian approach.
Also, how are businesses even going to learn which security measures are effective and which aren't? Counting the number of terrorist attacks that happen under each security regime? I don't think that would be tolerable either.
Fun fact about the TSA: The transporter booth you have to stand in spread-eagled every time you want to board a flight is an X-ray backscatter scanner. When the TSA introduced them, I got a copy of the paper cited to justify the claim that they were safe (which sadly I can't discover the title of now). The paper had shown that if you divide the amount of ionizing radiation shot at a passenger by the volume of the passenger, you got a radiation density considerably lower than levels considered safe for that particular type of radiation.
Unfortunately, most of the radiation is absorbed or scattered by the first 1.5 cm of human, so the radiation density in the parts actually irradiated (mostly skin and subq fat) is much greater than the average over the entire body's volume.
EDIT: BUT not nearly as much greater as I thought when I began typing this. Approximating myself as a perfect sphere of radius 120mm, I find that the average radiation density in the outmost 15mm of spherical me, which gets half of the radiation, is only 1.5 times as great as the average over my entire spherical volume.
The chemical problems are very different. The problem with digesting cellulose is prying apart the tightly-wound fibers to get access to the individual polysaccharide chains, and then cutting some of the glycosidic links (C-O-C linkages). That is, wood is chemically quite different from oil.
If you want to digest the most common plastics, the problem is that (1) they are not soluble, so it must be done in the gut by something that can digest things outside its body fluids, i.e. bacteria, and (2) your organism must be able to oxidize C-C bonds without any kind of chemical "handle" to hold onto them. (Our own mitochondria can oxidize them easily, but only if they have a carboxyl group at the head to grab onto.) Note that it can't oxidize just any old C-C bond, because of course then it would destroy itself.
Even then, it's going to be difficult to find or breed something that can break the benzene rings in polystyrene. They are very sturdy, and it is possible your bacteria, fed polystyrene, might only excrete a variety of small substituted benzenes, which would be worse than useless, since they have a decent chance of being carcinogens.
I think this whole process is awesome. I am excited to see follow-ups with the projects, and would also be interested in seeing which projects weren't funded but might still be interesting, like "runner-ups or honorary ideas". The beetle project and automated / hybrid tutoring seemed very cool to me; I don't know too much about biotech. but it would be cool if someone did a project to modify bacteria/insects/algae to be more nutritious (e.g., altering the taste of crickets or consumable-bacterial-growths) / produce more energy / consume-convert wastes into useful things.
My impression has always been that 4-7 or so, if you select them well, is a superior option for small scall education creating amazing outcomes. 1 on 1 is well ahead of 1-35 and has some advantages in curriculum tailoring, though.
>Despite the growing importance of this field, there are relatively few technical biosecurity centers in the US, and the West Coast is underrepresented. This causes serious problems like poor pandemic readiness, limited understanding of biowarfare risks, and the biosecurity grad student who I'm dating living 3,000 miles away from me.
One of these problems is not like the other ones 😂
Here in Boston, to me it seems like there are lots of biotech people but few biosecurity specialists. I think they're concentrated around Washington DC.
Washington D.C. makes sense. Why is Boston such a biotech hub? My initial hypothesis is proximity to a certain pair of prestigious universities, but I imagine there could be other reasons.
Development at Kendall Square (right behind MIT) has been targeted to biotech for a few years now. (This may be more of a description than an explanation, or half/half.)
The universities started it, but then comes clustering effects and targeted development. We just lost a beloved long-time food market to new lab space. What's really weird is driving around Cambridge and Watertown seeing billboards advertising reagents, next to the usual ads for fast food.
Before this Pandemic, Georgetown and Johns Hopkins were the only two universities that had a decent program in it I could find. Not surprising the experts would concentrate in DC, if that's where they studied. I predict many more will appear in coming years.
> This is the part where I post applications publicly on the blog (if you gave me permission) and readers can look at them and decide to support them or not. About 500 of you gave me permission to do this, and your applications together total about 1,500 pages of text. Substack probably won’t let me write a blog post this long, and you guys won’t read it even if I do, so I’m still thinking about how I want to handle this. Please give me until sometime in January to work something out, but rest assured, I haven’t forgotten about this.
I have a suggestion here. You have 500 ACX++ grants. I'd first weed out the ones you find objectionable. That not only do you not want to fund but you think lack merit. Then I'd pair them in small groups of highly different proposals. Then I'd let each person do a paragraph in a public Open Thread and tell them to be there to answer questions. The ones who won't agree are forfeiting their right to be included in ++.
I suspect this will cut down the number significantly. Those that remain will be rationed out over a year. But in exchange the grantees will get more exposure than people hunting through 2,000 pages or a huge database. And if you like it you can do this on a rolling basis instead of making it into a huge Christmas nut to be cracked like you did this year. (It'll also give you more content for however you weigh that.)
Thirding. As one of the 500 people, I would be in favor of some kind of “write an additional paragraph explaining your project” (perhaps a condensation of the original proposal) and I definitely want to second of the above commentor suggestion of an open thread participation for the AC X plus plus people.
Agreed. As someone with a rather speculative proposal that consumes almost no money apart from my own time, I was planning to post in an open thread if I didn't get "funded" anyway. (I already noted in my application that attention and "signal boosting" is more relevant for my project than money)
500 paragraphs is a whole lot of paragraphs. Even if half of them are meritless, 250 paragraphs is a whole lot of paragraphs. And lots of people never bother with comments on posts, especially open threads.
I think a better idea might be a single post, sorted into categories, with a very short sentence description of each grant as a link to the pages of text in the application, hosted somewhere else. It'd be a very long post, but if it's sorted into categories, people can peruse it for ideas they find interesting, link to the outside information to read more on it, and decide what to do. And maybe the links could include a place to ask and answer questions from founders.
It would probably take a lot of work to set this up, but Scott could probably pay someone to do it.
I like the idea, but this approach would require someone to read 500 ask-me-anything threads, summarize and interpret them, and then pick winners. I recommend also using some other web platform that can automatically collect votes or ratings from paying ACX subscribers, in order to circumvent the Scott-reads-10,000-comments step.
(Paying subscribers in order to reduce potential fraud, to get more paying subscribers, and because having the people who provide the money get to say where it goes would, by some long-standing but now aggressively-disputed standards, be more just, and motivate them to think more carefully.)
Earlier today Matthew Yglesias had a tweet about how despite widespread belief that numerous government institutions failed during the pandemic, there's been virtually no legislative effort to change anything. So I'm really glad to see ACX putting $100k toward a better FDA. Trying to fix what's broken instead of just having fun complaining. Hell yeah!
Oh, and as a biologist I want to strongly endorse the cell culture media testing project. This is the exact kind of thing that is 1) really important and 2) in a kind of hole where it's not the kind of thing any science funding agency would like to fund.
Strong agree - this one is tied with the beetle one for best application in my view. Making the process of doing science less onerous/painful is huge (not to mention the waste in having extremely smart people spending >50% of their time fiddling with petri dishes and whatnot, and another 20% writing grant applications...)
A lot of the stuff that did fail is just not sexy enough for politics to care. Stuff that might be handled by MY's "secret congress" but even there not a huge priority.
It seems like the incentives in American democracy don't actually work toward effective government on the national level. If your local city council can't keep the streets clean, they might get voted out, but very rarely will people vote for a different congressperson or president based on how competent the FDA is.
"SD, $5,000, to fund an honors' thesis on neutrino research. S is an undergraduate who wants to work on neutrino physics with one of his professors, but needs outside funding to be sure it will work."
SD, if you're out there and want any outside help / input / collaboration, please feel free to reach out. I'm a particle physics postdoc (mostly working on dark matter), know a thing or two about neutrinos, and think your idea is super interesting and useful. If nothing else I can connect you with people working in neutrino physics who may be more helpful than me. You can reach me at
SD here! I've emailed you - thanks for the help! I always, always appreciate advice from people who have been doing this for longer than I have.
If anyone else has any input or just wants to talk about the project, feel free to email me at sd10(at)williams(dot)edu (maybe should have asked Scott to put this in the main post but I figure I'll do the self-promo thing at some point closer to the time I begin my work, which should do the trick).
Yeah, I actually mentioned WATCHMAN (and Nucifer, although I don't think they take Americans) in my grant application. I want to join them ASAP or at least work in parallel with them. Thanks!
Well done Scott. It doesn’t look like being “outed” by the NYT was a bad thing in the end. I hope to do some small contribution. I’ll also sign up here.
Thank you for this initiative, Scott. It has surfaced many worthwhile endeavors. It is admirable that you took on all the related effort, with the help of many supporters and funders.
> if they're smart enough to attempt this project, they're smart enough to know about XYZ Grants which is better suited for them (...)
This sounds like the grant reviewers assume that "likely to succeed at project" is very strongly correlated with "able to navigate searching for funding". For some of the projects that got awarded this sounds eminently sensible to me (e.g. the affect public policy things), but not for others (e.g. most of biology-adjacent projects). I wonder if I'm wrong about the latter (i.e. if success at projects that ostensibly are not about influencing people is strongly correlated with ability to navigate social mazes). Thoughts?
I think the grant reviewers already took that into account because they're domain specialists, judging by (for instance) the remark for Todhunter's grant: "I'm not sure anyone will use this, except for me personally I WOULD LOVE THIS SO MUCH", seconded by commenter and biologist Ivan Fyodorovich above: "I want to strongly endorse the cell culture media testing project. This is the exact kind of thing that is 1) really important and 2) in a kind of hole where it's not the kind of thing any science funding agency would like to fund"
The bigger problem I have with this line of thinking is the assumption that
> if XYZ doesn't fund these people then that's strong evidence that they shouldn't be funded
If Scott doesn't fund an applicant, is that strong evidence they shouldn't be funded? It seems obvious the answer is no given the limited amount of people he can fund and the large number of applicants. Why should this not be true for other grantmakers? The assumption here seems to be some variant of the just world fallacy where P(is funded | should be funded) = 1, but it seems more likely that this number is quite low.
I realized this actually makes sense IF it means "since the applications get passed to ACXG+ anyway, let's pick up the ones they're unlikely to fund", but otherwise I agree it's weird (see my top-level comment below)
Beetles which can biodegrade plastic strike me as worryingly double-edged. Plus side, plastic not hanging out in the environment forever is probably better for us, and for the global ecosystem at large. Minus side, aren't there things made out of plastic where its non-biodegradingness is essential to their function? Is anyone in charge of knowing the full list of things which might fail catastrophically if plastic-biodegrading beetles were released into the environment?
My thoughts exactly. Termites (wood eating insects) are a huge problem in some parts of the world. It would be awful if similar issues arose with plastic.
We could always go full on science apocalypse and have an island where we ship plastics loaded with beetles that eventually results in our destruction when they learn to cover the distance to the mainland, possibly through micro-evolution. There are worse apocalyptic scenarios. This one seems almost fun by comparison. Michael Crichton would have to be simulated on a super-computer to write the novel.
According to the application, the beetles can only get a small amount of their subsistence from plastic and wouldn't be very good at this in any case. What the researcher actually wants is the gut microbes, with the hopes that once we understand them better we can figure out some way to use them at scale. Breeding the beetles in a plastic-enriched environment is a means to get better microbes. The microbes presumably can't escape the beetles without human help. At least this is my current understanding.
I can't help side-eyeing the "presumably" there. If plastic-biodegrading bacteria gained the ability to survive independently of the beetles, there's an awfully rich ecological niche waiting for them out there.
This is more worrisome than what I originally thought: first I thought, "Well, if we need boats or landfill liners or nuclear waste containers or whatever to survive the beetles, I guess the beetles can't eat them without oxygen, and we can always lace them with insecticides." I don't insecticides that will work that well with bacteria, and gut bacteria are presumably already anaerobic. So maybe we'll have to resort to teflon or geopolymers or something in order to build bacteria-proof things in the Grim Beetle Future.
On the plus side, the bacteria probably smell better than darkling beetles do.
(Just to be clear, I think the research should definitely be done; I'm just not sure whether deploying the results in the wild would be good or bad.)
I mean, anaerobic bacteria TYPICALLY are outcompeted by aerobic bacteria in situations with plenty of oxygen. Specialization is a big deal for bacteria, so I would expect it to be much much harder to get gut bacteria to thrive in a place they could digest plastic for us than it is to make efficient plastic-eating gut bacteria in the first place.
I don't understand, isn't this just a much *better* outcome than the one projected in the grant? It seems increasingly clear that plastic was a terrible mistake from the get-go, and if Scott's grant intervention leads ultimately to a world where we have to go back to glass, steel, ceramics and tin with a small dash of bakelite (a resin) for flavor because any plastic will just get devoured by omnipresent microbes before you have time to use it, that will be a massive improvement on all fronts.
My prior is that plastics are an integral part of modern life in such a way that we couldn't have much of what we have *at the price we have it for*. In other words, much of the world would be much worse off without the existence of plastics. The "terrible mistake" seems to be not using plastics judiciously, not that plastics exist.
That being said, I have no special insight or knowledge and could be convinced otherwise.
"My prior is that plastics are an integral part of modern life in such a way that we couldn't have much of what we have *at the price we have it for*."
I think this is almost certainly correct (notably, every post-celluloid data storage medium including film that doesn't burn your house down is plastic-based), and I'm willing to bite that bullet. To me, plastic seems like our time's equivalent of asbestos. It's no good telling people how practical it is that once your gloves get dirty you can just throw them into the fire and they're sparkling clean again.
Non-OBSF*: "The Great Fog" by H. F. Heard, in which a very fast mildew takes over the world and rots everything that can rot. There's a fog which makes smoke stay and be intolerable. No wood, no paper... as I recall, humanity is limited to using glass and stone. Fortunately, the fog moderates temperatures and produces sufficient stuff that's good to eat.
* In rec.arts.sf.written, it was a custom to add OBSF (obligatory sf) to posts that weren't about sf.
More that I'm pointing out that this is science fiction, even though I'm not obligated to include it. I tend to see the world through science fiction.
I reread the story. My description is mostly correct. except that I hadn't noticed that the climate-changing mold was the result of an effort to produce a fat-producing mold. Also, there was a big die-off of the people who couldn't handle living in a very humid environment. It was published in1943, so there was no notion of a high proportion of people dependent on medicine.
Well, among other things, no hip replacements, no arterial grafts, no cardiac catheterization to stop heart attacks, no single-use medical instruments so loads more iatrogenic infection and death from same.
The result of the great fog was presented as utopian, and it might actually be a net gain.
People couldn't see to go faster than four miles an hour, though I think Heard was underestimating how well people could go with cleared, familiar paths. It would still be running, at most. Horses couldn't see as well as people.
No war, no car accidents, nothing much to accumulate. Fine stone musical instruments might be about it for valuable items.
Note that this was written in the 1940s, so most of what you're mentioning didn't exist yet. If people are moving more slowly and not using a lot of edged tools, there would be fewer accidents of all sorts.
They live pretty much on food produced by the mold, so a lot of the less healthy food isn't there, and the premise seems to imply that the moss food is at least nutritionally adequate.
Any and all plastic could be totally re-used if only we chose to invest a little bit of energy. It isn't currently economically cost-effective to do so, but if we really wanted to get rid of plastic we could. We just lack the will.
Where can I read more about this? I was under the, likely mistaken, impression that we know how to reuse some plastics, but for vast majority of them once they are polymerized we don't have better ideas than "burn them at high temperature to recover simple compounds".
So warning that I am not a chemist, so I probably don't have all the details quite correct, and am just giving my semi-educated opinion. Burn them at high temperature is basically what I am talking about, but why is that such a problem? Well, because it ends up costing energy.
The way we make plastics now is we start with petroleum, which contains lots of molecules that contain more than three carbon atoms, break them down into mostly molecules that contain less than three (ok, we keep a lot of hex rings too) then re-combine those smaller molecules into large chains of polymer. There is no particular reason we can't start with carbon-containing trash, and use similar processes as we use to crack/refine oil to turn it into the same chemical feed-stocks we use to make plastic.
The reason we don't do this is energy. Oil is actually insanely cheap. I know Americans like to complain when gas costs more than $4 a gallon but. $75 for a 55 gallon drum of oil is extremely cheap. Do you know how much a 55 gallon drum it? It's like the size of a person. If you could buy jugs of oil in the supermarket it would be cheaper than almost anything else, except maybe bottled water.
Because is so cheap, oil refineries burn a substantial amount of their cheapest products to generate the energy required to form their more expensive products that are used to make things like plastic. We can't really do this with garbage because it would generate tons of pollution and garbage isn't as energy dense as petroleum. But if we had much more abundant, cheap energy we could. We wouldn't even need to burn the garbage, we could heat it in a kiln to recover and re-use essentially all of the carbon and other toxins.
I think you can get net energy out by burning plastic... (Hasn't everyone burned plastic by the camp fire?) I'd like to see plastics made that burn better. (Less crap in exhaust 'smoke') Think of plastics as a side step between oil and energy. We make all plastic knowing they will be burned. Or alternative 'green' idea. Keep putting them in land fills, but change the narrative to permeant carbon storage. Burn them or bury them... that's all I got. Putting 'em back into the ecosystem seems to be the worst thing to do.
Re Beetles: Why not termites? They already have the gut bacteria to digest wood.
Calling burying plastic permanent carbon storage seems pretty disingenuous - the carbon in it came from oil, and only helped us in its use as plastic. Burying it instead of burning it is the same idea as leaving oil in the ground.
It's a little more complicated than using high temperature, because many plastics contain heteroatoms (e.g. Cl, S,N) that can create noxious compounds (e.g. HCl, SO2, NOx) even at very high temperatures. So you need to add some fairly expensive flue-gas capture technology to pull out the compounds you don't want in the air -- and then find a way to dispose of them, or someone who wants to buy them. (It goes without saying that low-temperature combustion generates even more evil compounds, PAH and such.)
Really? You know how to remove nurdles from oceans, beaches and the innards of fish? Harmlessly remove microplastics from fetuses? I think you should have submitted an ACX grant application if this is true.
I think GP is talking about the problem of "I have a bucket of plastic trash and want to convert it into usable plastic of the shape I desire" and not "I have a lake that contains a bucketful of plastic in it and want to extract it".
I don't know who you mean by GP here, but I think we must be talking past one another. Let's recap: I asserted that a bacterium which destroys all plastic being released into the wild would be a net good despite the infrastructural damage it would cause, due to the fact that the bacteria would also obliterate all the horrible shit like oceanic nurdles and microplastics presently plaguing ourselves and the rest of the planet. Eric P. replied to this statement of mine with the claim that we can already recover any and all plastic. I replied to that claim in turn, expressing a skepticism which I assure you is very real and sincere. In other words, we're talking about something closer to your scenario #2 here.
I'm still under the impression that Eric meant #1 and not #2 (nonwithstanding the context). I eagerly await Eric's confirmation, refutal, and/or elaboration.
E: By GP I meant Eric's comment (grandparent of my comment).
Yeah, we were talking past each other. I don't have a solution for removing it from the environment, only re-using it industrially.
But while I hate plastic in the environment as much as the next person, I also don't think it is the most important environmental problem to solve. Micro-plastics are disgusting and surely harmful, but I don't believe they are so harmful that we should regret having ever used plastic. It will all eventually break down and settle into sediments. It will take hundreds of years, but the earth is old and life will go on.
The best way to remove plastic from the environment is to stop putting it there, and in my opinion the best way to do that is to develop a robust industrial system for re-processing it.
There's a lot of different types of plastic. I'd be surprised if there's not some high-grade polymer like UHMWPE or something with a bunch of fluorines that the beetles can't get at.
Would it be possible to engineer bacteria so that they were good at digesting micro-plastic, but not strong enough to pull bits off of plastic objects? Or at least cause deterioration very slowly?
I'm not an expert, but based on my limited knowledge of the subject, I don't think that's likely to be practical, because on the scale of a bacterium, basically any piece of plastic represents a large ragged surface. Even microplastics are much too large for bacteria to subsume, they'd still be attacking them in the same way that they would a larger piece. The more weathered the piece of plastic, the higher its surface area to volume ratio, so the more easily it ought to be attacked, but any plastic should be vulnerable to weathering.
Also, once eating plastic becomes a viable ecological niche, any bacteria which is better at it will tend to outcompete ones which are worse at it.
This is amazing. I mostly write to praise Scott and the team. But can't help but to contrast with the many philanthropies I know well....such exciting ideas/people he has curated, in such a short period, with so little red tape.
For peace of mind in the future: a way to avoid mistakes in the worse direction is to compile a list of terms that must not appear in a post (in this case, names of people who want to remain anonymous), and arrange for automation to scream loudly if a draft contains any of them. (Not my idea, this is a pretty common approach to ensuring that various not-intended-to-be-made-public things were redacted from a release of some software.)
This is just incredibly impressive that you pulled this venture out of thin air, roped in all the support and expertise you needed (or could find), let it grow way beyond what you anticipated, and then pulled off this very wide ranging list of worthy projects to support. Wow. It warms my heart to see.
The first part of my career involved a ton of fundraising, in and out of academia, and working with foundations on various joint endeavors. It can be a pretty slow-moving, trend-following world. But also, some of the expertise (to raise money well and to give it away well) are for sure real skills. It's astounding to me that you pulled this off in such a short time, and I hope you will write more about the process of it.
I'm interested in the question of what funded groups might want to talk to each other about. I know a lot of foundations try to start up those kinds of cross-pollinating conversations and that on the side of the people funded, they sometimes say "yes" to whatever the funders propose in the hopes it will lead to more funding, even if they think it's a waste of time. I think funders get impatient encountering the same kinds of organizational stumbling blocks in the groups they fund (I consult for organizations so I get that too). Some kind of honest assessment of when those networking opportunities are seen to be beneficial by the people in them would be nice to have.
Thank you for helping to bring so much good stuff into the world! I hope beyond the exhaustion it sounds like this produced, that you have a real sense of satisfaction in it. It's a remarkable thing you did, in a whole different direction from your writing and your psychiatry practice.
"Why would somebody working on biochemistry want to talk to someone working on political activism just because they got a grant from the same person?"
Going to conferences of people who all think and dress the same way and work on the same issue is beneficial in one way; going to conferences like EA or SciFoo is beneficial in a different way--optimizing for serendipity.
Yeah. People who received an ACX grant probably have many more things in common than two randomly selected people (even if we restrict the selection to compatriots). But also, sometimes just having common knowledge that everyone in a certain group actively wants to meet and chat with everyone else is already valuable.
It would have been useful to know that one option was funding, of some sort, for politics/political careers, not necessarily for me personally. There are some interesting Senate seats up this cycle that might be viable options for third party candidacy who knows what they are doing where the existing brand name options are mediocre or in conflict with the kinds of things you, Scott, care about.
There are, given the divergence in my interests vs Scott, more very exciting projects on this list than I would have expected. Though part of that may be all the extra money and effort many people contributed.
I'll also somewhat second the comment by someone that people involved in grants perhaps associate being able to get grant funding too much with having good ideas, as far as the "why are they asking you and not this other potential option for money".
I hesitate to ask because it seems like surely you've thought of this, but....whatever tax status your grant fund has, have you evaluated if giving to politicians impacts it?
Double-OF-COURSE-YOU-RESEARCHED-THIS-energy: do the laws in Australia allow you to give large sums to political actors, and the actor to accept it? I do note that the person is not running for office themselves, but, ya know.
The grants are coming from two pots: my money, and external funders' money.
External funders are donating through a tax-deductible EA Fund.
I can't do this for tax reasons, so my donations won't be inherently tax-deductible. I plan to cover a combination of whatever grants that the EA Fund can't, plus the recipients who are inherently tax-deductible themselves (eg charities).
Answer to the second part: yes, Australian laws do allow it.
It’s possible I’ll need to register my activities under the Foreign Influence Transparency Scheme (https://www.ag.gov.au/integrity/foreign-influence-transparency-scheme) but this doesn’t prohibit any such activities, just makes it public where the money is coming from. And clearly there is no difficulty with that being public in this case.
Very glad to see approval voting (AV) get funded (and promoted) on ACX. We see ranked choice voting (RCV) making inroads around the country, but there seems to be very little awareness of the pathologies of RCV that make it inferior to AV. [This is based on my remembrance of analysis on RangeVoting.org, which smells trustworthy to me.]
Approval voting is probably the best option, that could reasonably happen anyways, for American elections. Something of a limit to the problems that can be solved solely by better voting but at least at the municipal level it is pretty big.
The pathologies of ranked choice voting are that in certain weird theoretical combinations of circumstances, voting for your true preferences makes your preferences less likely to be satisfied.
The problems with Approval Voting are less obvious, because it's Never Been Tried (actually it has been tried, in a few low-stakes elections but not much that we'd learn from). The upside of Approval Voting is that it allows a charismatic centrist to be elected... the downside might be that it precludes anyone else from ever getting elected. In a world where Approval Voting was the norm, what kind of candidates would we get? Would it be a "race to the centre", where candidates work hard to avoid ever unambiguously taking any kind of position that anyone at all could possibly disagree with? And would this be better or worse than what we have now?
>The pathologies of ranked choice voting are that in certain weird theoretical combinations of circumstances, voting for your true preferences makes your preferences less likely to be satisfied.
By Arrow, is that not true for all (non-dictatorial) voting systems, including approval voting?
I think you're correct. The way you can kinda claim otherwise for approval voting is by changing the model of voter preferences (instead of them having preference orders, claim that they have sets of acceptable candidates).
Also: a nit: Arrow also requires determinism, but all the systems we talk about are deterministic.
(Trivial counterexample of a nondeterministic system: choose a voter at random and do as they wish. It's not dictatorial, because the voter is chosen at random.)
Yes, it is somewhat true for all (though Arrow iirc only applies to ordinal voting methods, which AV is not). But for AV, approving a candidate will never help a candidate you did not approve (though it may hurt a different candidate you approve of more). With IRV, your true preferences can potentially elect your least liked candidate.
In a highly polarized environment, you may have a candidate who is everyone’s second choice but no one’s first choice. They get eliminated in the first round of evaluation in the RCV algorithm.
RCV still has the benefit over our current system (PV) that the spoiler effect is avoided. So those third parties can exist, but they still need to beat the conventional parties in popularity.
With AV on the other hand, you're not forced to rank, only to approve, and so any acceptable compromiser has a huge advantage in an otherwise polarized election.
I'm not sure IRV completely avoids the spoiler effect.
With plurality, if a 3rd candidate is at 1-2%, it is not a big deal, but as soon as it is 5-10% it becomes a liability because of being a spoiler. Thus it is hard for them to grow large support. As soon as they start to grow, they become a liability and voters will strategically vote against them.
IRV does handle this better at low ranges of support. A candidate with low enough support to place last won't be a spoiler, but there is still a center squeeze and other cases where there could be a spoiler. This is because IRV does not satisfy Independence of Irrelevant Alternatives (https://en.wikipedia.org/wiki/Independence_of_irrelevant_alternatives)
I mean, doesn't it? in the limit where all the Irrelevant alternatives are truly identical, all their preferences will flow to one another and one of them at random will get all of the votes of the collective bloc of identical candidates. In the real world, of course, the candidates aren't truly identical, but then it is no longer obviously bad that eg. the votes for the Communist don't all preference the Socialist when the Communist party is eliminated.
Note that irrelevant does not mean identical, though it is true that a system that does not satisfy IIA will be subject to 'candidate cloning'. So your intuition is correct, IRV does a much better job at candidate cloning than plurality, but it still fails IIA.
In the communist/socialist example, you are correct it wont be an issue *if* the communist support is relatively low (i assume you are assuming this, since you say they will be eliminated). If their popularity grows, there is a potential for a center squeeze.
Consider this toy example (numbers are completely made up by me to illustrate. I am not implying any political preference here, just following the example provided)
Say there is an IRV election between Socialists and Capitalists with the following support:
Socialist: 55%
Capitalist: 45%
But what if the Communist party decided to enter... if they only have 1-5% support, their votes transfer to second choices, which I think is fair to say will be Socialist, so no difference in outcome, as we noted above.
But say instead, they drew a lot of support, and we had
Communist: 30%
Socialist: 25%
Capitalist: 45%
Now it is the socialists that will be eliminated first. And unlike the communists, where we could be confident that all of the 2nd place support would flow to Socialists, the Socialists 2nd place is more likely to be split. If just 6% of voters were socialist, but not full communist, and put Capitalist as 2nd place, then in the 2nd round we get
Communist: 49%
Capitalist: 51%
This is IIA failing (also 'center' squeeze' with regrads to IRV). By entering the race, the Communist party caused the winner to go from Socialist to Capitalist. This is a worse outcome for them.
Which suggests that approval voting has more problems than ranked voting in getting your “first choice” in, although AV does away with rankings which is an odd way of fixing that problem - pretending it doesn’t exist.
If that is even the problem (I’m not clear what your pathologies are).
As someone living in a country with ranked voting, specifically the single transferable vote, I would often vote for a first ranked candidate destined to lose. Then on the elimination of that candidate my vote goes to the next person. That person might also be destined to lose so the vote percolates up to elect somebody eventually if I kept ranking.
That’s pretty badly written, and it seeks to have handpicked an example where there might be a discrepancy. In order to prove the point the piece has the voter act suspiciously against her wishes.
“In the example, suppose the 1st voter, instead of honestly stating her top-preference was A, were to dishonestly vote C>A>B, i.e. pretending great love for her truly most-hated candidate C, and pretending a lack of affection for her true favorite A.”
There’s no explanation of why she did this, but apparently the voter doesn’t get A elected if she prefers C when voting. Why is that a problem and why are we second guessing her?
And I think it’s pretty clear that most people, if they get this system at all will use the highest rank available to them, 99 then 98 etc. Which is basically ranked voting.
i agree the article is pretty dense, and not very user friendly. There are more accessible writeups out there if you are interested.
But read the example again! The voter causes 'A' to win, not 'C'. By lying and claiming 'C' was her favorite, this causes 'A' to win.
To the 2nd point, "if they get this system at all will use the highest rank available to them, 99 then 98 etc. Which is basically ranked voting."
It is still different! They have implicitly 'ranked' their preferences; similar perhaps to IRV, but the way Score voting determines the winner does not make this equivalent.
And I don't think it is clear people will use the highest ranks. They will use the highest AND lowest (since they don't want the candidates they hate to be ranked high). And this.. is basically Approval Voting.
The voter didn’t lie. She voted as she voted. Why is the writer assuming somebodies internal mind choice as being illegitimate and why is she ever going to with out prior to the count what might have happen. It’s thoroughly contrived.
Of course most people will rank 99,98,97, if a die hard Republican doesn’t do that for the Republican candidates (and I assume there would be multiple candidates from each party) then she will worry about a democrat would rank the top Democrats 99,98,97.
1. I'm not sure I understand your objection... It is pretty standard for analysis of voting methods to consider strategic voting, this is not unique to this article. If you assume no strategic voting, that makes range even better.
If it is just a general 'this is contrived' objection, then yes I don't think anyone would disagree. Contrived, or 'toy', examples are often used to illustrate how a voting method can go wrong, as they are simpler, clearer and cleaner than a more complex example.
2. Yes, the die hard 'R' can rate Republicans '99,98,97...', or can rate them all '99' (you can reuse scores). But they wouldn't rate the Democrats this way. They would give them scores closer to 0. It would be foolish to leave them unranked.
I suppose it comes down to what our objective function is. I’d rather have an official of whom 80% of the population approves than have a candidate who is 53% of the countries first choice.
You make a good point about a strategy of priority ranking candidates unlikely to win, but we face an education barrier to get most of the electorate onboard. I suppose a similar educational barrier exists for AV as well. Voters may still adapt suboptimal strategies.
It could be interesting to write down some objective functions and voter behavior rules and then run simulations to see what would happen. Obviously lots of limitations to modeling, but it could fortify the discussion.
I agree approval voting (AV) is better than plurality voting (PV). Another system that could be beneficial in the current situation is approval/disapproval voting (ADV). You could vote for or against any number of candidates. ("I am for Alice, Bob and Carol, neutral about Dave and Fred, and against Elisabeth"). Where AV is giving [0,1] to each candidate, ADV would be giving [-1,0,1] to each.
AV's advantage indeed is to allow support of outsider candidates without "wasting the vote". ADV goes beyond that, it is the antidote to polarization - it would *actively punish* the "Shiri's Scissors candidates".
The said candidates utilize the scissors as a cheap way of mobilizing a voter base; whether it turns out on election day that they picked the "50%+" or the "50%-" side of the scissors, they have *certainly* swept aside their non-scissor competitors. In terms of ADV, a scissors candidate A would cut the voters pro:neutral:contra as 51:8:41 - often against a candidate B taking the opposite scissor position, with near-opposite ratios. Until the backlash of the next election, that is, where it will be 42:7:51. (it's about topics such as "defund the police" / "prohibit abortion" / etc). Main thing is, both A and B are sweeping aside their "non-scissors" competitors C to H, who'd be around 20:75:5, give or take, and who are getting virtually no votes in PV, and doing only slightly better in AV.
In the PV system, the A and B end up sucking out all the air in the room as far as C to H are concerned. The AV system gives *some* breathing room to the C-H candidates, yes. But the ADV system *punishes* the A & B candidates. They get "51-41=10" versus their "20-5=15" contenders.
Such a system would carry an obvious bias: against the "far positions" on all sides. It would have centrism baked into the cake. Some cite this as a flaw, that it would lead to electing candidates who aren't "anyone's favorite choice" or whatever. One good look at contemporary US proves that to not be a flaw, but as good an outcome as might be hoped for. Nowadays, Rep voters have mixed feelings about Rep candidates and hate the guts of Dem candidates, while Dem voters are in the exact opposite situation. This is the natural outcome of the PV process. The AV would slightly improve this, but the ADV has the potential to replace outcomes of "50%+ mixed-feelings / 50%- hating-guts" with "10% approval / 80% mixed feelings / 10% hating-guts". Arguably the latter *is* a better outcome for democratic governance, or at least a saner outcome for something that aspires to be a unified country.
I don’t see how any of this works without multiple candidates within parties, which can happen with the single transferable vote. That would bypass the primaries.
Using approval voting for primaries but plurality voting for general elections is an interesting case. The main benefit of plurality voting is that it avoids fracturing of parties; systems in which independence of irrelevant alternatives hold tend to have many parties which then can require large and unstable coalitions. Party primaries are already rather fractured. This also avoids a complete "race to the center" that could reduce political innovation.
In general work on how one might combine different voting systems for different purposes seems underdeveloped.
You appear to paint "race to the center" negatively, and "political innovation" positively. You also placed scare quotes around the former only, while I'm inclined to place them around the latter.
What "innovation"? I consider there is hardly any dynamic in contemporary US politics that isn't a rehash of the dynamics happening in end-of-18th-century France, for instance. (if excluding how unprecedentedly blockheaded most participants are around race questions, I would argue there is absolutely nothing new at all). People more versed in history might cite echoes of those dynamics going back to ancient Rome, Greece or even furhter.
"Fracturing of parties" is an illusion created by the perspective that there *should* be a duopoly of parties, rather than the perspective that the duopoly is the degenerate endgame of the plurality voting system. The PV system has the *rewarding* of scissor candidates baked into the cake, while the approval/disapproval voting system has the *punishing* of scissor candidates baked in. Yes, "centrism", if you prefer.
Democracy often labels itself as "rule with the approval of the governed". Running with that label, consider two systems who would give the following typical *outcomes* of the process.
Which of the two comes closer to the label of "rule with the approval of the governed"?
Also, if contemporary US is flip-flopping between "Trump US" and "Biden US" - both of them being 10:50:40 arrangements with flipping polarities - which of the two approaches would you label "stable" in a meaningful way?
I used actual quotes because I wanted to be clear I was referencing the concept in the previous comment. I don't do scare quotes.
I don't think any realistic voting system will come anywhere close to "rule of the governed." People have too varied of preferences. I am therefore generally extremely skeptical of government intervention, which I view as being nearly always and everywhere captured by special interests. That doesn't mean government should do nothing, but I don't see being elected as conferring a large amount of legitimacy. I would not be particularly supportive of democracy except it seems considerably more robust to providing for the general welfare than any other form of government, even if it doesn't do a great job in an absolute sense.
By political innovation I mean primarily policy innovation. If everyone is competing to be the least offensive, that to me suggests that they will avoid confronting entrenched special interests. That seems undesirable. I could care less about grand debates in political philosophy.
One also need only observe Italian, Brazilian, Israeli, or Belgian politics to see the impact of a high degree of fracturing.
I don't want to overly focus on the current political dynamic in the US. Trump wouldn't have been nominated with approval voting in a Republican primary.
Understood, and retracting my comments about scare quotes.
>Trump wouldn't have been nominated with approval voting in a Republican primary.
Correct. Neither would have been Hillary. I do agree AV is a step up from PV, I'm arguing why I consider ADV to be a step up from AV.
As for the entire US system having degenerated to "PV duopoly with PV primaries" in the first place, and that the only "tweak" it needs is to become a "PV duopoly with AV primaries" - because this avoids "party fracturing hell" - I remain skeptical of that view. I consider that having a multitude of parties that are elected by ADV on the general level, would reduce the creeping and seemingly unstoppable polarization of the country.
The examples of fracturing you gave *are* fracturings that evolved inside PV systems (except that they never spiraled down into a duopoly-with-primaries configuration). The ADV system would inherently affect the incentives in which any fracturings are evolving, so the comparisons need not directly apply.
About coming anywhere close to "rule with consent of the governed", if your key metric is the share of the population who does *not* feel unmitigated discontent with the way they are governed, then voting systems that punish scissor candidates provide a doubtlessly better approach than systems which reward them.
> democracy (...) seems considerably more robust to providing for the general welfare than any other form of government
In Christian and Islamic countries throughout past centuries, welfare has generally been organized by religious institutions more than by governments. While this has necessarily waned alongside the large-scale secularization trends - which largely coincide with establishment of democracies - equating government-provided welfare with democracy is confounding issues. I've been to some Islamic non-democratic countries that don't have -any- homeless people sleeping on the benches, and whose welfare stems from religious tradition more than from government-organized redirection of funds. Also, analysts like Sowell can point you to the reasons why the "blank-check government-provided welfare" that is de rigueur since the 1960s, has not really been an unalloyed good for the population receiving them, when compared to other historic welfare arrangements.
Gašo, would it be correct to say that, structurally, RV(2) == AV, and RV(3) == ADV? Here RV(n) refers to Range Voting with a range of n different options.
I believe your ADV proposal may be practically superior to RV(n) because Approval / Disapproval is easier to communicate than the Olympic-style scoring used in RV.
They would be functionally equivalent, yes, but I think that presenting it as zero-centered (i.e. [-1,0,+1] instead of [0,1,2])is better communicating to people the effect of their vote.
In your notation, the Score Voting system which proposes 0 "strongly disapproving" to 9 "strongly approving" plus an "X" for "no opinion", can be represented as "RV(11)"; its functionally equivalent zero-centered system would be [-5..+5], with mappings 0 → -5, 9 → +5 and "X" → 0. Whereas I'd still prefer the ADV ("zero-centered RV(3)") to the RV(11), as it has the advantages of simplicity; it is the simplest of the systems that can counteract the arguably largest pathology of PV systems, "VOTING AGAINST".
By this I mean voting for B because you *really* do not want A to win, and *not* because B is particularly well representing you - whereas voting for D or F (whom you're both reasonably aligned with) is pragmatically a wasted vote that will just let A win. This has been the most pathological dynamic for over a hundred years.
In ADV, if you're really against A, you can simply *vote against A*, and then vote for D and F. One could do a book-length presentation of major historical moments that would have gone quite differently under such a voting system, but the single-sentence summary is that it punishes scissor candidates instead of rewarding them.
Such an exciting list! And I was very happy to see Seattle Approves up there -- they're addressing the meta problem of "why can't the US political process solve more of the country's problems?"
"why can't the US political process solve more of the country's problems?"
Because of human nature. There is no one single utopian perfect nation in the world, even with a different political system. You can have a *bad* system that causes problems or make things worse, but you can also have a perfect, shiny system that people will then try to bend to their own ends or ignore.
Improving bad systems is a good idea, but don't think "we'll create a perfect system and that will be the panacea for the nation's problems" because it won't be.
Some really fascinating grants here (and a couple that I think might be ethically... questionable). Setting those aside, this is a really cool thing you've done here.
> Nils Kraus, $40,000, to experiment with new ways of measuring precision weighting in humans. The precision-weighting of mental predictions is one of the absolute basics of the predictive coding model of the mind, but we know very little about it and have trouble testing hypotheses about how it works.
I'm quite interested in this; is there a way to follow this work as it develops? (And is there a good source on the current state of understanding?)
I can send you the proposal where I have collected some previous studies that have tried to measure precision weighting via several different ways, if you'd like.
I'm one of the three cofounders of Mantic Markets, and I wanted to give a heartfelt thanks to Scott! Not just because we won a grant, or because he's namedropped us, or even because he's been so nice about us blatantly stealing the "Mantic" name...
But because it's fairly (~70%) likely this project would not have existed without ACX Grants. James and I were online friends, and we'd chatted before about collaborating on something. The existence of ACX Grants gave us an excuse to actually do something! I flew out to meet James and Stephen; and in our process of writing the grant application, we got so excited that we decided to just build out the prototype, whether we heard back or not. We've been hacking on it ever since (that's why our prototype is already live)!
Like Scott, I don't really know if Mantic will ultimately work out. (If you think _you_ know, bet on https://mantic.markets/ManticMarkets/will-mantic-markets-have-over-1m !) But this grant has already made me two good friends/cofounders, and a really exciting December so far. Thanks again for running this program!
Re: "Why would somebody working on biochemistry want to talk to someone working on political activism just because they got a grant from the same person?"
I can't speak for anyone else but I'd _love_ to be thrown into a Discord with everyone else who's gotten a grant. Beyond chatting, here are some things I'd look forward to (patterned off of YCombinator):
- Random 1:1s with other fundees, just to learn about their field of expertise
- Weekly check ins where everyone posts updates, as a shared accountability metric
- A "Demo Day", eg in 6 months where each fundee gives a 3min talk on what they accomplished, in front of potential future funders
Everything optional, of course! I'm happy to help organize; in fact, I may just go and do this for the forecasting people. We've already chatted with Nuno before, and I'm planning on reaching out to Nikos and Nathan too!
I winced. It’s certainly a joke, but it’s hard to make something like that land in our environment when so many people treat all Republicans as evil.
I chalk this up to the influence of living in the Bay Area and among his tribe, where casual insults to Republicans come off as harmless political humor that nobody is hurt by, because no actual Republicans would dare out themselves by objecting.
"Other times a reviewer was concerned that if you were successful, your work might be used by terrorists / dictators / AI capabilities researchers / Republicans and cause damage in ways you couldn't foresee"
I agree that this is a joke, but to my reading it is a joke that gently pokes fun at the reviewer (who seemingly equates Republicans with terrorists). If anything, it reads to me as sympathetic to Republicans, not pointed at them.
Maybe it is! The problem is that it can be read both ways, and since the double meaning is implicit and there’s nothing to read against the equivalency, it’s tone-deaf in an environment where just this kind of casual associational slur is terrifically common.
1Day Sooner staff member here—thanks so much, Scott! These grants are an awesome idea, and honestly I’m kind of blown away by the company we’re in. Good luck with your projects, everyone!
> My father has been stalled on an important research project for years for lack of the right kind of statistician
Just out of curiosity, what is it and what would the right kind of statistician be? I'm a statistician and I know a lot of statisticians. Often we feel we are short on important projects, while society is yelling that it is short on statisticians. It's hard to square that circle.
I am a statistician who meets the latter requirement. I have done biostats, though I don't know that I'm A Biostatistician. Is he still looking for help?
Edit: I guess there's enough grading in level of access that my comment might not be super-helpful. Does he need someone *at* the VA who can query their database directly, or somebody who's clear to stare at such data?
Thanks for the reply - I think for now I've got enough on my plate that I probably can't squeeze in another project without a sprinkle of the "Great Family" glamour. ;)
Thank you Scott at everyone who contributed to this new grant. I'm honored to have RaDVaC included in this group of creative, important, really interesting projects. This will help take our work at RaDVaC to new levels; in my view this is a big moment for vaccinology and the future of global health. We're going to do some kickass science.
This legitimately made me cry. I'm writing this with tears coming down my face right now, because this is the group of people that are going to save the world.
I just want to state my most heartfelt gratitude towards Scott and the other funders for their munificent generosity. Thank you *so much* especially for supporting the www.endohazard.org project, which I strongly believe can lead to huge beneficial second and third order effects. Thank you!
That symbol has an interesting angry face appearance. Unsure if intended, and also unsure if beneficial; certainly amusing, and memorable. It seems, complicated? I wouldn't remember how to make it, although I'd certainly recognize it. Of course, I'm not exactly the sort of person who'd be applying this to things.
Regardless, congrats on the grant, and thank you for your work on this!
Thank you! Yes, it's a stylized fusion of BPA atop Dioxin – two molecules that have particularly strong and widespread endocrine disrupting properties. The 'face' is designed to create an instinctive feeling of danger. Even if one isn't aware of its meaning on the side of a barrel or component, it's intended to clearly convey 'bad mojo'.
The spike elements are inspired by the biohazard symbol, which has a similar instinctive effect. I agree that it could be simplified, and may well be in due course as the design is further refined. Thank you for your kind comment, and the insightful feedback! :)
"These elements together form the grotesque Endohazard Wraith, intended to convey an arresting, instinctive sensation of danger and revulsion across all cultures, irrespective of familiarity with the symbol’s meaning."
Yeah, is this intended to be a chemical warning symbol or your new horror video game?
There are a range of symbols in existence already, from the general and simple:
If you're trying to get a new symbol into general use and widely accepted, row back on the breathless scaremongering and be a lot simpler in design and a lot cooler in "Dear regulatory bodies, we would like to suggest this symbol for new purpose" rather than the florid B-list horror movie photos.
The GHS symbols and SDS sheets provide a strong basis for environmental and health protection. This includes the Serious Health Hazard symbol intended for substances that can cause long-term disease.
However, the uniquely insidious impact of endocrine disrupting chemicals warrants its own endohazard label.
EDCs have transgenerational effects that persist long after the original organism has been exposed, and the alterations may be permanent. This is unlike almost any other hazard class, with the potential exception of hard radiation.
Their symptoms can manifest as ordinary degenerative diseases, or things that people don't necessarily consider diseases such as alterations to gender identity, parental instincts, irritability and mood, mental fog, coldness and sluggishness, impaired sexual desire, etc, yet which may have a major impact upon individuals, societies, and ecologies. This is different from an oncogenic substance, where symptoms are overt and a clear differential diagnosis may be made.
EDCs are absolutely everywhere in modern society, and the problem is getting much worse very quickly. By introducing a specific hazard symbol, a justifiable 'badge of shame' can be designated by regulators to products that are injurious, targeting the worst offenders to begin with, and gradually expanding the scope over time. This way, the power of the market can be brought to bear to curtail their usage by innovating to find less hazardous alternatives.
EDCs are poorly understood or recognized, and indeed often appear benign. Their nature as forever chemicals means that we need to consider future civilizations as well as our own. This is why the symbol needs to be emotionally arresting on an instinctive and universal level – danger, stay away!
The design was more simple in the past, but became necessarily more complex when it came to ensuring it could be used as a stencil. If it needs to be simplified again it can be.
I accept your feedback on the photo, it may be counterproductive. As I have stated most of the effects of EDCs are subtle and insidious and are not easily telegraphed in a physical manner. The photo is a genuine example of EDC exposure at high concentrations, and can represent chronic exposure over a lifetime, or indeed across generations.
It's important that the public understand viscerally how much suffering can be caused by exposure to this class of chemicals.
Thank you for sharing your impressions.
EDIT: I have since changed the editorial image to something more generic.
I'm getting the strong sense that I should be scared but I'm very unclear on what I can actually do about this. Labelling would be helpful, yes, but until then — where do I find out about alternative uses? What should I be using my purchasing power for? Who should I be writing letters to? What should be outlawed? Where can I find out more that I can actually do?
Labelling is a beginning. It's a way to at least make people more aware of the issue, to avoid unnecessary risk of contamination, to enable better regulatory practices, and to provide a basis for free market competition to innovate away from the use of this dangerous (yet often very useful) class of chemicals.
Of course, it's possible that BPA is slightly harmful. That would be pretty bad, given how widespread it is.
But merging it in the same category as *dioxins*? Are you crazy? Dioxins are horrible things that I wouldn't touch with a 20-foot pole. BPA is in the canned beans I ate yesterday.
Putting them in the same umbrella seems likely to confuse more than clarify. Your symbol will become the new "known toxic to the State of California" -- a message everyone has learned to ignore, because it is on everything. Which is too bad, because dioxins truly are monsters from the deep.
(You show a photo of a child with chloracne due to dioxin exposure. I note that you don't have anything to show for BPA exposure.)
There are several systematic reviews and metanalyses that illustrate significant effects of BPA on human and non-human life, particularly in early life stages:
However, I agree that the evidence is much weaker than for dioxins, which everyone can agree are nightmare fuel.
I take your point that Dioxin and BPA are radically different the manifestation of their effects. That said, both present serious health challenges with transgenerational effects. BPA is a cousin of DES, a drug that has caused cancers and gender issues in grandchildren of the original ingesters of it.
Both Dioxin and BPA are two sides of the EDC coin – both are forever chemicals that cause damage across generations, whether from Agent Orange or cash register receipts.
BPA is more slow and insidious, easily ignored or mistaken for 'diseases of modern lifestyle', but perhaps just as devastating in the long run due to its ubiquity (~90% of the human population), and the concentration of effects across generations. There is a possibility that EDCs present an existential risk to the future of humanity.
If you have a suggestion for an alternative chemical than BPA, or simply to base it purely on dioxins (as an earlier iteration of the symbol did), I will gladly consider it.
My hope is for the symbol to be primary used to denote the most hazardous kinds of EDCs, though which kind qualify is something I would leave to regulators.
I accept that there may be a risk of public confusion about the intensity of threat, or of the symbol becoming undermined over time by 'crying wolf'. I am considering some kind of gradation or 'traffic light' system to be applied with it. This is why the standardization process is so crucial – to hammer out these kinds of details, and to carefully define the scope of use.
Thank you for your feedback, which I appreciate very much.
Thanks for your response. You link to 3 meta-analyses, 2 of which are about humans. For the human ones, they summarize 16 and 10 studies respectively (presumably with some overlap) and find correlations between BPA exposure and (1) Type 2 diabetes (in the first meta-analysis), (2) obesity (in the second).
Almost all studies included in these meta-analyses are cross-sectional. None of them are randomized controlled studies. What this means is that we know there is a *correlation* between BPA and obesity/diabetes, but this is a far cry from showing causation.
Of course, RCTs are really hard to conduct in such a setting. I'm not really blaming these researchers. What I'm saying instead is that we just don't know. It's *possible* that BPA contributes to obesity and/or diabetes. It's just as possible that it doesn't.
Many, many other things correlate with obesity. Basically, controlling for SES (socioeconomic class) is always imperfectly done, and so SES ends up confounding all correlational health studies. In other words, poor people have higher obesity and higher BPA exposure. You'd see the same result if you studied whether (say) breastfeeding prevents obesity later in life: the studies will all point to "yes", because it's all SES confounded, even after attempts at controlling for SES.
Again, maybe BPA causes obesity. I can't rule it out. I just don't think we know that right now. Perhaps the precautionary principle means we should avoid BPA (but beware its understudied substitutes like BPS and BPF; for all we know, they might be worse). But they are not the same as dioxins.
(The one mouse study you link is perhaps more concerning, but I can't immediately tell how high the dose given to the mice was; I'd guess it's much higher than the typical human exposure levels for BPA.)
It's also worth noting that BPA does not bioaccumulate. You say there's a concentration effect across generations, but I don't know what you mean by that. It's true that we use it more now, but I wouldn't call increased usage a concentration effect. (Dioxins do bioaccumulate.)
I note that Bioaccumulation of BPA (in human beings, rather than the environment) may be underreported, as some research suggests that it can be excreted in sweat but not detected in blood or urine.
By concentration of effects across generations I refer to transgenerational and multigenerational epimutations stemming from EDCs. Transgenerational effects (due to prenatal exposure) from BPA are strongly reported in literature:
The real nightmarish issue is multigenerational effects which compound over generations. This has been documented in human beings, with the grandchildren of women who ingested DES.
This is why I consider EDCs to be a potential existential risk. The long term danger of this class of chemicals to public health (of human and non-human life) may be orders of magnitude worse than Tetraethyl lead.
I certainly take your point that direct prima facie evidence for BPA as a hazard remains far less strong than dioxins. Perhaps PFAS might be a more immediate example.
Hmm. I didn't realize you were referring to an epigenetic pathway.
I'm in principle skeptical of meaningful effects happening via epigenetic pathways. I was a bit surprised to see your links to so many papers finding such effects in mouse models; I will keep an eye out for this, and maybe investigate it later when I have time. It would indeed be concerning if damage from endocrine disruptors lasted for multiple generations.
I'll still note that the damage from endocrine disruptors in these mouse models was *fading* over the generations, rather than becoming worse. Of course, this is in the context of the researchers not exposing the subsequent generations to the endocrine disruptors, so I see your point about how the damage may get worse across generations with constant background exposure. I guess all I have to say is that I'm vaguely suspicious of these mouse studies and think they must have done something wrong, but this is not a very principled stance so I acknowledge I could be wrong here.
Small side note: I'm ignoring the studies done in zebrafish or other non-mouse, non-human models, because zebrafish are quite far from humans and rodents are a much better model.
Another small side note: the study in humans who ingested DES is not very convincing. This is because DES was not given randomly: so far as I understand, it was given to at-risk pregnant women. Their daughters would therefore inherit whatever genes caused their mothers to have at-risk pregnancies, and therefore would be more likely to have at-risk pregnancies, even without DES exposure. This is ordinary genetics combined with selection bias, and not epigenetic effects from DES. Also, preterm birth straightforwardly affects the next generation (no epigenetics needed), so if DES exposure in utero caused the DES daughters to have preterm births themselves, this would affect the DES grandchildren without any epigenetics involved. So there are two possible explanations here that don't have anything to do with epigenetic damage.
Still, you've linked to a lot of mouse studies that aren't subject to these objections. I'll have to think about them.
Yeah, as someone who understands a fair amount of the organic chemistry, I find the idea that BPA is some monster poison laughable, particularly given (1) it's not water soluble, (2) even in the optimal laboratory case, it's a very poor substitute for estrogen, and (3) it's been used so widely and for so long that the idea that it could have massive health effects which have remained unnoticed for 50 years is dubious. If massive exposure for decades *does* do any actual harm, it would be barely measureable, and the relative risk compared to, say, the aflatoxins in your peanut butter or the radon you breathe every time you go down into the basement close to zero.
"Other times a reviewer was concerned that if you were successful, your work might be used by terrorists / dictators / AI capabilities researchers / Republicans"
Right. "Republicans." Because if we've learned anything over the past two years, it's that only one side of the political spectrum is capable of doing harm.
Seeing another comment like this above makes me feel like maybe there are some new readers on here who just aren't attuned to Scott's sense of humor? Rest assured, etc.
FWIW, I've been a huge admirer of Scott's writing in general and sense of humor in particular, for many years, and am not a Republican (though I was, kinda, an AI capabilities researcher, just a bit!), and yet I still cringed. It simply isn't funny enough to justify the potential misreading or sense of alienation enough people might have.
ETA: for the sake of new readers, though, rest assured etc. indeed.
I guess YMMV with this. Being familiar with Scott's humour and the context, I would have been unoffended with whatever word was in the place of 'Republicans'.
Mostly because the joke is set up by 'AI capabilities researchers' - after that Scott is free to drop in whatever he feels like. Republicans, Democrats, Georgists, Hospice nurses.......
But we all know that Scott really does worry about what AI capability researchers might do. So it doesn't so really set up a joke. Not that it can't end in a joke, but I don't doubt the AI bit at least is meant seriously.
If it had ended at "AI researchers", I'd have laughed. Going on for a political opponent made me wince, since that doesn't help with partisanship. Then again, I have seen people absolutely convinced Republicans are indeed evil moustache-twirling, widow-evicting, orphans' tears-drinking villains and not their next door neighbours, so it's dry humour I guess.
To overanalyze a split-second writing decision that I'm not sure I have much more insight into than the rest of you:
Terrorists, dictators, and AI capabilities researchers were actual things that various reviewers were concerned about. When I put them up there it seemed funny in a "murder, arson, and jaywalking" way, so I decided to go for broke and add "Republicans" to give it lots of different categories that people worry about for different reasons and at widely varying levels of certainty about how bad they are.
Speaking of "murder, arson, and jaywalking", I got that vibe from the endocrine disruptors link, where they group together dioxins and BPA into one category.
Dioxins are horrible substances that are both acutely toxic and a class 1 carcinogen, and they bioaccumulate to boot.
BPA is none of those things, and there is controversy over whether it is (a) totally harmless, or (b) possibly an endocrine disruptor causing mild harmful effects. But even in case (b), everyone agrees the common substitutes for it can easily be worse (they are just understudied). In other words, plastics that are "BPA free" are just as likely to be endocrine disruptors.
First, I’m very much in favour of doing good things in efficient and effective ways. In my book that makes me pro-EA.
Second, even if you don’t believe that, you can rest assured that I do believe in making people who give me money feel glad they gave me money. I’m a professional, and I intend to deliver results.
One point and one point only. You're a professional in a field that's known for professional ethics which foster only the highest levels of objectivity, rationality, altruism, integrity, honesty, efficiency and transparency. Being able to use something like Reddit to reveal preferences that appear to be, at best, totally unrelated to any of the grant objectives is not an undue risk in light of the high degree of professionalism you claim to possess.
Ash, what are your EA policy proposals, and who do you intend to lobby about them? Governments spend money all the time, so it can't just be about getting them to spend more. Do you have some ideas for making spending more efficiently targeted, or is it like an awareness raising thing? I've worked for a couple of Australian politicians and I'm interested to know what your plans are.
This is a question I'm interested in hearing the answer to as well. Are you in contact with the various EA groups in Australia, Ash? Melbourne has a reasonably active one and I assume that's also true of the other big cities - on that note, I'm curious the extent to which you're working for Scott or whether he's employed you on behalf of Australian groups with more local knowledge of what to lobby for?
The comment history is pretty large but shows no clear evidence of involvement in or advocacy for anything that would meet a definition for EA that basically every normal person wouldn't also meet. You would expect from a person who intends to undertake completely unspecified and unsupervised work in support of a policy framework at least some evidence of prior engagement with it.
Conversely I think the Reddit history strongly suggests that you've hired a pretty hardcore culture warrior who's otherwise a basically a standard political operative. The reason I think you've been taken for a ride is that political operations are extremely well known for their transparency and ease of progress measurement, while those who work in the area are universally upheld by all societies as paradigms of sincerity and honesty with a proven track record of never pursuing ulterior agendas most especially not in federal election years.
I would suggest that with a Federal election due in the next four months, the odds that a good political operative is focussed on the objectives of a foreign grant are not high. I would suggest that the odds of your goals being misaligned with his over the next four and twelve months are nearly 100% and I would suggest that structuring incentives to align those interests sufficiently to get value from your grant spend is probably going to be impossible even if none of the resources you invest are directly reinvested into things that are contrary to your goals.
What this suggests is that for the grant that seems to have the least clear and measurable outcomes and which is in a field where outcomes are notoriously unclear, hard to measure and hard to attribute you'll be most reliant on setting extremely clear, attributable and measurable objectives yourself.
My suggestion is that you set some objectives around land value taxation uptake. There is already very high degrees of support for it at multiple levels of Australian government in most parties and most of the public service, including limited adoption in one jurisdiction. Everyone agrees that more of the tax base being more efficient is good and that LVT is about as efficient as a broad tax can be. Finally it aligns with NA's interests and beliefs so it helps to minimise risks from misaligned incentives. The only disadvantage is that it is electorally unpopular which aligns poorly with NA's skills (elite vs popular advocacy). It is extremely difficult to imagine any other area where you will be able to find clear and even vaguely measurable goals within Australia's policy landscape that you desire for NA to achieve.
"The only disadvantage is that it is electorally unpopular". Unfortunately, this can be a huge and crippling disadvantage.
I know people have been shilling the land tax reform thing for a while but I can't remember what all the details are or how it's played out. Didn't they do it in Canberra or something like that?
Yeah, ACT. The only place where the wonk density is high enough to get it on the agenda easily. I know for an absolute fact that the government in at least three other states would love to do tax reform with LVT but no one's sure how to put it on the agenda.
Edit: Which is to say that it hasn't played out anywhere yet.
Which states? Can I ask that or is it Secret Insider Information?
I'm inclined to agree with you that the lobbyist grant seems dodgy, just based on the general vibe of the thing. It doesn't take $90,000 to learn about politics.
I mean, Victoria has a Land Tax but it's a weird thing with 'progressive' tax rates designed more to punish owning lots of land than incentivise good use of land. IME most state-level taxes are implemented in awful ways, and I'd love to see us get what was promised with GST, which is for taxation to shift almost entirely to the federal level with the money being sent back to the states but without all the weird distortions to economic incentives.
Victoria's quote unquote land tax is for vacant property. When they increased the rights of renters a few years ago they ended up having to change the land tax conditions because there was a wave of landlords who disabled their own property so it would qualify as uninhabitable so they'd be exempt the tax.
I think state laws are such a mess mostly because state politics is less competitive. It's the political B Teams of parties and some states are very much stitched up by certain factions and parties. Add to that that states had taxation power before the commonwealth so there's more legacy junk around and I think the picture is complete. The good news is that it makes it easier to fix.
"D, $5,000, to help interview for CS professor positions."
Huh? Interviewing for CS professor positions was taxing in all sorts of ways, but it wasn't expensive. In fact, universities were super accommodating, saving me the cost of a personal flight to Toronto in between interviews.
"Everyone with experience in movement-building says that getting your members into top positions at top colleges is important, and this is a surprisingly cheap opportunity to make that happen."
Huh. I'm guess I'm one of those members. Do we have any coordination mechanisms?
Decent list. Compared to some other EA-adjacent funding, this was relatively light on "$X to my friend for personal growth, because they are awesome and after personal growth they will affect the world for the better". That's a good thing. But it still had a bit of this category/pattern, which is a bad thing.
For example, getting money to apply for faculty positions in CS is pretty useless -- the interviewing universities will refund all the plane tickets anyway.
Yeah, the "please fund my lifestyle for a year while I sit around and decide if I want to stay on the course I picked" was a bit tooth-grinding for me: get a job, ya bum, like the rest of us if you're not going to be in training/education!
But hey, I was reared working class/peasant, where it was up to you to get there on time, pay for transport, and forego a day's pay if you had to take the day off for interview, and not "of course the people inviting you for interview will reimburse the cost of transport and pay to put you up overnight if necessary" class of job. How the other half lives! 😁
I support further experiments in connecting EA to political structures. For instance, there are many serious ongoing openings for engagement in the U.S. that should be appealing to the EA community, some of these likely among the very highest-leverage near-term opportunities to improve lives across the world at scale. Just some of these include: better vaccine distribution; the opportunity to provide aid to poorer countries at a scale of hundreds of billions of dollars to help them through the pandemic; augmenting sanctions policies so fewer people starve to death; ending various foreign escapades (the U.S. is complicit in a blockade of Yemen that has the potential to cause hundreds of thousands of surplus deaths) and more. And it can be striking how much difference can be made with relatively modest engagement, especially in light of the magnitudes of the potential outputs.
Numeracy, sorting out priorities based on even a vague sense of utilitarian impact, is an underdeveloped skill in DC. And EA community would do well to learn more about politics, legislation, rule-making, etc to come to understand highest-leverage mechanisms for potential influence and how to manipulate them. (I say this as what I suppose one might call an American analog to the Australian grantee.)
Yes, absolutely! It’s politics and nothing is guaranteed, but I consistently find it shocking how much impact just basic engagement with the political process can have.
Have you by chance seen any organizing activity in Australia around the IMF/Special Drawing Rights? That's the most profound opportunity I'm working on here. Conceptually could get on the order of $1 trillion or so out to poorer countries to help contend with pandemic fallout (direct health and economic), with around $200B or so already released.
Thanks to Scott and all the generous donors on behalf of me, Delia Grace, and the pigs, farmers and pork-lovers in Uganda! Very exciting to be part of such an eclectic and talented group
This was an astonishingly fast submission, review and grant allocation process, compared to how this is traditionally done. Impressive, not least taking into account the large number and diversity of the projects seeking a grant. Based only on reading this blog post, it seems that the review process succeeded in selecting exiting projects. Made me think if there are more detailed tips or tricks concerning how Scott organised and implemented the decision-making process, that could be useful for other public or private organisations (or wealthy individuals) funding research. Consider this a suggestion for a possible future blog post.
"Why would somebody working on biochemistry want to talk to someone working on political activism..." Well, why would anyone talk about those topics on the same blog? It is all part of the codex, while all fields of research have their specific tools, they also useful tools of the general craft. As a PhD-student, I really like the events where you get to talk to PhD-students from other fields, because you have a good opportunity to learn about the field of others, and more importantly, learn how your field is seen by others and try to give it a more positive/reasonable image to the person talking to you.
Incredible work, huge praise to the applicants and the work of Scott and the reviewers picking out these exciting projects. Very interested to hear about how the various projects progress.
Well, I imagine you can't really sue the mobile slaughterhouses out of existence *before* they come into existence. Give the Chicken Farms Suing people time to get round to it!
More seriously, it does raise the moral question: is it better to be a dead pig or a live but sick pig? If the live pigs are spreading disease, is the benefit to them of their lives more than the benefit to the healthy pigs of their death? If you're going to argue that raising and killing chickens for food is wrong, this is another question you have to address. Will the slaughter houses sell the pig carcasses for food and if so, isn't that immoral on animal rights grounds? If they have to kill infected pigs, shouldn't they destroy the carcasses rather than encourage meat eating and so make it profitable for farmers to raise pigs for food?
(Scott is the one who used "moral" in a couple of these awards descriptions, so if you're going to argue morality as a ground, you have to be prepared for those counter-arguing your proposition is immoral).
"Other times a reviewer was concerned that if you were successful, your work might be used by terrorists / dictators / AI capabilities researchers / Republicans and cause damage in ways you couldn't foresee. " -- I cannot believe that you actually wrote this. Very disappointed. (Not in the grants, in what how you chose to write the intro.)
Scott's making fun of people who think Republicans and AI capabilities researchers are as bad as dictators and terrorists. It's a classic Scott joke, I laughed.
Is there any way to support the next ACX grant with private donations? I am sure many of your readers would love to participate in such a high-impact and diverse portfolio of projects
On a different note, I would like to invite you to visit by blog. I am Dutch and writing in English. It would be great to get feedback, both regarding my English and the general writing style of my blog. I would greatly appreciate it.
I like your writing style - very easy on the brain. Your English is also excellent. I only noticed one idiomatic oddity which was the phrase "3 o clock at night" - most native English speakers would say "3 o clock in the morning. So... your English is even better than the average Dutch person :)
I would prefer longer, more in-depth and more striking articles but each to their own.
I wish you well in your writing, drawing, dancing, playing and singing!
> they're smart enough to know about XYZ Grants which is better suited for them, which means they're mostly banking on XYZ funding and using you as a backup
Um, doesn't this confuse "smart enough" with "experienced with applying for grants"?
Sure, if it's considered a norm to apply for N grants in parallel, this means an experienced grant reviewer would encounter N times more applications like that, and it's reasonable to assume this as the default, and to try to adjust for this kind of selection bias.
But as a person who had applied for the first time, this sounds... surprising, to say the least.
Intuitively, it sounds intrusive and rude to just spam every grant program out there. Is this really expected and OK?
Would really appreciate comments from an experienced grant reviewer.
Still, thanks a lot for highlighting this point! Will keep it in mind.
(I've never previously considered grants as a realistic source of funding -- coming from the industry background, and given anecdotal evidence about the amount of bureacracy from friends who successfully received ones. Decided to apply only because the initial post was so much more welcoming and really different from what I've heard about grants previously)
On a second thought, if what this really means is "since the applications are going to have a second chance at ACXG+ anyway, let's focus on the ones that are unlikely get funded via ACXG+" then it's totally reasonable, of course.
(1) If you're looking for funding, don't put all your eggs in one basket. Apply everywhere, you'll probably get some positive reply, rather than just applying to one source.
(2) This supposes that people will know that XYZ Funding exists. They may not do! It wasn't until I started working in one area that I became aware of (a) exactly how much the organisation covered and (b) how many people had no idea they could apply for grants from that same organisation.
Someone smart but inexperienced or just not plugged into the "grants and funding application network" may not be aware at all that they could apply for and be awarded XYZ Funding.
Point 2 is really important. I think it'd be a great service (though possibly also a lot of work) if Scott or associates were to go through every candidate who got that comment and still ahs a funding gap and let them know that the reviewers thought they should be applying for XYZ Funding.
It seems like the system you used to publish the rest of book reviews worked pretty well. Perhaps something similar would work with the grant applications.
I am very surprised that experiment.com was not mentioned. I am constantly frustrated that there are not more medical science projects there to support.
And the ones that I _do_ want to support get funded almost instantly. Come on, researchers, take my money!
"Alice Evans, $60,000, for sabbatical and travel to fund her research and associated book on "the Great Gender Divergence", ie why some countries developed gender equality norms while others didn't."
The chickens one is also pretty bad (and I'm not going to claim originality, this was pointed out on Reddit too, I didn't even notice the clause): to help kickstart their project of suing factory farms that violate animal cruelty laws *or otherwise expose themselves to legal action.* Translation: If they don't violate animal cruelty laws, we'll use the $72000 to find something else to get them on.
That some of the funded projects would be trash like this is hardly surprising. Always gonna be some duds in a program like this. Luckily blowing 60k on 1 really dumb thing is actually not even close to a worst case scenario. It also doesn't say who got their money from Scott vs others. Maybe this funding came from someone who agrees with the other stuff.
Basically this. There were a few in the list that looked like wastes of money to me, but out of this many projects it really wasn't that many and they tended to be on the smaller side of funds spent.
I think the chickens one is bad, too, because it's basically funding nuisance lawsuits. The group think factory farming is bad, eating chickens for food is bad, and want to get commercial and private 'chickens for food' ended. Going after factory farms is low-hanging fruit; I'm all for prosecuting abusive industry, but "or otherwise" means "if we can get them on having 0.99 instead of 1.0 hectare per 100 chickens, we'll sue them and with any luck force them to close down". No.
I don't know how a lawsuit like that would even succeed. Chickens don't have any constitutional rights, and if they did, only chickens could sue for their violation, since you can't sue on some other...entity's...behalf unless you are its parent or something. So what would be the basis? Violation of assorted local regulations on cage size, et cetera, one assumes -- but it's unlikely any such regulations allow a private lawsuit as a method of enforcement (unlike the novel and much maligned recent Texas anti-abortion law). Normally the only plaintiff with a cause of action for the violation of regulations set by the state is the state itself (meaning the state can sue, but not any individual). I guess you could try animal cruelty laws, but, again, the usual cause of action lies with the state. The only thing that comes to mind is suing on the basis of your plaintiff's mental distress -- OMG! I ate a chicken that had been cruelly raised! I feel so guilty now I'm going to hang myself -- or a much more difficult to prove claim of some actual physical damages (e.g. the chickens from Factory Farm X turned out to contain high levels of Compound Y because they were under constant stress and Y caused your asthma).
If I had to take a wild guess, I'd say nuisance lawsuits -- with no real hope of success -- is the actual game plan. You sue for any wild reason you can get a lawyer to commit to paper without fearing for his license -- and hope to get into the news, embarass the defendant, get a 6-figure settlement which you use to fund the next round, et cetera.
I suspect Scott's reasoning below is correct (that actual full on abuses are sufficiently common to avoid needing to sue on technicalities).
However, even if that's wrong, if you think factory farming is basically immoral (I do), then making the industry act overly cautiously to avoid lawsuits seems great. If Tyson Chickens ends up giving all its chickens 1.2 hectares because it's worried it might slip under 1.0 and get sued, that seems like good news. If Tyson Chickens ends up going bankrupt, all the better.
I'm against nuisance lawsuits in medicine etc because I think they have negative externalities and reduce the quality and quantity of a good with positive externalities and high social value. I'm at least queasy about Peter Thiel suing Gawker into non-existence because it impinges on a fundamental political freedom (freedom of the press). None of that applies to factory farming.
If the critique of this charity is that they might shut down factory farming for $72,000, that is probably the best thing Scott will ever do.
I think enough factory farms genuinely violate animal cruelty laws (and whatever your common sense notion of animal cruelty is) to burn through $72,000 (or $72 million) without getting to edge cases.
Dr. Evans has some beliefs I find annoying, but her core work is about things of the level of forced marriage or women being banned from owning property and so on. I feel like people on every side of first-world debates about social justice can agree those are bad and be interested in changing them, and I'm told Dr. Evans' work in those fields is rigorous and important. I think if I were to think less of her based on some of her political comments I find annoying, and refuse to back someone's rigorous / important / potentially beneficial work on those grounds, that would be akin to cancelling / witch hunting / whatever you want to call the thing where you blacklist someone because of their politics.
I agree with Deiseach (below) about the chicken people. If they were motivated to keep factory farms to the spirit, or even letter of the law I'd say fine, why not. But they aren't - they expressly want to shut down law-abiding businesses for ideological reasons. However your defence of them is fair, and pragmatically they will merely harass the worst actors so their influence will (in my view) be mildly positive.
I'm more impressed with your ability to put aside the annoyingness of Dr Evans. I couldn't do it, despite wishing that I had the courage and charitableness. Bravo!
I think your sociological reasoning is in error. The most likely targets for harassment are actually the *best* businesses, those most concerned for their public image, and those which are more profitable through efficient and effective operations -- but who have made a mistake here and there (and what big operation has not?) You don't sue people who are barely making it, financially, because they won't be able to pay out if you win. You don't sue people who are already shady operators, because ipso facto they don't care as much for public opinion. To have the most impact (to your accounts receivable), you sue the people who have a high public profile, who are big and wealthy and successful, and who appear in general to be nice clean folks -- and who prize that reputation. Those are exactly the kind of people who will give you $1.5 million to just go away as soon as you serve them.
I mean, people sue brain surgeons and obstetricians at famous clinics, not used-car salesmen from the wrong side of the tracks who chisel them on a beater. As Willie Sutton said about why he robs banks, because that's where the money is.
Yes, I agree with you about the targets. However I don't think the motivation is money and the likelihood of payouts - Tyson foods gets sued fairly regularly and pays out a few million here and there for various breaches of regulations. But they're not paying out to the activists who want them completely shut down.
I think the objective - at least the proximal one - is simply to get negative images (of suffering/malformed/generally oppressed chickens) into the public eye as often and as viscerally as possible. To tie those things into "industrial farming" would obviously be a bonus. Because on the slow march to a virtuously vegan world there will probably have to be some other destinations along the way including generally higher welfare standards, free-range and organic.
My objection to the support of this project is the dishonesty involved, and I admit that I may just be naive on this point. But the project isn't to genuinely keep the industry sticking to the regulations - or even to change the regulations themselves. It's to harass people because of an ideology which, if admitted more honestly, would result in them receiving a great deal less support.
>I'm more impressed with your ability to put aside the annoyingness of Dr Evans. I couldn't do it, despite wishing that I had the courage and charitableness. Bravo!
It's admirable in one sense, though I think extremely naive to think that these two areas are two totally independent things and that good insights can be expected to come from somebody so absolutely wrong about a related area.
I'd expect her beliefs about structural racism to be associated with particular beliefs about gender equality. It's not as if she's the person working on autoimmune disorders or neutrino physics--it's not "cancellation" because her beliefs are relevant to what you wanted to fund her for. Everyone here can agree that forced marriages are bad, but she's not just going to be saying "these are forced marriages. Stop them." She's going to have recommendations and analysis and I don't see someone into social justice as doing well with those.
(And if you say "someone can have bad beliefs about racism but be perfectly sane when it comes to gender relations"--that's logically possible, but my priors are strongly against that.)
Her actual work does not seem to be 'structural racism but gender', especially the stuff on South Asian history. I ironically came across her via Razib Khan of all people: https://www.draliceevans.com/blog
>I feel like people on every side of first-world debates about social justice can agree those are bad and be interested in changing them, and I'm told Dr. Evans' work in those fields is rigorous and important.
Okay? She's an academic employed by two elite universities. She doesn't need funding.
SURELY there are people doing rigorous and important work like this who actually need funding and don't have completely hateful and wrong-headed views like Evans, so?
And I sincerely doubt that her work can possibly even be rigorous when her view of the world in this other, related domain (race) is so completely false. Do you really think she's going to produce some stunning insights on "gender equality" when she is totally, 180 degrees wrong (and extremely obnoxious) about other social issues?
I'm all for funding work on reducing gender equality, but I struggle to believe that this was even the tenth best option available for doing so, even if you had to ask to get people to go looking for those opportunities for you.
>I think if I were to think less of her based on some of her political comments I find annoying, and refuse to back someone's rigorous / important / potentially beneficial work on those grounds, that would be akin to cancelling / witch hunting / whatever you want to call the thing where you blacklist someone because of their politics.
It certainly doesn't help that you say this after essentially saying "we don't want to give money to those evil republicans haha just a joke but not really".
If Scott couldn't fund your project, but it might be a good fit for a Focused Research Organization, please get in touch with us at https://convergentresearch.org/.
Hi, I'm would like to make up some of the $150,000 (I can't give the entire thing but I am willing to give what I can) that you say Kartik Akileswaran and Jonathan Mazumdar are looking for but you didn't list a way to get in tough with them. I'm hesitant to use a contact I found on Google for fear of being scammed by the wrong Kartik Akileswaran or Jonathan Mazumdar. I also can't seam to find, Dr. Alexander's e-mail listed on ACX or Substack anywhere. I'm probably just missing it.
I was raised to be quite paranoid and I'm trying to figure out how to verifiability give money to the right people. Are there any suggestions on how I should do this? If Dr. Alexander can see the e-mail associated with this comment then feel free to e-mail me with contact information for the pair. If it is known that Scott Alexander checks his Twitter DMs then I can probably contact him through a friend of mine.
Okay, here's where I need some clarification. Take this:
"Still other times, my grant reviewers tied themselves up in knots with 4D chess logic like "if they're smart enough to attempt this project, they're smart enough to know about XYZ Grants which is better suited for them, which means they're mostly banking on XYZ funding and using you as a backup, but if XYZ doesn't fund these people then that's strong evidence that they shouldn't be funded, so even though everything about them looks amazing, please reject them."
Then take this:
"Alice Evans, $60,000, for sabbatical and travel to fund her research and associated book on "the Great Gender Divergence", ie why some countries developed gender equality norms while others didn't. A large body of research shows that gender equality, aside from its moral benefits, is also deeply important for economic development. Dr. Evans is an expert on the interaction of gender, history, and economics, whose work has been cited on BBC, Al Jazeera, and Sky News. She blogs here and podcasts here."
That is a project that should be funded either by her publisher for the forthcoming book on it, or her university department. Are Princeton University Press suddenly so penniless they no longer pay out advances? https://press.princeton.edu/news/the-great-gender-divergence
Anyway, there a couple of things I would like to know further:
(1) How do you define "gender equality"?
(2) There are plenty who complain that we don't have gender equality even now, today, in the developed Western world. Are we going to see talk of the pay gap, or will all this be subsumed into "women can be prime ministers, too!"
(3) I know it's a blurb, but there seems to be already a structure in place for why this is so:
"Inspired by research on The Great Divergence, Evans asks why, as societies around the world have become more gender-equal, some regions have become more equal than others. Something radical happened over the twentieth century: women entered the workforce and became political leaders across the world. This had never happened before, not in the entirety of human history. What changed? And why is progress so uneven?
The Great Gender Divergence will explain the causes of Europe’s “precocious equality,” how East Asia and Latin America caught up, why gender equity in the Middle East, South Asia, and Sub-Saharan Africa has lagged behind, and why South-East Asia was always ahead. Evans offers a comparative history of how societies come to support gender equality and why this varies around the world, telling a story about geography, economic growth, strong states, and militant activism."
So it would seem that Dr. Evans has already answered the question posed and it is about "geography, economic growth, strong states, and militant activism".
Myself, I would put the difference in one word: "education". Can I now hold my hand out for a share of $60,000 for a pleasant junket jetting around internationally while I 'research' a topic I have already made my mind up on?
However, let me quote one grant that I do think is a good and useful idea:
"Michael Todhunter, $40,000, to continue work on automating testing cell culture media."
Yep, badly needed, and I wish him good luck on this.
Good job raising all that money. Congratulations to the donors and grant winners.
> illegal fuel enrichment produces neutrinos which could theoretically be detected from thousands of miles away
To nitpick: I do not understand why fuel enrichment (as in U-235 enrichment, e.g. in a centrifuge, which involves neither nuclear reactions nor the weak interaction in particular) would produce neutrinos. Are you sure you are not talking about monitoring the anti-neutrinos from reactors used to breed Plutonium?
Fuel enrichment obviously does not create neutrinos itself, but if you had a super-duper sensitive neutrino telescope, I imagine you could probably tell that the concentration of U-235 was going up in some location from the higher level of spontaneous fission reactions there. Although the level of sensitivity required is incredible. Normally neutrino detectors the size of swimming pools are needed to detect the enormous blast of neutrinos put out by a supernova, or by the Sun's fusion at its core. (Which also points to the incredible level of noise-discrimination you'd need.)
"Still other times, my grant reviewers tied themselves up in knots with 4D chess logic like "if they're smart enough to attempt this project, they're smart enough to know about XYZ Grants which is better suited for them, which means they're mostly banking on XYZ funding and using you as a backup, but if XYZ doesn't fund these people then that's strong evidence that they shouldn't be funded, so even though everything about them looks amazing, please reject them."
Completely understandable, but there are many reasons that a given philanthropy might not donate to a seemingly obvious recipient. Only one of the reasons (and, I suspect, not the most common one) is that they made a full, fair, and objective investigation into the recipient, and decided to reject it for reasons that everyone else would agree with.
If Foundation A gives a lot of money to Issue B, and if a proposal relates to Issue B but got turned down by Foundation A, there could be lots of reasons for that:
1. Foundation A took a solid look at the proposal, and decided for objective reasons that it wasn't worth funding compared to everything else related to Issue B. (Fair enough).
2. It's December, and Foundation A didn't look at the proposal at all yet, because their internal process requires the proposal to have been submitted by June 1 for a decision to be made by the Board on Nov. 1, and there's now a new funding cycle that won't make decisions until March of next year.
3. Foundation A already exhausted its internal budget for Issue B for now.
4. Foundation A hasn't quite exhausted its internal budget for Issue B, but staff are saving the rest while they argue about an expansion into Issue C.
5. Foundation A has a new President with a different vision for the organization, and unbeknownst to anyone has decided not to invest in Issue B going forward.
6. Foundation A has decided to revisit its entire strategy on Issue B, and the strategy planning and discussion will take a year before they make any more grants.
7. Foundation A's staff had already put a lot of time and internal capital into arguing for a particular approach to Issue B, and this new proposal takes a slightly different approach, and they are embarrassed to change direction so quickly.
8. Foundation A's staff have been involved in an internal squabble for reasons unrelated to Issue B, but which make it impossible for now to get the necessary agreement.
9. Foundation A hasn't funded it *yet* . . . but might soon!
10. Foundation A is itself playing a strategic game where they are trying to anticipate what other funders might do, and that is delaying everything.
11. Prominent grantees of Foundation A are skeptical or insecure about the newcomer proposal, and Foundation A wants to preserve good relationships with them.
12. The decisionmakers at Foundation A were somehow rubbed the wrong way by the proposal and/or accompanying PowerPoint.
13. The decisionmakers at Foundation A just don't have a good feeling about the grant, for reasons that they can't articulate.
None of these rationales are only hypothetical--they do happen. And I could keep going. There are many reasons why we wouldn't expect the efficient market hypothesis to apply in philanthropy. And it doesn't. Due to the lack of market feedback and the prevalence of idiosyncratic philanthropist/staff preferences, there are many $20 bills lying on the ground for anyone to pick up. This is 1) exciting for potential philanthropists and 2) cause for concern about the overall incentive structure in philanthropy.
I'm really happy to see the endohazard project here. I've been finding the research into (and huge public blind spot around) EDCs and POPs so alarming that it's basically displaced all of my concerns about climate change. I think there's a good chance that we're basically manufacturing another Great Oxygenation Event.
There are three properties that worry me: (1) they are dangerous to human and ecological health, (2) they are basically never broken down and so accumulate in human bodies and the environment, and (3) the rate at which we're inventing and manufacturing them is growing.
It's especially number 3 that I think has the biggest tail risk. In the time it took to discover that just BPA was harmful, who knows how many similarly dangerous compounds had entered the supply chain. As more of these chemicals are invented, they'll have more and more useful properties and start replacing ever more previously-harmless materials in new applications. We already store, transport, and prepare/serve food and water in plastic. Medical equipment, toiletries and textiles are often made with PFAS.
If POPs/EDCs are partly responsible for obesity, which seems likely, then obesity will only get worse (and the increasing rate of use of these compounds means they'll likely become the leading cause if they're not already). Obesity, at higher and higher levels, will be a catastrophic blow not only to health, but to our economic and cognitive capabilities as well.
The other (possibly silly) thing I worry about is why I referenced the Great Oxygenation Event. When (not if) bacteria/fungi/something adapt to metabolize plastics, several things might happen: more leeching of EDCs/POPs from this metabolism, loss of plastic-based infrastructure, major damage to food/water/medicine safety (eg. plastic food packaging suddenly being useless), and a new family of pollutants from whatever we replace plastic with.
And with climate change, economic forces are at least partly in the right direction, where solar power and batteries are getting cheaper, fast. But I don't see any natural economic incentives pushing against POPs/EDCs.
"The other (possibly silly) thing I worry about is why I referenced the Great Oxygenation Event. When (not if) bacteria/fungi/something adapt to metabolize plastics, several things might happen: more leeching of EDCs/POPs from this metabolism, loss of plastic-based infrastructure, major damage to food/water/medicine safety (eg. plastic food packaging suddenly being useless), and a new fsmily of pollutants from whatever we replace plastic with."
That concern would seem to mitigate against the grant for plastic-eating beetles!
I do still think it's worthwhile research, but I also fear it misses the main problem of plastic pollution. As I understand it, it's the polymers themselves which have the energy that can be metabolized. But the biologically harmful pollutants (BPA probably being the most famous) tend to be additives to the polymers. So breaking down the polymers risks just releasing more toxins from existing pollution.
Thank you so much for your support and encouragement. I have wondered if focusing on POPs rather than EDCs (a rough subset of many POPs) might be a better focus for labelling, as there is longstanding political backing to phase out POPs, and generally strong evidence of their dangers. However, there is considerable overlap between EDCs and POPs, and the general public encounters EDCs every day, whereas POPs are supposed to be rather restricted. Hence, the focus on EDCs themselves.
If you proposed an ACX grant in an area related to alt-ac (alternative models to the standard University system, broadly conceived), please consider dropping me a line. You can find me by e-mail at sdedeo[at]andrew.cmu.edu
This is because, in tandem with a few colleagues within the system, I’ve been thinking about alternatives to how we provide what is currently described as “elite” education (small seminars, research mentorship, residential, etc). It would be helpful to learn what other people are thinking about.
I really want to know what the Australian political lobbyist guy is planning to do with $90,000. Seems like it could take a lot of different directions, depending on what your policy proposals were and which party you were talking to.
One thing that would be cool is if there was some feedback for the rejected proposals, on why it was rejected, even if it's very brief. Even if you disagree with it, it would be awesome to know what potential funders have thought of your idea when first hearing about it.
Personally, like many in the field I think the classical evolutionary theory of aging* is more plausible, or more widely relevant: that it's the result of evolutionary neglect, not itself an adaptation [selection cares less about late life than about early life; if there are trade-offs between them selection favours early life**]. However, I'm also glad that there's a minority of researchers pursuing alternative theories, in case they're true or in case e.g. they affect the rate of ageing even if not being the reason it exists. Here's another recent review, this one focusing on colonial organisms, and emphasizing kin selection not group selection***.
>L, $17,000, to breed a line of beetles that can digest plastic. Darkling beetles (and their associated gut microbes) can already do this a little. Maybe if someone selectively bred them for this ability, they could do it better. Plastic is generally considered bad for the environment because it's "not biodegradable", but maybe everything is biodegradable if you have sufficiently advanced beetles. This project will find out!
The contrarian perspective here is that since plastic is made from hydrocarbons, it's actually a good thing that it doesn't degrade. Throwing non-biodegradable hydrocarbons in the dump = carbon sequestration.
I hope these new beetles don't fart out CO2 or methane after eating plastic...
Interesting! A lot depends on where the plastic is. We don’t consider microplastic in the ocean to be sequestration… maybe we need plastic-eating algae. I suppose we would have to balance the badness of landfills against the badness of atmospheric carbon.
Thanks for doing this and for this post! I'm one of the guest fund managers on one of the EA Funds (the EA Infrastructure Fund, specifically), and *I would really like many of these people to apply to EA Funds for a top up or a substantially larger grant right now (if they haven't already), and for many others to apply later on for "next phases" of these projects or for new projects*. https://funds.effectivealtruism.org/apply-for-funding
This can pretty easily be worthwhile in expectation because:
1. It should take 0.5-2 hours to write an application (setting aside time actually planning the project)
2. The actual evaluation process is typically pretty quick too, for both the applicant and the grant evaluators
3. It's faster for things that don't end up funded (so it's relatively unlikely for lots of time to be spent without impactful-in-expectation work ending up funded)
4. EA Funds's impact is probably most bottlenecked by number of good applications received (more so than by fund manager time or money available) (I'm most confident of this for the Long-Term Future Fund and the Infrastructure Fund)
Scott, did you or someone else already heavily emphasise roughly that message to the grantees? If not, could you do so? Let me know if there's any way I can help (I can be reached at michaeljamesaird AT gmail DOT com )
Also, many readers of this comment should probably consider applying too.
(Caveat: This is mostly a message I spam repeatedly in lots of EA-adjacent contexts, rather than something I'm saying because I think lots of these projects in particular sound extremely impactful and funding constrained. And there are many projects listed here that I don't feel very excited about from an impartially altruistic perspective, even if they sound cool from a vaguely-progress-studies perspective. That said, many do sound either likely to be great or *likely enough* to be great that submitting an application is worthwhile in expectation.)
This is kinda what I imagined early university/bell labs or google/steam research looked like. Eclectic, smart people asking for money to try to solve interesting problems and getting money to do so. This was one of the most inspiring posts you've ever put together, intentionally or not. Makes me more optimistic than I've been in a while about what the future holds. Thank you.
Scott, you are probably already doing this, but if not… My wife is a professional fundraiser for a granting organization. I told her about the ACX grants. She recommends that, for tax reasons alone, you establish yourself as a foundation under the appropriate tax code. Especially if this is not going to be a one-off, make sure the IRS doesn’t come and bite you. Anyway, I hope you have already talked to lawyers about this.
I've seen some concern about bias against Republicans, so I asked at datasecretlox.com about what Republican or (as it turned out, Republican-adjacent) grants would look like, and got some interesting answers, and I'm asking again here.
I would be interested in seeing the version of this post with the unsuccessful candidates, maybe those 500 people can submit their own TLDR blurb with a link, and Scott just post the whole thing?
As for a ‘group’ of ‘forum’: the SSC reddit sounds definitely like a nice place for the winners to share follow up stuff and status updates etc etc back to the community! maybe with subject prefixed by an agreed on [ACX 2021 Grantee] tag or something?
This is amazing. I do worry that Scott is too trusting to transfer millions of dollars around, and may get into sticky legal situations. E.g., researcher Z has spent $20,000 on an approved project, but the funders haven't yet delivered Scott enough money to pay researcher Z $20,000; or a major donor died leaving Scott without a legally-binding promise to deliver the money; and Scott is on the hook for it. The usual practice is to have a legal team and some full-time staff to manage contracts and payments.
How will payments work? Will Scott have the money, or legally-binding promises for them, before the grantees start spending money? Do grantees get the money up-front or after-the fact, is it in installments, do the installments require milestone deliverables? Does each project need to deliver a final report for public distribution before receiving their final payment? What is the status of the intellectual property produced? I have no need to know myself, but hope these things get addressed before it's too late.
In the past, before grants came with excruciatingly-detailed contracts outlining milestones, deliverables, and payments, grantees sometimes spent all of the money doing the research, without leaving enough time or money to organize, summarize, and publish the results. There were archaealogical digs which excavated important sites, yet never published any of their findings; so all we have left are boxes of artifacts without information beyond which site they came from.
Today, grantees must deliver reports, but these reports might not be saved, or even read, let alone be published or publicly available.
I don't know if it's possible to sign a binding contract to make a donation in the US. US contracts must have "consideration" to be enforceable, meaning that a donor signing a contract to hand over money must, by that contract, receive something of approximately equal value to that donor in return. So a contract with a donor to donate some sum of money couldn't be binding unless it's tied to specific research projects, and to the donor receiving some sort of compensation from those projects. Perhaps the compensation can be an emotional value, but I doubt it can be anything as general as "the feeling of satisfaction from making the world a better place".
Without contracts for donations, it would be wise to get the donations up-front, so Scott doesn't end up on the hook if some donors fail to pay. Better yet might be for Scott not to be in the loop handling money for all of these grants, but to link each funder with specific grantees and let them work out their own contracts.
My plan is to give everyone their money ASAP. If they take it and run, then they've successfully scammed me and I will feel sad, but I chose people who I thought wouldn't do that, and this is the only option available to me given that I don't have the infrastructure to constantly monitor everyone.
I should have all the money I've been promised within a week or two, after which it will go to the grantees. I've told all of them that until it's in their hands, these aren't legally binding promises. I have a backup plan for who I'm going to ashamedly defund if I get a 95th or 99th percentile bad outcome in terms of people shirking.
Don't mean to be rude, but did Alice Evans really need $60k? She's an academic at two elite universities and is doing research which strikes me as broadly indistinguishable from other people working in gender studies. Would this work really not be done without such funding?
And from (an admittedly brief) perusal of her work, it seems she completely shirks consideration of any heritable biological differences between sexes or populations that could at least partly explain the phenomena she investigates, which must surely limit the depth of insight (e.g. concluding that cultural/political factors explain X difference between countries with the implicit idea that these things are just a matter of random chance like which parts of the world have arable land or something).
This seems somewhat similar, except (1) most of my discussions are focused on career development, only a few are on technical skills, and (2) I'm approaching the scaling problem differently
I wonder if we'd change each other's minds about something, or collaborate about something, I have no idea really, but it seems worth reaching out
"A large body of research shows that gender equality, aside from its moral benefits, is also deeply important for economic development." [[Citation Needed]]
I see a large correlation between gdp per capita and gender equality norms, but that could just be that the countries that were smart enough to develop a high GDP per capita were also smart enough to abandon or attenuate their traditional religions.
On the other hand there's a large anticorrelation between gdp per capita and gender equality in choice of profession. So the countries with more gender-egalitarian ideologies have less gender-egalitarian labor allocation. This undermines the OP. My first paragraph seems like a much more likely explanation.
[Edit: maybe they are just talking about the very basic first steps of gender egalitarianism, i.e., giving women access to contraception and control over their own fertility, so that reproduction doesn't outpace the capacity of schools and other infrastructure. That seems very plausibly beneficial to growing GDP per capita in third world countries. But I'm still very doubtful that taking it further than the gender norms of a TFR<2.5 country confers any additional benefit to GDP per capita]
Nuclear tests already generate shockwaves detectable by seismometers thousands of miles away.
The stated plan to detect mere enrichment via neutrinos is impossible. Maybe he means transmutation of U-238 to Pu-239? That involves a beta decay that would release a neutrino.
Solar neutrino flux is circa 10^15 per m^2 per second.
Let's calculate the neutrino flux due to someone producing a critical mass (10kg) of Pu-239 in a year, 1km away.
Total neutrinos = number of plutonium atoms = (10000g/239g)*6.022e23 = 2.5e25
Neutrinos per second = 2.5e25 / (365*24*3600) = 8e17
Surface area of 1km radius sphere = 1.25e7 m^2
Neutrinos per second intersecting 1m^2 target at 1km distance = 6.4e10
So that's four orders of magnitude below the background noise.
But maybe you can very precisely detect the energy of the neutrinos and thus distinguish between plutonium neutrinos and all others. You'd also need to simultaneously get a precise direction vector, because there are hundreds of legitimate plutonium producers around the globe. (e.g., every uranium reactor has a side effect of making some plutonium). I don't know enough physics to say whether it's possible to design a neutrino detector with those parameters. But I know that the largest neutrino detector in the world (https://icecube.wisc.edu/about-us/facts/#:~:text=IceCube%20detects%20275%20atmospheric%20neutrinos%20daily%20and%20about%20100%2C000%20per%20year) gets about 275 neutrinos per day out of a solar flux ~15000 times bigger than the signal we're looking for. So in the case of the guy making 10kg of plutonium in a year, 1km away, it'd receive about 7 neutrinos in a year. So in terms of quantity alone, detection is just barely possible if you're within 1km of the largest neutrino detector in the world. If you're not on the same continent, (it's located at the south pole), forget about it.
What a cool suite of ideas and initiatives. And your insights on grantmaking and the process are just as valuable. Delighted to see the good you are doing (even though my idea didn't get up).
Except for the adopted ones, the ART-conceived (assisted reproductive technology) kids wouldn’t exist otherwise. And besides, the whole point of IVG (in-vitro gametogenesis) is to reduce the need for donor gametes.
How is being conceived in a different way to the norm being 'commodified'? I want to have kids with my partner, and right now that's not possible for us. If the Ivy Natal Fertility group succeed, it would be. The ability to have biological children is no more than what straight couples have, with the added bonus that all children born of this method would be intentional and desperately wanted.
Such an impressive and inspiring range of projects. I am excited to live in the world where these dreams become reality. What a cool project, Scott.
Yes. And people seeing these get funded and what happens as a consequence may well be amazing. :)
Definitely bookmarking this post. Excited to follow up over the next few months and years to see where a lot of them end up :)
Scott has some interesting thoughts outside the mainstream such that I am tentatively excited for some of these grants. If 20%+ end up paying off at least somewhat that would considerably raise the prestige and interest in later ACX grants I'd think.
I assume Scott has already done a forecasting success thing that he will reveal some years from now about which ones he feels most confident about. Of course like start ups having a 1 or even 2 in 10 hit rate is actually quite good.
I love the beetle one. Can we make that an ACX meme somehow?
(I also like that a nuclear war one showed up. EAers have always rated nuclear war and pandemic/bioterrorism as two of the 3 worst existential risks, but generally had an attitude that those two big problems were mostly better left to establishment government & science. Covd-19 suggests that this may have been misplaced confidence when it came to public health; perhaps we have misplaced confidence in the other one too.)
Finally a scientific advance related to insects that doesn't require me to eat them! Here's hoping it works.
Amen.
Of course we wouldn't eat the beetles. Obviously.
Unless we ran out of other food or something.
Of course we wouldn't eat the beetles, except to fend off starvation in desperate🪲circumstances. Obviously.
But then, there's some elasticity around what counts as "desperate circumstances," depending on how the beetles taste.🪲And we can't very well claim to know the exact hedonic exchange rate between eating the beetles versus other🪲food sources, if we haven't tried them even once...
🪲Of course we wouldn't eat the beetles, except for experimental purposes, to determine under what circumstances🪲we should reconsider our anti-beetle-eating🪲stance. Obviously.
Also we should take the relative supply🪲🪲of beetles and other goods into consideration.🪲We do produce a lot of plastic waste,🪲after all. Look at them. Just crawling around,🪲not🪲being put to any better use...🪲🪲🪲
🪲🪲Of course🪲we should🪲put🪲🪲the beetles🪲to the🪲most🪲🪲🪲economically efficient🪲use.🪲Obviously.🪲And if that🪲🪲means eating them,🪲well,🪲it's not like🪲they🪲were so🪲bad that🪲one🪲🪲time...🪲🪲🪲🪲
🪲🪲🪲🪲Of🪲course🪲we should🪲🪲eat🪲the🪲beetles.🪲🪲🪲Obviously.🪲What🪲🪲else🪲are🪲🪲🪲our🪲🪲beetle-suppression🪲🪲🪲🪲crews🪲going🪲to🪲🪲eat?🪲🪲It's just🪲a🪲🪲🪲temporary emergency🪲🪲🪲measure.🪲The🪲🪲tide🪲🪲🪲🪲of🪲🪲🪲beetles🪲🪲will🪲🪲🪲recede🪲🪲once🪲🪲🪲🪲they🪲🪲🪲run out🪲🪲of🪲🪲🪲plastic🪲🪲to🪲🪲eat.🪲🪲🪲🪲🪲
🪲🪲🪲🪲Unless🪲🪲🪲they🪲🪲🪲🪲🪲turn🪲🪲🪲🪲🪲🪲🪲🪲to🪲🪲🪲another🪲🪲🪲🪲🪲food🪲🪲🪲source.🪲🪲🪲🪲🪲🪲
🪲🪲🪲🪲🪲🪲Of🪲🪲🪲course🪲🪲🪲🪲the🪲🪲🪲🪲🪲beetles🪲🪲🪲🪲🪲🪲🪲🪲wouldn't🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲eat🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲us.🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲Obviously.🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲🪲
A beetle like that, you wouldn't eat all at once.
That was strangely disturbing. Good job.
Big Food has been feeding us bugs for decades. Ground red beetle is the standard red food coloring.
Half of me loves the beetle one and half of me is thinking, if 20 years from now we're looking back on this and one of these has gone horribly wrong it's going to be that one.
step 1: use plastic for almost everything, everywhere
step 2: breed beetles that can digest plastic
step 3: beetles escape from the lab, civilization collapses
Beetles mutate to digest crude oil, quickly consume all remaining oil reserves.
How would they get to the deep sea reserves?
Yellow submarine.
This comment is amazing
There are already many species of bacteria that consume crude oil. They're all over the floor of the Gulf of Mexico, for example, where oil seeps from the sea floor and has for umpty millions of years. They're sometimes used to hasten the breakdown of oil when it's spilled accidentally.
Estimated total plastic ever produced: c. 8.3bn tonnes (https://www.unep.org/interactive/beat-plastic-pollution/#:~:text=Researchers%20estimate%20that%20more%20than,landfill%20or%20the%20natural%20environment.).
CO2 emissions from burning fossil fuels per year: c.35bn tonnes (https://ourworldindata.org/co2-emissions)
So about three months equivalent of fossil fuel burning - not really a material consideration for a one-off even assuming a full plastic to CO2 conversion.
I think the main concern is less the CO2 and more the plastic not being plastic anymore.
Thank you. (Hmm I think there is a math mistake here. One ton of CO2 is 12/(12+2*16) tons of carbon, call it 1/3, so total carbon in plastic is equal to carbon (w/ hydrogen) that we burn every year. We should just burn all plastic and get whatever energy we can out of it.
Yes, you're right, I didn't take into account the combustion chemistry (albeit it's not an order of magnitude difference).
You joke about burning plastic (I assume) but some forms are reasonably good feedstock for energy from waste power plants (assuming you've got all your filters up to date to clean out the toxic emissions resulting from it). That said, it's not a particularly efficient fuel by any means, but it's just a way to get some use out of what would otherwise go to landfill.
I specifically came here to say the same thing - these are generally great, but the plastic beetle one seems the most perfect fit for this kind of grant. Some smart biologist just needs a bit of funding for some equipment to keep beetles in; low probability of success, potentially hugely impactful.
To address a comment below, I'm not too concerned about catastrophic risk on this one as the plastic itself is unlikely to be nourishing enough to sustain explosive population growth. Perhaps the ideal outcome is that there is some kind of cheap spray that makes the plastic much more digestible to the beetles, thereby controlling spread (but that's a problem that only happens if this project is *wildly* successful).
I'm not even sure the probability of success is very low...
Covid-19 has boosted my skepticism of establishment government & science so much as to make me fear for the entire EA project - I feel like if we don't find a way to address rampant demosclerosis nothing else matters, nothing else will work. I'd feel a lot happier if any of Scott's grants or Zvi's grants explicitly addressed the issue. Or if I had an idea of my OWN to address it - I might then have applied for a grant myself!
I just keep noticing that in areas of concern for which people successfully "raise awareness" the government's response is to build a bureaucracy to "address" the issue...and that bureaucracy near-inevitably makes the problem WORSE.
The TSA, created to make travel statistically safer, makes airports SO "safe" as to become so inconvenient/expensive/annoying it causes more people to drive instead of fly. The increased driving means more car accidents; net effect (at least on the current margin) is to make travel statistically less safe.
Nuclear power regulations, created to make power generation safer, require SO MUCH "safety" that nuclear power plants are unaffordable and take too long to develop, so states end up relying more on coal/oil/gas for baseline power. Which is more polluting and kills more people (mining accidents and air pollution); net effect is to make power generation less safe.
The FDA/CDC/WHO, created to make us healthier, have killed over 100,000 Americans in the last few years (and a great many foreigners) by postponing vaccine approval, by preventing cheap instant covid tests or challenge tests, by not letting firms update vaccines to new variants or letting them optimize dosage levels to produce more doses or letting them optimize vaccine timing to improve effectiveness etcetera; net effect is to make us less healthy. (and fewer of us alive).
If we can't fix this dynamic I have to expect the result of *successfully* bringing more attention to nuclear war prevention or asteroid strike prevention or unfriendly AI prevention or what-have-you is that once we show the way - once we demonstrate progress along a promising path toward improving things - our government will make these issues a priority and form an Official Department to be in charge of each issue whereupon said departments will end up making THESE problems worse too.
We might be better off trying to fix these kind of problems IN SECRET to avoid regulatory attention! A private EA cabal with secret symbols, handshakes, meetings. Like the FreeMasons.
I hadn't thought about the net effects of nuclear power regulations and TSA hassles. But you have to think about the public reaction to those: how would the public feel about having a nuclear reactor built if the government wasn't endorsing it as safe to an extreme extent? And would people be willing to fly in a post-9/11 world without the security theater of TSA? It's true that the regulatory agencies may have swung the pendulum too far toward safety, but the public probably wouldn't accept is as far the other way as you might like.
Without the TSA (especially once we privatize the airports and air control) it would be between airlines, their insurers, and their customers what level of security is chosen. Security measures that aren’t cost-effective (eg taking off shoes because one crazy guy once tried a thing involving shoes that failed and couldn’t possibly have worked) would tend to fall by the wayside. Measures which *actually matter* (eg, hardening the cockpit doors) would tend to stay. The reason the dumb measures would go away is that companies could offer lower prices by getting rid of that stuff.
Personally I would pay EXTRA for a flight with NO SECURITY AT ALL at the customer level. Let people walk straight to the gate without lining up anywhere at all. Let ‘em even have open-carry or concealed-carry guns if they like!
The funny thing is that this level of security *already exists* in the form of small-airport private planes. Very rich people CAN get on a plane with no security; it’s just us plebes in the cattle-car planes that can’t!
Allow companies to COMPETE on security and some may pick more, others less. If everybody LIKES getting felt up by wand-wielding morons after a Rapiscan pass then that will remain the standard but maybe one or two lines experiment with less to drop prices; if it works then more follow.
The main argument FOR federalizing this stuff is externality-based - if a low-security plane gets hijacked and hits a building that affects the people in the building too. But that specific kind of attack became impossible moments after 9/11 and remains so; it’s just not worth worrying about.
9/11 revealed the massive externalities possible from unsecured plane flight, and not just hijacking - also bombing, or bargaining with passenger lives. That security theater wasn't just for passengers' peace of mind, but for everyone in a building that might be struck by a runaway plane. I don't think the mass public would tolerate a full libertarian approach.
Also, how are businesses even going to learn which security measures are effective and which aren't? Counting the number of terrorist attacks that happen under each security regime? I don't think that would be tolerable either.
Fun fact about the TSA: The transporter booth you have to stand in spread-eagled every time you want to board a flight is an X-ray backscatter scanner. When the TSA introduced them, I got a copy of the paper cited to justify the claim that they were safe (which sadly I can't discover the title of now). The paper had shown that if you divide the amount of ionizing radiation shot at a passenger by the volume of the passenger, you got a radiation density considerably lower than levels considered safe for that particular type of radiation.
Unfortunately, most of the radiation is absorbed or scattered by the first 1.5 cm of human, so the radiation density in the parts actually irradiated (mostly skin and subq fat) is much greater than the average over the entire body's volume.
EDIT: BUT not nearly as much greater as I thought when I began typing this. Approximating myself as a perfect sphere of radius 120mm, I find that the average radiation density in the outmost 15mm of spherical me, which gets half of the radiation, is only 1.5 times as great as the average over my entire spherical volume.
But why not termites? They already have the gut biome that breaks down wood.
The chemical problems are very different. The problem with digesting cellulose is prying apart the tightly-wound fibers to get access to the individual polysaccharide chains, and then cutting some of the glycosidic links (C-O-C linkages). That is, wood is chemically quite different from oil.
If you want to digest the most common plastics, the problem is that (1) they are not soluble, so it must be done in the gut by something that can digest things outside its body fluids, i.e. bacteria, and (2) your organism must be able to oxidize C-C bonds without any kind of chemical "handle" to hold onto them. (Our own mitochondria can oxidize them easily, but only if they have a carboxyl group at the head to grab onto.) Note that it can't oxidize just any old C-C bond, because of course then it would destroy itself.
Even then, it's going to be difficult to find or breed something that can break the benzene rings in polystyrene. They are very sturdy, and it is possible your bacteria, fed polystyrene, might only excrete a variety of small substituted benzenes, which would be worse than useless, since they have a decent chance of being carcinogens.
Thanks Carl, so you predict no plastic eating bugs in the near future.
Yes, that would be my odds-on guess.
I think this whole process is awesome. I am excited to see follow-ups with the projects, and would also be interested in seeing which projects weren't funded but might still be interesting, like "runner-ups or honorary ideas". The beetle project and automated / hybrid tutoring seemed very cool to me; I don't know too much about biotech. but it would be cool if someone did a project to modify bacteria/insects/algae to be more nutritious (e.g., altering the taste of crickets or consumable-bacterial-growths) / produce more energy / consume-convert wastes into useful things.
My impression has always been that 4-7 or so, if you select them well, is a superior option for small scall education creating amazing outcomes. 1 on 1 is well ahead of 1-35 and has some advantages in curriculum tailoring, though.
>Despite the growing importance of this field, there are relatively few technical biosecurity centers in the US, and the West Coast is underrepresented. This causes serious problems like poor pandemic readiness, limited understanding of biowarfare risks, and the biosecurity grad student who I'm dating living 3,000 miles away from me.
One of these problems is not like the other ones 😂
Here in Boston, to me it seems like there are lots of biotech people but few biosecurity specialists. I think they're concentrated around Washington DC.
Washington D.C. makes sense. Why is Boston such a biotech hub? My initial hypothesis is proximity to a certain pair of prestigious universities, but I imagine there could be other reasons.
Development at Kendall Square (right behind MIT) has been targeted to biotech for a few years now. (This may be more of a description than an explanation, or half/half.)
The universities started it, but then comes clustering effects and targeted development. We just lost a beloved long-time food market to new lab space. What's really weird is driving around Cambridge and Watertown seeing billboards advertising reagents, next to the usual ads for fast food.
>What's really weird is driving around Cambridge and Watertown seeing billboards advertising reagents
Oh yeah, the first time I saw it, I was shocked.
The other Cambridge (uk) is also a biotech hub.
the universities help, but we also have a number of really great hospitals.
Before this Pandemic, Georgetown and Johns Hopkins were the only two universities that had a decent program in it I could find. Not surprising the experts would concentrate in DC, if that's where they studied. I predict many more will appear in coming years.
Scott is usually so private! I'm so surprised he snuck that in there!
Congrats Scott!
I'm glad to see several worthy projects got funded! I'm especially looking forward to Michael Todhunter's results.
You are a cool dude
This is delightful to see and makes me feel optimistic about the rationalist community having a growing, positive impact on the world.
> This is the part where I post applications publicly on the blog (if you gave me permission) and readers can look at them and decide to support them or not. About 500 of you gave me permission to do this, and your applications together total about 1,500 pages of text. Substack probably won’t let me write a blog post this long, and you guys won’t read it even if I do, so I’m still thinking about how I want to handle this. Please give me until sometime in January to work something out, but rest assured, I haven’t forgotten about this.
I have a suggestion here. You have 500 ACX++ grants. I'd first weed out the ones you find objectionable. That not only do you not want to fund but you think lack merit. Then I'd pair them in small groups of highly different proposals. Then I'd let each person do a paragraph in a public Open Thread and tell them to be there to answer questions. The ones who won't agree are forfeiting their right to be included in ++.
I suspect this will cut down the number significantly. Those that remain will be rationed out over a year. But in exchange the grantees will get more exposure than people hunting through 2,000 pages or a huge database. And if you like it you can do this on a rolling basis instead of making it into a huge Christmas nut to be cracked like you did this year. (It'll also give you more content for however you weigh that.)
Seconding this. You need some kind of filter and this one has several advantages.
Thirding. As one of the 500 people, I would be in favor of some kind of “write an additional paragraph explaining your project” (perhaps a condensation of the original proposal) and I definitely want to second of the above commentor suggestion of an open thread participation for the AC X plus plus people.
Agreed. As someone with a rather speculative proposal that consumes almost no money apart from my own time, I was planning to post in an open thread if I didn't get "funded" anyway. (I already noted in my application that attention and "signal boosting" is more relevant for my project than money)
500 paragraphs is a whole lot of paragraphs. Even if half of them are meritless, 250 paragraphs is a whole lot of paragraphs. And lots of people never bother with comments on posts, especially open threads.
I think a better idea might be a single post, sorted into categories, with a very short sentence description of each grant as a link to the pages of text in the application, hosted somewhere else. It'd be a very long post, but if it's sorted into categories, people can peruse it for ideas they find interesting, link to the outside information to read more on it, and decide what to do. And maybe the links could include a place to ask and answer questions from founders.
It would probably take a lot of work to set this up, but Scott could probably pay someone to do it.
I like the idea, but this approach would require someone to read 500 ask-me-anything threads, summarize and interpret them, and then pick winners. I recommend also using some other web platform that can automatically collect votes or ratings from paying ACX subscribers, in order to circumvent the Scott-reads-10,000-comments step.
(Paying subscribers in order to reduce potential fraud, to get more paying subscribers, and because having the people who provide the money get to say where it goes would, by some long-standing but now aggressively-disputed standards, be more just, and motivate them to think more carefully.)
Earlier today Matthew Yglesias had a tweet about how despite widespread belief that numerous government institutions failed during the pandemic, there's been virtually no legislative effort to change anything. So I'm really glad to see ACX putting $100k toward a better FDA. Trying to fix what's broken instead of just having fun complaining. Hell yeah!
Oh, and as a biologist I want to strongly endorse the cell culture media testing project. This is the exact kind of thing that is 1) really important and 2) in a kind of hole where it's not the kind of thing any science funding agency would like to fund.
Strong agree - this one is tied with the beetle one for best application in my view. Making the process of doing science less onerous/painful is huge (not to mention the waste in having extremely smart people spending >50% of their time fiddling with petri dishes and whatnot, and another 20% writing grant applications...)
Kinda disagree on 2) Nsf funded Dr block $3.5million, part of which is on cell culture media optimisation.( https://www.ucdavis.edu/food/news/uc-davis-establishes-research-training-cultivated-meat). Will be intresting to see how this builds on existing academic work that uses genetic algorithms and robotic handling machined.
https://www.nature.com/articles/s42003-019-0296-7
Optimization of muscle cell culture media using nonlinear design of experiments
Agreed.
A lot of the stuff that did fail is just not sexy enough for politics to care. Stuff that might be handled by MY's "secret congress" but even there not a huge priority.
It seems like the incentives in American democracy don't actually work toward effective government on the national level. If your local city council can't keep the streets clean, they might get voted out, but very rarely will people vote for a different congressperson or president based on how competent the FDA is.
Agreed.
Sadly, 100k would be a drop in the bucket regarding FDA lobbying.
love everything about this, thank you for putting in so much effort Scott <3
This is super amazing work from all involved.
This is inspiring!
Especially so when you consider that the initial announcement was made on November 12
"SD, $5,000, to fund an honors' thesis on neutrino research. S is an undergraduate who wants to work on neutrino physics with one of his professors, but needs outside funding to be sure it will work."
SD, if you're out there and want any outside help / input / collaboration, please feel free to reach out. I'm a particle physics postdoc (mostly working on dark matter), know a thing or two about neutrinos, and think your idea is super interesting and useful. If nothing else I can connect you with people working in neutrino physics who may be more helpful than me. You can reach me at
joshaebyATgmailDOTCOM
SD here! I've emailed you - thanks for the help! I always, always appreciate advice from people who have been doing this for longer than I have.
If anyone else has any input or just wants to talk about the project, feel free to email me at sd10(at)williams(dot)edu (maybe should have asked Scott to put this in the main post but I figure I'll do the self-promo thing at some point closer to the time I begin my work, which should do the trick).
Again, thanks a bunch!
I've worked on various neutrino experiments during my (ongoing) career as a particle physicist, though neutrinos aren't my primary focus. Regarding your idea, i just wanted to make sure you're aware of WATCHMAN: https://arxiv.org/abs/1502.01132, http://svoboda.ucdavis.edu/experiments/watchman/, https://www.sheffield.ac.uk/physics/research/particle/neutrino/watchman. (Meant to be informative, not discouraging - new ideas in this realm or contributions to the existing project are both valuable!)
Yeah, I actually mentioned WATCHMAN (and Nucifer, although I don't think they take Americans) in my grant application. I want to join them ASAP or at least work in parallel with them. Thanks!
Well done Scott. It doesn’t look like being “outed” by the NYT was a bad thing in the end. I hope to do some small contribution. I’ll also sign up here.
Thank you for this initiative, Scott. It has surfaced many worthwhile endeavors. It is admirable that you took on all the related effort, with the help of many supporters and funders.
> if they're smart enough to attempt this project, they're smart enough to know about XYZ Grants which is better suited for them (...)
This sounds like the grant reviewers assume that "likely to succeed at project" is very strongly correlated with "able to navigate searching for funding". For some of the projects that got awarded this sounds eminently sensible to me (e.g. the affect public policy things), but not for others (e.g. most of biology-adjacent projects). I wonder if I'm wrong about the latter (i.e. if success at projects that ostensibly are not about influencing people is strongly correlated with ability to navigate social mazes). Thoughts?
I think the grant reviewers already took that into account because they're domain specialists, judging by (for instance) the remark for Todhunter's grant: "I'm not sure anyone will use this, except for me personally I WOULD LOVE THIS SO MUCH", seconded by commenter and biologist Ivan Fyodorovich above: "I want to strongly endorse the cell culture media testing project. This is the exact kind of thing that is 1) really important and 2) in a kind of hole where it's not the kind of thing any science funding agency would like to fund"
The bigger problem I have with this line of thinking is the assumption that
> if XYZ doesn't fund these people then that's strong evidence that they shouldn't be funded
If Scott doesn't fund an applicant, is that strong evidence they shouldn't be funded? It seems obvious the answer is no given the limited amount of people he can fund and the large number of applicants. Why should this not be true for other grantmakers? The assumption here seems to be some variant of the just world fallacy where P(is funded | should be funded) = 1, but it seems more likely that this number is quite low.
I share this concern. It can be obviated with a note saying "Sorry ACX didn't fund you, but you should seriously consider applying for an XYZ grant".
I realized this actually makes sense IF it means "since the applications get passed to ACXG+ anyway, let's pick up the ones they're unlikely to fund", but otherwise I agree it's weird (see my top-level comment below)
awesome
2021: the year when a blog about medicine, rationality, and fighting Moloch funded a research program about breeding beetles.
If they're good enough for God, they're good enough for me! https://manoa.hawaii.edu/exploringourfluidearth/biological/invertebrates/phylum-arthropoda/weird-science-inordinate-fondness-beetles
Seriously this is one of the ones I'm most interested in hearing follow-ups about
I know right? I'm going to continue joking about beetle eugenics but the project is super interesting (also) for non-meme reasons.
Beetles which can biodegrade plastic strike me as worryingly double-edged. Plus side, plastic not hanging out in the environment forever is probably better for us, and for the global ecosystem at large. Minus side, aren't there things made out of plastic where its non-biodegradingness is essential to their function? Is anyone in charge of knowing the full list of things which might fail catastrophically if plastic-biodegrading beetles were released into the environment?
My thoughts exactly. Termites (wood eating insects) are a huge problem in some parts of the world. It would be awful if similar issues arose with plastic.
"Society will be destroyed by a bug!" ... no, a literal bug.
We could always go full on science apocalypse and have an island where we ship plastics loaded with beetles that eventually results in our destruction when they learn to cover the distance to the mainland, possibly through micro-evolution. There are worse apocalyptic scenarios. This one seems almost fun by comparison. Michael Crichton would have to be simulated on a super-computer to write the novel.
According to the application, the beetles can only get a small amount of their subsistence from plastic and wouldn't be very good at this in any case. What the researcher actually wants is the gut microbes, with the hopes that once we understand them better we can figure out some way to use them at scale. Breeding the beetles in a plastic-enriched environment is a means to get better microbes. The microbes presumably can't escape the beetles without human help. At least this is my current understanding.
I can't help side-eyeing the "presumably" there. If plastic-biodegrading bacteria gained the ability to survive independently of the beetles, there's an awfully rich ecological niche waiting for them out there.
Seconded.
This is more worrisome than what I originally thought: first I thought, "Well, if we need boats or landfill liners or nuclear waste containers or whatever to survive the beetles, I guess the beetles can't eat them without oxygen, and we can always lace them with insecticides." I don't insecticides that will work that well with bacteria, and gut bacteria are presumably already anaerobic. So maybe we'll have to resort to teflon or geopolymers or something in order to build bacteria-proof things in the Grim Beetle Future.
On the plus side, the bacteria probably smell better than darkling beetles do.
(Just to be clear, I think the research should definitely be done; I'm just not sure whether deploying the results in the wild would be good or bad.)
Glass and metal still work good.
I mean, anaerobic bacteria TYPICALLY are outcompeted by aerobic bacteria in situations with plenty of oxygen. Specialization is a big deal for bacteria, so I would expect it to be much much harder to get gut bacteria to thrive in a place they could digest plastic for us than it is to make efficient plastic-eating gut bacteria in the first place.
A somewhat related alternative - making some plastics/materials more biodegradable eg. The BioMask https://www.canadianshieldppe.ca/blogs/the-canadian-shield-blog/the-biomask™-infographic
I don't understand, isn't this just a much *better* outcome than the one projected in the grant? It seems increasingly clear that plastic was a terrible mistake from the get-go, and if Scott's grant intervention leads ultimately to a world where we have to go back to glass, steel, ceramics and tin with a small dash of bakelite (a resin) for flavor because any plastic will just get devoured by omnipresent microbes before you have time to use it, that will be a massive improvement on all fronts.
My prior is that plastics are an integral part of modern life in such a way that we couldn't have much of what we have *at the price we have it for*. In other words, much of the world would be much worse off without the existence of plastics. The "terrible mistake" seems to be not using plastics judiciously, not that plastics exist.
That being said, I have no special insight or knowledge and could be convinced otherwise.
"My prior is that plastics are an integral part of modern life in such a way that we couldn't have much of what we have *at the price we have it for*."
I think this is almost certainly correct (notably, every post-celluloid data storage medium including film that doesn't burn your house down is plastic-based), and I'm willing to bite that bullet. To me, plastic seems like our time's equivalent of asbestos. It's no good telling people how practical it is that once your gloves get dirty you can just throw them into the fire and they're sparkling clean again.
What would a world without plastic look like?
Non-OBSF*: "The Great Fog" by H. F. Heard, in which a very fast mildew takes over the world and rots everything that can rot. There's a fog which makes smoke stay and be intolerable. No wood, no paper... as I recall, humanity is limited to using glass and stone. Fortunately, the fog moderates temperatures and produces sufficient stuff that's good to eat.
* In rec.arts.sf.written, it was a custom to add OBSF (obligatory sf) to posts that weren't about sf.
And SF in this case is... science fiction? So you're pointing out that this is science fiction, even when you're not obligated to point it out?
Yes, SF is science fiction.
More that I'm pointing out that this is science fiction, even though I'm not obligated to include it. I tend to see the world through science fiction.
I reread the story. My description is mostly correct. except that I hadn't noticed that the climate-changing mold was the result of an effort to produce a fat-producing mold. Also, there was a big die-off of the people who couldn't handle living in a very humid environment. It was published in1943, so there was no notion of a high proportion of people dependent on medicine.
Well, among other things, no hip replacements, no arterial grafts, no cardiac catheterization to stop heart attacks, no single-use medical instruments so loads more iatrogenic infection and death from same.
The result of the great fog was presented as utopian, and it might actually be a net gain.
People couldn't see to go faster than four miles an hour, though I think Heard was underestimating how well people could go with cleared, familiar paths. It would still be running, at most. Horses couldn't see as well as people.
No war, no car accidents, nothing much to accumulate. Fine stone musical instruments might be about it for valuable items.
Note that this was written in the 1940s, so most of what you're mentioning didn't exist yet. If people are moving more slowly and not using a lot of edged tools, there would be fewer accidents of all sorts.
They live pretty much on food produced by the mold, so a lot of the less healthy food isn't there, and the premise seems to imply that the moss food is at least nutritionally adequate.
Any and all plastic could be totally re-used if only we chose to invest a little bit of energy. It isn't currently economically cost-effective to do so, but if we really wanted to get rid of plastic we could. We just lack the will.
Where can I read more about this? I was under the, likely mistaken, impression that we know how to reuse some plastics, but for vast majority of them once they are polymerized we don't have better ideas than "burn them at high temperature to recover simple compounds".
So warning that I am not a chemist, so I probably don't have all the details quite correct, and am just giving my semi-educated opinion. Burn them at high temperature is basically what I am talking about, but why is that such a problem? Well, because it ends up costing energy.
The way we make plastics now is we start with petroleum, which contains lots of molecules that contain more than three carbon atoms, break them down into mostly molecules that contain less than three (ok, we keep a lot of hex rings too) then re-combine those smaller molecules into large chains of polymer. There is no particular reason we can't start with carbon-containing trash, and use similar processes as we use to crack/refine oil to turn it into the same chemical feed-stocks we use to make plastic.
The reason we don't do this is energy. Oil is actually insanely cheap. I know Americans like to complain when gas costs more than $4 a gallon but. $75 for a 55 gallon drum of oil is extremely cheap. Do you know how much a 55 gallon drum it? It's like the size of a person. If you could buy jugs of oil in the supermarket it would be cheaper than almost anything else, except maybe bottled water.
Because is so cheap, oil refineries burn a substantial amount of their cheapest products to generate the energy required to form their more expensive products that are used to make things like plastic. We can't really do this with garbage because it would generate tons of pollution and garbage isn't as energy dense as petroleum. But if we had much more abundant, cheap energy we could. We wouldn't even need to burn the garbage, we could heat it in a kiln to recover and re-use essentially all of the carbon and other toxins.
I think you can get net energy out by burning plastic... (Hasn't everyone burned plastic by the camp fire?) I'd like to see plastics made that burn better. (Less crap in exhaust 'smoke') Think of plastics as a side step between oil and energy. We make all plastic knowing they will be burned. Or alternative 'green' idea. Keep putting them in land fills, but change the narrative to permeant carbon storage. Burn them or bury them... that's all I got. Putting 'em back into the ecosystem seems to be the worst thing to do.
Re Beetles: Why not termites? They already have the gut bacteria to digest wood.
Calling burying plastic permanent carbon storage seems pretty disingenuous - the carbon in it came from oil, and only helped us in its use as plastic. Burying it instead of burning it is the same idea as leaving oil in the ground.
It's a little more complicated than using high temperature, because many plastics contain heteroatoms (e.g. Cl, S,N) that can create noxious compounds (e.g. HCl, SO2, NOx) even at very high temperatures. So you need to add some fairly expensive flue-gas capture technology to pull out the compounds you don't want in the air -- and then find a way to dispose of them, or someone who wants to buy them. (It goes without saying that low-temperature combustion generates even more evil compounds, PAH and such.)
Really? You know how to remove nurdles from oceans, beaches and the innards of fish? Harmlessly remove microplastics from fetuses? I think you should have submitted an ACX grant application if this is true.
I think GP is talking about the problem of "I have a bucket of plastic trash and want to convert it into usable plastic of the shape I desire" and not "I have a lake that contains a bucketful of plastic in it and want to extract it".
I don't know who you mean by GP here, but I think we must be talking past one another. Let's recap: I asserted that a bacterium which destroys all plastic being released into the wild would be a net good despite the infrastructural damage it would cause, due to the fact that the bacteria would also obliterate all the horrible shit like oceanic nurdles and microplastics presently plaguing ourselves and the rest of the planet. Eric P. replied to this statement of mine with the claim that we can already recover any and all plastic. I replied to that claim in turn, expressing a skepticism which I assure you is very real and sincere. In other words, we're talking about something closer to your scenario #2 here.
I'm still under the impression that Eric meant #1 and not #2 (nonwithstanding the context). I eagerly await Eric's confirmation, refutal, and/or elaboration.
E: By GP I meant Eric's comment (grandparent of my comment).
Yeah, we were talking past each other. I don't have a solution for removing it from the environment, only re-using it industrially.
But while I hate plastic in the environment as much as the next person, I also don't think it is the most important environmental problem to solve. Micro-plastics are disgusting and surely harmful, but I don't believe they are so harmful that we should regret having ever used plastic. It will all eventually break down and settle into sediments. It will take hundreds of years, but the earth is old and life will go on.
The best way to remove plastic from the environment is to stop putting it there, and in my opinion the best way to do that is to develop a robust industrial system for re-processing it.
There's a lot of different types of plastic. I'd be surprised if there's not some high-grade polymer like UHMWPE or something with a bunch of fluorines that the beetles can't get at.
Would it be possible to engineer bacteria so that they were good at digesting micro-plastic, but not strong enough to pull bits off of plastic objects? Or at least cause deterioration very slowly?
And would it be possible to make the bacteria salt-dependent? This would help with cleaning up the oceans, though not fresh water.
I'm not an expert, but based on my limited knowledge of the subject, I don't think that's likely to be practical, because on the scale of a bacterium, basically any piece of plastic represents a large ragged surface. Even microplastics are much too large for bacteria to subsume, they'd still be attacking them in the same way that they would a larger piece. The more weathered the piece of plastic, the higher its surface area to volume ratio, so the more easily it ought to be attacked, but any plastic should be vulnerable to weathering.
Also, once eating plastic becomes a viable ecological niche, any bacteria which is better at it will tend to outcompete ones which are worse at it.
Oh, well. It seemed plausible, but I didn't know enough about bacteria.
This is amazing. I mostly write to praise Scott and the team. But can't help but to contrast with the many philanthropies I know well....such exciting ideas/people he has curated, in such a short period, with so little red tape.
You refer to Nils Kraus as N in a later sentence. Not sure if you forgot to anonymize, or just a typo.
Thankfully, the error was the other direction - he was originally anonymized, and then gave me permission to name him. I've corrected this, thank you.
For peace of mind in the future: a way to avoid mistakes in the worse direction is to compile a list of terms that must not appear in a post (in this case, names of people who want to remain anonymous), and arrange for automation to scream loudly if a draft contains any of them. (Not my idea, this is a pretty common approach to ensuring that various not-intended-to-be-made-public things were redacted from a release of some software.)
This is just incredibly impressive that you pulled this venture out of thin air, roped in all the support and expertise you needed (or could find), let it grow way beyond what you anticipated, and then pulled off this very wide ranging list of worthy projects to support. Wow. It warms my heart to see.
The first part of my career involved a ton of fundraising, in and out of academia, and working with foundations on various joint endeavors. It can be a pretty slow-moving, trend-following world. But also, some of the expertise (to raise money well and to give it away well) are for sure real skills. It's astounding to me that you pulled this off in such a short time, and I hope you will write more about the process of it.
I'm interested in the question of what funded groups might want to talk to each other about. I know a lot of foundations try to start up those kinds of cross-pollinating conversations and that on the side of the people funded, they sometimes say "yes" to whatever the funders propose in the hopes it will lead to more funding, even if they think it's a waste of time. I think funders get impatient encountering the same kinds of organizational stumbling blocks in the groups they fund (I consult for organizations so I get that too). Some kind of honest assessment of when those networking opportunities are seen to be beneficial by the people in them would be nice to have.
Thank you for helping to bring so much good stuff into the world! I hope beyond the exhaustion it sounds like this produced, that you have a real sense of satisfaction in it. It's a remarkable thing you did, in a whole different direction from your writing and your psychiatry practice.
"Why would somebody working on biochemistry want to talk to someone working on political activism just because they got a grant from the same person?"
Going to conferences of people who all think and dress the same way and work on the same issue is beneficial in one way; going to conferences like EA or SciFoo is beneficial in a different way--optimizing for serendipity.
Yeah. People who received an ACX grant probably have many more things in common than two randomly selected people (even if we restrict the selection to compatriots). But also, sometimes just having common knowledge that everyone in a certain group actively wants to meet and chat with everyone else is already valuable.
It would have been useful to know that one option was funding, of some sort, for politics/political careers, not necessarily for me personally. There are some interesting Senate seats up this cycle that might be viable options for third party candidacy who knows what they are doing where the existing brand name options are mediocre or in conflict with the kinds of things you, Scott, care about.
There are, given the divergence in my interests vs Scott, more very exciting projects on this list than I would have expected. Though part of that may be all the extra money and effort many people contributed.
I'll also somewhat second the comment by someone that people involved in grants perhaps associate being able to get grant funding too much with having good ideas, as far as the "why are they asking you and not this other potential option for money".
I hesitate to ask because it seems like surely you've thought of this, but....whatever tax status your grant fund has, have you evaluated if giving to politicians impacts it?
Double-OF-COURSE-YOU-RESEARCHED-THIS-energy: do the laws in Australia allow you to give large sums to political actors, and the actor to accept it? I do note that the person is not running for office themselves, but, ya know.
The grants are coming from two pots: my money, and external funders' money.
External funders are donating through a tax-deductible EA Fund.
I can't do this for tax reasons, so my donations won't be inherently tax-deductible. I plan to cover a combination of whatever grants that the EA Fund can't, plus the recipients who are inherently tax-deductible themselves (eg charities).
Answer to the second part: yes, Australian laws do allow it.
It’s possible I’ll need to register my activities under the Foreign Influence Transparency Scheme (https://www.ag.gov.au/integrity/foreign-influence-transparency-scheme) but this doesn’t prohibit any such activities, just makes it public where the money is coming from. And clearly there is no difficulty with that being public in this case.
Thanks to both for the answers to my worries.
I’m sort of curious if there are any ideas that you specifically didn’t fund that you’d feel comfortable talking about
I hope to showcase all the ideas I didn't get to fund in Grants ++.
Looking forward to it :)
Very glad to see approval voting (AV) get funded (and promoted) on ACX. We see ranked choice voting (RCV) making inroads around the country, but there seems to be very little awareness of the pathologies of RCV that make it inferior to AV. [This is based on my remembrance of analysis on RangeVoting.org, which smells trustworthy to me.]
Approval voting is probably the best option, that could reasonably happen anyways, for American elections. Something of a limit to the problems that can be solved solely by better voting but at least at the municipal level it is pretty big.
Agreed.
What are the pathologies?
The pathologies of ranked choice voting are that in certain weird theoretical combinations of circumstances, voting for your true preferences makes your preferences less likely to be satisfied.
The problems with Approval Voting are less obvious, because it's Never Been Tried (actually it has been tried, in a few low-stakes elections but not much that we'd learn from). The upside of Approval Voting is that it allows a charismatic centrist to be elected... the downside might be that it precludes anyone else from ever getting elected. In a world where Approval Voting was the norm, what kind of candidates would we get? Would it be a "race to the centre", where candidates work hard to avoid ever unambiguously taking any kind of position that anyone at all could possibly disagree with? And would this be better or worse than what we have now?
Good points. I look forward to seeing it tried!
>The pathologies of ranked choice voting are that in certain weird theoretical combinations of circumstances, voting for your true preferences makes your preferences less likely to be satisfied.
By Arrow, is that not true for all (non-dictatorial) voting systems, including approval voting?
I think you're correct. The way you can kinda claim otherwise for approval voting is by changing the model of voter preferences (instead of them having preference orders, claim that they have sets of acceptable candidates).
Also: a nit: Arrow also requires determinism, but all the systems we talk about are deterministic.
(Trivial counterexample of a nondeterministic system: choose a voter at random and do as they wish. It's not dictatorial, because the voter is chosen at random.)
Yes, it is somewhat true for all (though Arrow iirc only applies to ordinal voting methods, which AV is not). But for AV, approving a candidate will never help a candidate you did not approve (though it may hurt a different candidate you approve of more). With IRV, your true preferences can potentially elect your least liked candidate.
In a highly polarized environment, you may have a candidate who is everyone’s second choice but no one’s first choice. They get eliminated in the first round of evaluation in the RCV algorithm.
RCV still has the benefit over our current system (PV) that the spoiler effect is avoided. So those third parties can exist, but they still need to beat the conventional parties in popularity.
With AV on the other hand, you're not forced to rank, only to approve, and so any acceptable compromiser has a huge advantage in an otherwise polarized election.
I'm not sure IRV completely avoids the spoiler effect.
With plurality, if a 3rd candidate is at 1-2%, it is not a big deal, but as soon as it is 5-10% it becomes a liability because of being a spoiler. Thus it is hard for them to grow large support. As soon as they start to grow, they become a liability and voters will strategically vote against them.
IRV does handle this better at low ranges of support. A candidate with low enough support to place last won't be a spoiler, but there is still a center squeeze and other cases where there could be a spoiler. This is because IRV does not satisfy Independence of Irrelevant Alternatives (https://en.wikipedia.org/wiki/Independence_of_irrelevant_alternatives)
I mean, doesn't it? in the limit where all the Irrelevant alternatives are truly identical, all their preferences will flow to one another and one of them at random will get all of the votes of the collective bloc of identical candidates. In the real world, of course, the candidates aren't truly identical, but then it is no longer obviously bad that eg. the votes for the Communist don't all preference the Socialist when the Communist party is eliminated.
Note that irrelevant does not mean identical, though it is true that a system that does not satisfy IIA will be subject to 'candidate cloning'. So your intuition is correct, IRV does a much better job at candidate cloning than plurality, but it still fails IIA.
In the communist/socialist example, you are correct it wont be an issue *if* the communist support is relatively low (i assume you are assuming this, since you say they will be eliminated). If their popularity grows, there is a potential for a center squeeze.
Consider this toy example (numbers are completely made up by me to illustrate. I am not implying any political preference here, just following the example provided)
Say there is an IRV election between Socialists and Capitalists with the following support:
Socialist: 55%
Capitalist: 45%
But what if the Communist party decided to enter... if they only have 1-5% support, their votes transfer to second choices, which I think is fair to say will be Socialist, so no difference in outcome, as we noted above.
But say instead, they drew a lot of support, and we had
Communist: 30%
Socialist: 25%
Capitalist: 45%
Now it is the socialists that will be eliminated first. And unlike the communists, where we could be confident that all of the 2nd place support would flow to Socialists, the Socialists 2nd place is more likely to be split. If just 6% of voters were socialist, but not full communist, and put Capitalist as 2nd place, then in the 2nd round we get
Communist: 49%
Capitalist: 51%
This is IIA failing (also 'center' squeeze' with regrads to IRV). By entering the race, the Communist party caused the winner to go from Socialist to Capitalist. This is a worse outcome for them.
Here is a decent independent rundown of the limitations of both https://www.youtube.com/watch?v=yhO6jfHPFQU
A quick google lands me here:
https://www.fairvote.org/electoral_systems_rcv_vs_approval_voting
Which suggests that approval voting has more problems than ranked voting in getting your “first choice” in, although AV does away with rankings which is an odd way of fixing that problem - pretending it doesn’t exist.
If that is even the problem (I’m not clear what your pathologies are).
As someone living in a country with ranked voting, specifically the single transferable vote, I would often vote for a first ranked candidate destined to lose. Then on the elimination of that candidate my vote goes to the next person. That person might also be destined to lose so the vote percolates up to elect somebody eventually if I kept ranking.
here is a more thorough analysis https://www.rangevoting.org/rangeVirv.html
That’s pretty badly written, and it seeks to have handpicked an example where there might be a discrepancy. In order to prove the point the piece has the voter act suspiciously against her wishes.
“In the example, suppose the 1st voter, instead of honestly stating her top-preference was A, were to dishonestly vote C>A>B, i.e. pretending great love for her truly most-hated candidate C, and pretending a lack of affection for her true favorite A.”
There’s no explanation of why she did this, but apparently the voter doesn’t get A elected if she prefers C when voting. Why is that a problem and why are we second guessing her?
And I think it’s pretty clear that most people, if they get this system at all will use the highest rank available to them, 99 then 98 etc. Which is basically ranked voting.
i agree the article is pretty dense, and not very user friendly. There are more accessible writeups out there if you are interested.
But read the example again! The voter causes 'A' to win, not 'C'. By lying and claiming 'C' was her favorite, this causes 'A' to win.
To the 2nd point, "if they get this system at all will use the highest rank available to them, 99 then 98 etc. Which is basically ranked voting."
It is still different! They have implicitly 'ranked' their preferences; similar perhaps to IRV, but the way Score voting determines the winner does not make this equivalent.
And I don't think it is clear people will use the highest ranks. They will use the highest AND lowest (since they don't want the candidates they hate to be ranked high). And this.. is basically Approval Voting.
The voter didn’t lie. She voted as she voted. Why is the writer assuming somebodies internal mind choice as being illegitimate and why is she ever going to with out prior to the count what might have happen. It’s thoroughly contrived.
Of course most people will rank 99,98,97, if a die hard Republican doesn’t do that for the Republican candidates (and I assume there would be multiple candidates from each party) then she will worry about a democrat would rank the top Democrats 99,98,97.
1. I'm not sure I understand your objection... It is pretty standard for analysis of voting methods to consider strategic voting, this is not unique to this article. If you assume no strategic voting, that makes range even better.
If it is just a general 'this is contrived' objection, then yes I don't think anyone would disagree. Contrived, or 'toy', examples are often used to illustrate how a voting method can go wrong, as they are simpler, clearer and cleaner than a more complex example.
2. Yes, the die hard 'R' can rate Republicans '99,98,97...', or can rate them all '99' (you can reuse scores). But they wouldn't rate the Democrats this way. They would give them scores closer to 0. It would be foolish to leave them unranked.
This one may be easier to read https://electionscience.org/approval-voting-faqs/
I suppose it comes down to what our objective function is. I’d rather have an official of whom 80% of the population approves than have a candidate who is 53% of the countries first choice.
You make a good point about a strategy of priority ranking candidates unlikely to win, but we face an education barrier to get most of the electorate onboard. I suppose a similar educational barrier exists for AV as well. Voters may still adapt suboptimal strategies.
It could be interesting to write down some objective functions and voter behavior rules and then run simulations to see what would happen. Obviously lots of limitations to modeling, but it could fortify the discussion.
I agree approval voting (AV) is better than plurality voting (PV). Another system that could be beneficial in the current situation is approval/disapproval voting (ADV). You could vote for or against any number of candidates. ("I am for Alice, Bob and Carol, neutral about Dave and Fred, and against Elisabeth"). Where AV is giving [0,1] to each candidate, ADV would be giving [-1,0,1] to each.
AV's advantage indeed is to allow support of outsider candidates without "wasting the vote". ADV goes beyond that, it is the antidote to polarization - it would *actively punish* the "Shiri's Scissors candidates".
The said candidates utilize the scissors as a cheap way of mobilizing a voter base; whether it turns out on election day that they picked the "50%+" or the "50%-" side of the scissors, they have *certainly* swept aside their non-scissor competitors. In terms of ADV, a scissors candidate A would cut the voters pro:neutral:contra as 51:8:41 - often against a candidate B taking the opposite scissor position, with near-opposite ratios. Until the backlash of the next election, that is, where it will be 42:7:51. (it's about topics such as "defund the police" / "prohibit abortion" / etc). Main thing is, both A and B are sweeping aside their "non-scissors" competitors C to H, who'd be around 20:75:5, give or take, and who are getting virtually no votes in PV, and doing only slightly better in AV.
In the PV system, the A and B end up sucking out all the air in the room as far as C to H are concerned. The AV system gives *some* breathing room to the C-H candidates, yes. But the ADV system *punishes* the A & B candidates. They get "51-41=10" versus their "20-5=15" contenders.
Such a system would carry an obvious bias: against the "far positions" on all sides. It would have centrism baked into the cake. Some cite this as a flaw, that it would lead to electing candidates who aren't "anyone's favorite choice" or whatever. One good look at contemporary US proves that to not be a flaw, but as good an outcome as might be hoped for. Nowadays, Rep voters have mixed feelings about Rep candidates and hate the guts of Dem candidates, while Dem voters are in the exact opposite situation. This is the natural outcome of the PV process. The AV would slightly improve this, but the ADV has the potential to replace outcomes of "50%+ mixed-feelings / 50%- hating-guts" with "10% approval / 80% mixed feelings / 10% hating-guts". Arguably the latter *is* a better outcome for democratic governance, or at least a saner outcome for something that aspires to be a unified country.
I don’t see how any of this works without multiple candidates within parties, which can happen with the single transferable vote. That would bypass the primaries.
Using approval voting for primaries but plurality voting for general elections is an interesting case. The main benefit of plurality voting is that it avoids fracturing of parties; systems in which independence of irrelevant alternatives hold tend to have many parties which then can require large and unstable coalitions. Party primaries are already rather fractured. This also avoids a complete "race to the center" that could reduce political innovation.
In general work on how one might combine different voting systems for different purposes seems underdeveloped.
You appear to paint "race to the center" negatively, and "political innovation" positively. You also placed scare quotes around the former only, while I'm inclined to place them around the latter.
What "innovation"? I consider there is hardly any dynamic in contemporary US politics that isn't a rehash of the dynamics happening in end-of-18th-century France, for instance. (if excluding how unprecedentedly blockheaded most participants are around race questions, I would argue there is absolutely nothing new at all). People more versed in history might cite echoes of those dynamics going back to ancient Rome, Greece or even furhter.
"Fracturing of parties" is an illusion created by the perspective that there *should* be a duopoly of parties, rather than the perspective that the duopoly is the degenerate endgame of the plurality voting system. The PV system has the *rewarding* of scissor candidates baked into the cake, while the approval/disapproval voting system has the *punishing* of scissor candidates baked in. Yes, "centrism", if you prefer.
Democracy often labels itself as "rule with the approval of the governed". Running with that label, consider two systems who would give the following typical *outcomes* of the process.
PV: 10% approval, 50% mixed-feelings, 40% unmitigated discontent
ADV: 15% approval, 75% mixed-feelings, 10% unmitigated discontent
Which of the two comes closer to the label of "rule with the approval of the governed"?
Also, if contemporary US is flip-flopping between "Trump US" and "Biden US" - both of them being 10:50:40 arrangements with flipping polarities - which of the two approaches would you label "stable" in a meaningful way?
I used actual quotes because I wanted to be clear I was referencing the concept in the previous comment. I don't do scare quotes.
I don't think any realistic voting system will come anywhere close to "rule of the governed." People have too varied of preferences. I am therefore generally extremely skeptical of government intervention, which I view as being nearly always and everywhere captured by special interests. That doesn't mean government should do nothing, but I don't see being elected as conferring a large amount of legitimacy. I would not be particularly supportive of democracy except it seems considerably more robust to providing for the general welfare than any other form of government, even if it doesn't do a great job in an absolute sense.
By political innovation I mean primarily policy innovation. If everyone is competing to be the least offensive, that to me suggests that they will avoid confronting entrenched special interests. That seems undesirable. I could care less about grand debates in political philosophy.
One also need only observe Italian, Brazilian, Israeli, or Belgian politics to see the impact of a high degree of fracturing.
I don't want to overly focus on the current political dynamic in the US. Trump wouldn't have been nominated with approval voting in a Republican primary.
Understood, and retracting my comments about scare quotes.
>Trump wouldn't have been nominated with approval voting in a Republican primary.
Correct. Neither would have been Hillary. I do agree AV is a step up from PV, I'm arguing why I consider ADV to be a step up from AV.
As for the entire US system having degenerated to "PV duopoly with PV primaries" in the first place, and that the only "tweak" it needs is to become a "PV duopoly with AV primaries" - because this avoids "party fracturing hell" - I remain skeptical of that view. I consider that having a multitude of parties that are elected by ADV on the general level, would reduce the creeping and seemingly unstoppable polarization of the country.
The examples of fracturing you gave *are* fracturings that evolved inside PV systems (except that they never spiraled down into a duopoly-with-primaries configuration). The ADV system would inherently affect the incentives in which any fracturings are evolving, so the comparisons need not directly apply.
About coming anywhere close to "rule with consent of the governed", if your key metric is the share of the population who does *not* feel unmitigated discontent with the way they are governed, then voting systems that punish scissor candidates provide a doubtlessly better approach than systems which reward them.
> democracy (...) seems considerably more robust to providing for the general welfare than any other form of government
In Christian and Islamic countries throughout past centuries, welfare has generally been organized by religious institutions more than by governments. While this has necessarily waned alongside the large-scale secularization trends - which largely coincide with establishment of democracies - equating government-provided welfare with democracy is confounding issues. I've been to some Islamic non-democratic countries that don't have -any- homeless people sleeping on the benches, and whose welfare stems from religious tradition more than from government-organized redirection of funds. Also, analysts like Sowell can point you to the reasons why the "blank-check government-provided welfare" that is de rigueur since the 1960s, has not really been an unalloyed good for the population receiving them, when compared to other historic welfare arrangements.
Would direct democracy qualify as rule of the governed for you?
I like the phrase "Scissors Candidate". Apt.
Gašo, would it be correct to say that, structurally, RV(2) == AV, and RV(3) == ADV? Here RV(n) refers to Range Voting with a range of n different options.
I believe your ADV proposal may be practically superior to RV(n) because Approval / Disapproval is easier to communicate than the Olympic-style scoring used in RV.
They would be functionally equivalent, yes, but I think that presenting it as zero-centered (i.e. [-1,0,+1] instead of [0,1,2])is better communicating to people the effect of their vote.
In your notation, the Score Voting system which proposes 0 "strongly disapproving" to 9 "strongly approving" plus an "X" for "no opinion", can be represented as "RV(11)"; its functionally equivalent zero-centered system would be [-5..+5], with mappings 0 → -5, 9 → +5 and "X" → 0. Whereas I'd still prefer the ADV ("zero-centered RV(3)") to the RV(11), as it has the advantages of simplicity; it is the simplest of the systems that can counteract the arguably largest pathology of PV systems, "VOTING AGAINST".
By this I mean voting for B because you *really* do not want A to win, and *not* because B is particularly well representing you - whereas voting for D or F (whom you're both reasonably aligned with) is pragmatically a wasted vote that will just let A win. This has been the most pathological dynamic for over a hundred years.
In ADV, if you're really against A, you can simply *vote against A*, and then vote for D and F. One could do a book-length presentation of major historical moments that would have gone quite differently under such a voting system, but the single-sentence summary is that it punishes scissor candidates instead of rewarding them.
I'm awed and inspired to read about so many cool projects happening around this community. Thank for doing this, Scott!
Also I'm super proud to be Beny's brother even if I don't understand enough biochemistry to know what he's actually doing :)
Such an exciting list! And I was very happy to see Seattle Approves up there -- they're addressing the meta problem of "why can't the US political process solve more of the country's problems?"
"why can't the US political process solve more of the country's problems?"
Because of human nature. There is no one single utopian perfect nation in the world, even with a different political system. You can have a *bad* system that causes problems or make things worse, but you can also have a perfect, shiny system that people will then try to bend to their own ends or ignore.
Improving bad systems is a good idea, but don't think "we'll create a perfect system and that will be the panacea for the nation's problems" because it won't be.
GIGO
Some really fascinating grants here (and a couple that I think might be ethically... questionable). Setting those aside, this is a really cool thing you've done here.
Lots of fascinating projects, thanks for sharing
> Nils Kraus, $40,000, to experiment with new ways of measuring precision weighting in humans. The precision-weighting of mental predictions is one of the absolute basics of the predictive coding model of the mind, but we know very little about it and have trouble testing hypotheses about how it works.
I'm quite interested in this; is there a way to follow this work as it develops? (And is there a good source on the current state of understanding?)
Happy to hear that people are interested :)
I can send you the proposal where I have collected some previous studies that have tried to measure precision weighting via several different ways, if you'd like.
Yes, please!
I'm one of the three cofounders of Mantic Markets, and I wanted to give a heartfelt thanks to Scott! Not just because we won a grant, or because he's namedropped us, or even because he's been so nice about us blatantly stealing the "Mantic" name...
But because it's fairly (~70%) likely this project would not have existed without ACX Grants. James and I were online friends, and we'd chatted before about collaborating on something. The existence of ACX Grants gave us an excuse to actually do something! I flew out to meet James and Stephen; and in our process of writing the grant application, we got so excited that we decided to just build out the prototype, whether we heard back or not. We've been hacking on it ever since (that's why our prototype is already live)!
Like Scott, I don't really know if Mantic will ultimately work out. (If you think _you_ know, bet on https://mantic.markets/ManticMarkets/will-mantic-markets-have-over-1m !) But this grant has already made me two good friends/cofounders, and a really exciting December so far. Thanks again for running this program!
Re: "Why would somebody working on biochemistry want to talk to someone working on political activism just because they got a grant from the same person?"
I can't speak for anyone else but I'd _love_ to be thrown into a Discord with everyone else who's gotten a grant. Beyond chatting, here are some things I'd look forward to (patterned off of YCombinator):
- Random 1:1s with other fundees, just to learn about their field of expertise
- Weekly check ins where everyone posts updates, as a shared accountability metric
- A "Demo Day", eg in 6 months where each fundee gives a 3min talk on what they accomplished, in front of potential future funders
Everything optional, of course! I'm happy to help organize; in fact, I may just go and do this for the forecasting people. We've already chatted with Nuno before, and I'm planning on reaching out to Nikos and Nathan too!
Send me an invite when you do!
Thanks Austin! I 100% agree and add my thanks to Scott for this grant that brought us together.
I am only here to say that "sufficiently advanced beetles" is the sort of Vorkosigan chaos energy that I hope there is more of in the world.
Makes me wonder if it’s the beatles or their micro-/mycobiome. There’s some exciting work happening with plastic-eating fungi.
Interesting how "sufficiently advanced beetles" mentally primed me to read that word as 'Volkswagen' on first glance.
Wow. Republicans are evil. I wish I hadn't read that.
I'm actually fairly surprised by that too. I expected it to be a set up for a joke that never came.
Seems to pretty obviously be a joke.
Scott's already admitted to hating Republicans in https://astralcodexten.substack.com/p/a-modest-proposal-for-republicans
in a post where it was just joking enough that he would have plausible deniability if anyone called him on it and he wanted to deny it.
Comedians--and non-comedians playing at comedy--like to say things they may need to later deny as part of humor.
Unless Scott also believes that AI capabilities researchers are evil too (seems unlikely), then this is obviously a joke.
I winced. It’s certainly a joke, but it’s hard to make something like that land in our environment when so many people treat all Republicans as evil.
I chalk this up to the influence of living in the Bay Area and among his tribe, where casual insults to Republicans come off as harmless political humor that nobody is hurt by, because no actual Republicans would dare out themselves by objecting.
"Other times a reviewer was concerned that if you were successful, your work might be used by terrorists / dictators / AI capabilities researchers / Republicans and cause damage in ways you couldn't foresee"
I agree that this is a joke, but to my reading it is a joke that gently pokes fun at the reviewer (who seemingly equates Republicans with terrorists). If anything, it reads to me as sympathetic to Republicans, not pointed at them.
This was my impression as well. A little eyeroll at partisanship that was bound to crop up in some reviews.
Maybe it is! The problem is that it can be read both ways, and since the double meaning is implicit and there’s nothing to read against the equivalency, it’s tone-deaf in an environment where just this kind of casual associational slur is terrifically common.
1Day Sooner staff member here—thanks so much, Scott! These grants are an awesome idea, and honestly I’m kind of blown away by the company we’re in. Good luck with your projects, everyone!
> My father has been stalled on an important research project for years for lack of the right kind of statistician
Just out of curiosity, what is it and what would the right kind of statistician be? I'm a statistician and I know a lot of statisticians. Often we feel we are short on important projects, while society is yelling that it is short on statisticians. It's hard to square that circle.
Biostatistician who is allowed to look at VA data.
I am a statistician who meets the latter requirement. I have done biostats, though I don't know that I'm A Biostatistician. Is he still looking for help?
Edit: I guess there's enough grading in level of access that my comment might not be super-helpful. Does he need someone *at* the VA who can query their database directly, or somebody who's clear to stare at such data?
He's retired now, but if you're really interested I can query whoever he passed his project to.
Thanks for the reply - I think for now I've got enough on my plate that I probably can't squeeze in another project without a sprinkle of the "Great Family" glamour. ;)
Did your Dad apply for a grant?
I got such a rush reading this. I hope that there are periodic follow up posts.
Thank you Scott at everyone who contributed to this new grant. I'm honored to have RaDVaC included in this group of creative, important, really interesting projects. This will help take our work at RaDVaC to new levels; in my view this is a big moment for vaccinology and the future of global health. We're going to do some kickass science.
This legitimately made me cry. I'm writing this with tears coming down my face right now, because this is the group of people that are going to save the world.
Aww, don’t cry. Dry yourself with my fronds. I miss you over at DSL!
This was cool to read. My best wishes to everyone here, in taking their ideas to completion.
I just want to state my most heartfelt gratitude towards Scott and the other funders for their munificent generosity. Thank you *so much* especially for supporting the www.endohazard.org project, which I strongly believe can lead to huge beneficial second and third order effects. Thank you!
That symbol has an interesting angry face appearance. Unsure if intended, and also unsure if beneficial; certainly amusing, and memorable. It seems, complicated? I wouldn't remember how to make it, although I'd certainly recognize it. Of course, I'm not exactly the sort of person who'd be applying this to things.
Regardless, congrats on the grant, and thank you for your work on this!
Thank you! Yes, it's a stylized fusion of BPA atop Dioxin – two molecules that have particularly strong and widespread endocrine disrupting properties. The 'face' is designed to create an instinctive feeling of danger. Even if one isn't aware of its meaning on the side of a barrel or component, it's intended to clearly convey 'bad mojo'.
The spike elements are inspired by the biohazard symbol, which has a similar instinctive effect. I agree that it could be simplified, and may well be in due course as the design is further refined. Thank you for your kind comment, and the insightful feedback! :)
You're over-egging the pudding:
"These elements together form the grotesque Endohazard Wraith, intended to convey an arresting, instinctive sensation of danger and revulsion across all cultures, irrespective of familiarity with the symbol’s meaning."
Yeah, is this intended to be a chemical warning symbol or your new horror video game?
There are a range of symbols in existence already, from the general and simple:
https://echa.europa.eu/regulations/clp/clp-pictograms
To ones for transport and handling of such substances:
https://www.royalchemical.com/blog/hazardous-materials-classes
If you're trying to get a new symbol into general use and widely accepted, row back on the breathless scaremongering and be a lot simpler in design and a lot cooler in "Dear regulatory bodies, we would like to suggest this symbol for new purpose" rather than the florid B-list horror movie photos.
The GHS symbols and SDS sheets provide a strong basis for environmental and health protection. This includes the Serious Health Hazard symbol intended for substances that can cause long-term disease.
However, the uniquely insidious impact of endocrine disrupting chemicals warrants its own endohazard label.
EDCs have transgenerational effects that persist long after the original organism has been exposed, and the alterations may be permanent. This is unlike almost any other hazard class, with the potential exception of hard radiation.
Their symptoms can manifest as ordinary degenerative diseases, or things that people don't necessarily consider diseases such as alterations to gender identity, parental instincts, irritability and mood, mental fog, coldness and sluggishness, impaired sexual desire, etc, yet which may have a major impact upon individuals, societies, and ecologies. This is different from an oncogenic substance, where symptoms are overt and a clear differential diagnosis may be made.
EDCs are absolutely everywhere in modern society, and the problem is getting much worse very quickly. By introducing a specific hazard symbol, a justifiable 'badge of shame' can be designated by regulators to products that are injurious, targeting the worst offenders to begin with, and gradually expanding the scope over time. This way, the power of the market can be brought to bear to curtail their usage by innovating to find less hazardous alternatives.
EDCs are poorly understood or recognized, and indeed often appear benign. Their nature as forever chemicals means that we need to consider future civilizations as well as our own. This is why the symbol needs to be emotionally arresting on an instinctive and universal level – danger, stay away!
The design was more simple in the past, but became necessarily more complex when it came to ensuring it could be used as a stencil. If it needs to be simplified again it can be.
I accept your feedback on the photo, it may be counterproductive. As I have stated most of the effects of EDCs are subtle and insidious and are not easily telegraphed in a physical manner. The photo is a genuine example of EDC exposure at high concentrations, and can represent chronic exposure over a lifetime, or indeed across generations.
It's important that the public understand viscerally how much suffering can be caused by exposure to this class of chemicals.
Thank you for sharing your impressions.
EDIT: I have since changed the editorial image to something more generic.
I'm getting the strong sense that I should be scared but I'm very unclear on what I can actually do about this. Labelling would be helpful, yes, but until then — where do I find out about alternative uses? What should I be using my purchasing power for? Who should I be writing letters to? What should be outlawed? Where can I find out more that I can actually do?
Thank you for your comment, and your enthusiasm about this significant challenge to public health.
There are a few sources I can point to for further info about EDCs:
Dr Shanna Swan's book, Count Down, is an accessible read on this subject: https://www.goodreads.com/en/book/show/50892327-count-down
– https://www.hormone.org/your-health-and-hormones/endocrine-disrupting-chemicals-edcs
– https://diethylstilbestrol.co.uk/studies/
– http://www.germlineexposures.org/
Labelling is a beginning. It's a way to at least make people more aware of the issue, to avoid unnecessary risk of contamination, to enable better regulatory practices, and to provide a basis for free market competition to innovate away from the use of this dangerous (yet often very useful) class of chemicals.
The evidence that BPA is harmful is decidedly mixed. I think there's quite a good chance that BPA is totally harmless.
https://www.nytimes.com/wirecutter/blog/bpa-panic-accomplished-nothing/
Of course, it's possible that BPA is slightly harmful. That would be pretty bad, given how widespread it is.
But merging it in the same category as *dioxins*? Are you crazy? Dioxins are horrible things that I wouldn't touch with a 20-foot pole. BPA is in the canned beans I ate yesterday.
Putting them in the same umbrella seems likely to confuse more than clarify. Your symbol will become the new "known toxic to the State of California" -- a message everyone has learned to ignore, because it is on everything. Which is too bad, because dioxins truly are monsters from the deep.
(You show a photo of a child with chloracne due to dioxin exposure. I note that you don't have anything to show for BPA exposure.)
There are several systematic reviews and metanalyses that illustrate significant effects of BPA on human and non-human life, particularly in early life stages:
https://bmcendocrdisord.biomedcentral.com/articles/10.1186/s12902-018-0310-y
https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC7153198/
https://onlinelibrary.wiley.com/doi/abs/10.1111/gcb.15127
However, I agree that the evidence is much weaker than for dioxins, which everyone can agree are nightmare fuel.
I take your point that Dioxin and BPA are radically different the manifestation of their effects. That said, both present serious health challenges with transgenerational effects. BPA is a cousin of DES, a drug that has caused cancers and gender issues in grandchildren of the original ingesters of it.
https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC3140020/
Both Dioxin and BPA are two sides of the EDC coin – both are forever chemicals that cause damage across generations, whether from Agent Orange or cash register receipts.
BPA is more slow and insidious, easily ignored or mistaken for 'diseases of modern lifestyle', but perhaps just as devastating in the long run due to its ubiquity (~90% of the human population), and the concentration of effects across generations. There is a possibility that EDCs present an existential risk to the future of humanity.
If you have a suggestion for an alternative chemical than BPA, or simply to base it purely on dioxins (as an earlier iteration of the symbol did), I will gladly consider it.
My hope is for the symbol to be primary used to denote the most hazardous kinds of EDCs, though which kind qualify is something I would leave to regulators.
I accept that there may be a risk of public confusion about the intensity of threat, or of the symbol becoming undermined over time by 'crying wolf'. I am considering some kind of gradation or 'traffic light' system to be applied with it. This is why the standardization process is so crucial – to hammer out these kinds of details, and to carefully define the scope of use.
Thank you for your feedback, which I appreciate very much.
Thanks for your response. You link to 3 meta-analyses, 2 of which are about humans. For the human ones, they summarize 16 and 10 studies respectively (presumably with some overlap) and find correlations between BPA exposure and (1) Type 2 diabetes (in the first meta-analysis), (2) obesity (in the second).
Almost all studies included in these meta-analyses are cross-sectional. None of them are randomized controlled studies. What this means is that we know there is a *correlation* between BPA and obesity/diabetes, but this is a far cry from showing causation.
Of course, RCTs are really hard to conduct in such a setting. I'm not really blaming these researchers. What I'm saying instead is that we just don't know. It's *possible* that BPA contributes to obesity and/or diabetes. It's just as possible that it doesn't.
Many, many other things correlate with obesity. Basically, controlling for SES (socioeconomic class) is always imperfectly done, and so SES ends up confounding all correlational health studies. In other words, poor people have higher obesity and higher BPA exposure. You'd see the same result if you studied whether (say) breastfeeding prevents obesity later in life: the studies will all point to "yes", because it's all SES confounded, even after attempts at controlling for SES.
Again, maybe BPA causes obesity. I can't rule it out. I just don't think we know that right now. Perhaps the precautionary principle means we should avoid BPA (but beware its understudied substitutes like BPS and BPF; for all we know, they might be worse). But they are not the same as dioxins.
(The one mouse study you link is perhaps more concerning, but I can't immediately tell how high the dose given to the mice was; I'd guess it's much higher than the typical human exposure levels for BPA.)
It's also worth noting that BPA does not bioaccumulate. You say there's a concentration effect across generations, but I don't know what you mean by that. It's true that we use it more now, but I wouldn't call increased usage a concentration effect. (Dioxins do bioaccumulate.)
Thank you, I appreciate your response very much.
I note that Bioaccumulation of BPA (in human beings, rather than the environment) may be underreported, as some research suggests that it can be excreted in sweat but not detected in blood or urine.
https://www.hindawi.com/journals/jeph/2012/185731/
By concentration of effects across generations I refer to transgenerational and multigenerational epimutations stemming from EDCs. Transgenerational effects (due to prenatal exposure) from BPA are strongly reported in literature:
https://www.nature.com/articles/laban.796
https://pubmed.ncbi.nlm.nih.gov/30362448/
https://pubmed.ncbi.nlm.nih.gov/31272786/
https://www.biorxiv.org/content/10.1101/2020.11.20.391672v1
https://academic.oup.com/endo/article/160/8/1854/5513493
https://academic.oup.com/humrep/article/35/8/1740/5869374
The real nightmarish issue is multigenerational effects which compound over generations. This has been documented in human beings, with the grandchildren of women who ingested DES.
https://pubmed.ncbi.nlm.nih.gov/30758926/
Such multigenerational damage may lead to an eventual loss of viability for affected organisms if that damage cannot be repaired.
https://www.ncbi.nlm.nih.gov/labs/pmc/articles/PMC4662893/
https://www.frontiersin.org/articles/10.3389/fmars.2020.00471/full
https://www.sciencedirect.com/science/article/pii/S001393512101358X
This is why I consider EDCs to be a potential existential risk. The long term danger of this class of chemicals to public health (of human and non-human life) may be orders of magnitude worse than Tetraethyl lead.
I certainly take your point that direct prima facie evidence for BPA as a hazard remains far less strong than dioxins. Perhaps PFAS might be a more immediate example.
Hmm. I didn't realize you were referring to an epigenetic pathway.
I'm in principle skeptical of meaningful effects happening via epigenetic pathways. I was a bit surprised to see your links to so many papers finding such effects in mouse models; I will keep an eye out for this, and maybe investigate it later when I have time. It would indeed be concerning if damage from endocrine disruptors lasted for multiple generations.
I'll still note that the damage from endocrine disruptors in these mouse models was *fading* over the generations, rather than becoming worse. Of course, this is in the context of the researchers not exposing the subsequent generations to the endocrine disruptors, so I see your point about how the damage may get worse across generations with constant background exposure. I guess all I have to say is that I'm vaguely suspicious of these mouse studies and think they must have done something wrong, but this is not a very principled stance so I acknowledge I could be wrong here.
Small side note: I'm ignoring the studies done in zebrafish or other non-mouse, non-human models, because zebrafish are quite far from humans and rodents are a much better model.
Another small side note: the study in humans who ingested DES is not very convincing. This is because DES was not given randomly: so far as I understand, it was given to at-risk pregnant women. Their daughters would therefore inherit whatever genes caused their mothers to have at-risk pregnancies, and therefore would be more likely to have at-risk pregnancies, even without DES exposure. This is ordinary genetics combined with selection bias, and not epigenetic effects from DES. Also, preterm birth straightforwardly affects the next generation (no epigenetics needed), so if DES exposure in utero caused the DES daughters to have preterm births themselves, this would affect the DES grandchildren without any epigenetics involved. So there are two possible explanations here that don't have anything to do with epigenetic damage.
Still, you've linked to a lot of mouse studies that aren't subject to these objections. I'll have to think about them.
Yeah, as someone who understands a fair amount of the organic chemistry, I find the idea that BPA is some monster poison laughable, particularly given (1) it's not water soluble, (2) even in the optimal laboratory case, it's a very poor substitute for estrogen, and (3) it's been used so widely and for so long that the idea that it could have massive health effects which have remained unnoticed for 50 years is dubious. If massive exposure for decades *does* do any actual harm, it would be barely measureable, and the relative risk compared to, say, the aflatoxins in your peanut butter or the radon you breathe every time you go down into the basement close to zero.
Speaking of which, how concerned should I be about aflatoxins in my peanut butter?
"Other times a reviewer was concerned that if you were successful, your work might be used by terrorists / dictators / AI capabilities researchers / Republicans"
Right. "Republicans." Because if we've learned anything over the past two years, it's that only one side of the political spectrum is capable of doing harm.
Seeing another comment like this above makes me feel like maybe there are some new readers on here who just aren't attuned to Scott's sense of humor? Rest assured, etc.
FWIW, I've been a huge admirer of Scott's writing in general and sense of humor in particular, for many years, and am not a Republican (though I was, kinda, an AI capabilities researcher, just a bit!), and yet I still cringed. It simply isn't funny enough to justify the potential misreading or sense of alienation enough people might have.
ETA: for the sake of new readers, though, rest assured etc. indeed.
If a joke doesn't offend someone it probably wasn't funny to begin with.
I guess YMMV with this. Being familiar with Scott's humour and the context, I would have been unoffended with whatever word was in the place of 'Republicans'.
Mostly because the joke is set up by 'AI capabilities researchers' - after that Scott is free to drop in whatever he feels like. Republicans, Democrats, Georgists, Hospice nurses.......
But we all know that Scott really does worry about what AI capability researchers might do. So it doesn't so really set up a joke. Not that it can't end in a joke, but I don't doubt the AI bit at least is meant seriously.
If it had ended at "AI researchers", I'd have laughed. Going on for a political opponent made me wince, since that doesn't help with partisanship. Then again, I have seen people absolutely convinced Republicans are indeed evil moustache-twirling, widow-evicting, orphans' tears-drinking villains and not their next door neighbours, so it's dry humour I guess.
Yes, it's dry humour, but surely Scott is just gently mocking his reviewer?
To overanalyze a split-second writing decision that I'm not sure I have much more insight into than the rest of you:
Terrorists, dictators, and AI capabilities researchers were actual things that various reviewers were concerned about. When I put them up there it seemed funny in a "murder, arson, and jaywalking" way, so I decided to go for broke and add "Republicans" to give it lots of different categories that people worry about for different reasons and at widely varying levels of certainty about how bad they are.
Speaking of "murder, arson, and jaywalking", I got that vibe from the endocrine disruptors link, where they group together dioxins and BPA into one category.
Dioxins are horrible substances that are both acutely toxic and a class 1 carcinogen, and they bioaccumulate to boot.
BPA is none of those things, and there is controversy over whether it is (a) totally harmless, or (b) possibly an endocrine disruptor causing mild harmful effects. But even in case (b), everyone agrees the common substitutes for it can easily be worse (they are just understudied). In other words, plastics that are "BPA free" are just as likely to be endocrine disruptors.
Well, if you're so sensitive about slights to your ingroup, maybe it would help to look across the board, and see SA poking Marxists as well:
https://slatestarcodex.com/2014/09/13/book-review-singer-on-marx/
15 minutes of reading NA's Reddit history has convinced me that he's probably not actually interested in EA and that you've been taken for a ride.
Two points.
First, I’m very much in favour of doing good things in efficient and effective ways. In my book that makes me pro-EA.
Second, even if you don’t believe that, you can rest assured that I do believe in making people who give me money feel glad they gave me money. I’m a professional, and I intend to deliver results.
You wangled a grant out of this, so you are indeed a professional and successful at what you do, congrats! 😂
One point and one point only. You're a professional in a field that's known for professional ethics which foster only the highest levels of objectivity, rationality, altruism, integrity, honesty, efficiency and transparency. Being able to use something like Reddit to reveal preferences that appear to be, at best, totally unrelated to any of the grant objectives is not an undue risk in light of the high degree of professionalism you claim to possess.
Ash, what are your EA policy proposals, and who do you intend to lobby about them? Governments spend money all the time, so it can't just be about getting them to spend more. Do you have some ideas for making spending more efficiently targeted, or is it like an awareness raising thing? I've worked for a couple of Australian politicians and I'm interested to know what your plans are.
This is a question I'm interested in hearing the answer to as well. Are you in contact with the various EA groups in Australia, Ash? Melbourne has a reasonably active one and I assume that's also true of the other big cities - on that note, I'm curious the extent to which you're working for Scott or whether he's employed you on behalf of Australian groups with more local knowledge of what to lobby for?
Can you explain this?
The comment history is pretty large but shows no clear evidence of involvement in or advocacy for anything that would meet a definition for EA that basically every normal person wouldn't also meet. You would expect from a person who intends to undertake completely unspecified and unsupervised work in support of a policy framework at least some evidence of prior engagement with it.
Conversely I think the Reddit history strongly suggests that you've hired a pretty hardcore culture warrior who's otherwise a basically a standard political operative. The reason I think you've been taken for a ride is that political operations are extremely well known for their transparency and ease of progress measurement, while those who work in the area are universally upheld by all societies as paradigms of sincerity and honesty with a proven track record of never pursuing ulterior agendas most especially not in federal election years.
I would suggest that with a Federal election due in the next four months, the odds that a good political operative is focussed on the objectives of a foreign grant are not high. I would suggest that the odds of your goals being misaligned with his over the next four and twelve months are nearly 100% and I would suggest that structuring incentives to align those interests sufficiently to get value from your grant spend is probably going to be impossible even if none of the resources you invest are directly reinvested into things that are contrary to your goals.
What this suggests is that for the grant that seems to have the least clear and measurable outcomes and which is in a field where outcomes are notoriously unclear, hard to measure and hard to attribute you'll be most reliant on setting extremely clear, attributable and measurable objectives yourself.
My suggestion is that you set some objectives around land value taxation uptake. There is already very high degrees of support for it at multiple levels of Australian government in most parties and most of the public service, including limited adoption in one jurisdiction. Everyone agrees that more of the tax base being more efficient is good and that LVT is about as efficient as a broad tax can be. Finally it aligns with NA's interests and beliefs so it helps to minimise risks from misaligned incentives. The only disadvantage is that it is electorally unpopular which aligns poorly with NA's skills (elite vs popular advocacy). It is extremely difficult to imagine any other area where you will be able to find clear and even vaguely measurable goals within Australia's policy landscape that you desire for NA to achieve.
"The only disadvantage is that it is electorally unpopular". Unfortunately, this can be a huge and crippling disadvantage.
I know people have been shilling the land tax reform thing for a while but I can't remember what all the details are or how it's played out. Didn't they do it in Canberra or something like that?
Yeah, ACT. The only place where the wonk density is high enough to get it on the agenda easily. I know for an absolute fact that the government in at least three other states would love to do tax reform with LVT but no one's sure how to put it on the agenda.
Edit: Which is to say that it hasn't played out anywhere yet.
Which states? Can I ask that or is it Secret Insider Information?
I'm inclined to agree with you that the lobbyist grant seems dodgy, just based on the general vibe of the thing. It doesn't take $90,000 to learn about politics.
Would betray some confidences.
I mean, Victoria has a Land Tax but it's a weird thing with 'progressive' tax rates designed more to punish owning lots of land than incentivise good use of land. IME most state-level taxes are implemented in awful ways, and I'd love to see us get what was promised with GST, which is for taxation to shift almost entirely to the federal level with the money being sent back to the states but without all the weird distortions to economic incentives.
Victoria's quote unquote land tax is for vacant property. When they increased the rights of renters a few years ago they ended up having to change the land tax conditions because there was a wave of landlords who disabled their own property so it would qualify as uninhabitable so they'd be exempt the tax.
I think state laws are such a mess mostly because state politics is less competitive. It's the political B Teams of parties and some states are very much stitched up by certain factions and parties. Add to that that states had taxation power before the commonwealth so there's more legacy junk around and I think the picture is complete. The good news is that it makes it easier to fix.
"D, $5,000, to help interview for CS professor positions."
Huh? Interviewing for CS professor positions was taxing in all sorts of ways, but it wasn't expensive. In fact, universities were super accommodating, saving me the cost of a personal flight to Toronto in between interviews.
"Everyone with experience in movement-building says that getting your members into top positions at top colleges is important, and this is a surprisingly cheap opportunity to make that happen."
Huh. I'm guess I'm one of those members. Do we have any coordination mechanisms?
Great work Scott! Lots of love to all involved. I can't wait for the follow ups!
This is the most exciting and innovative thing I have encountered in months. It's gone way, way beyond my expectations. Kudos once again, Scott.
Decent list. Compared to some other EA-adjacent funding, this was relatively light on "$X to my friend for personal growth, because they are awesome and after personal growth they will affect the world for the better". That's a good thing. But it still had a bit of this category/pattern, which is a bad thing.
For example, getting money to apply for faculty positions in CS is pretty useless -- the interviewing universities will refund all the plane tickets anyway.
Yeah, the "please fund my lifestyle for a year while I sit around and decide if I want to stay on the course I picked" was a bit tooth-grinding for me: get a job, ya bum, like the rest of us if you're not going to be in training/education!
But hey, I was reared working class/peasant, where it was up to you to get there on time, pay for transport, and forego a day's pay if you had to take the day off for interview, and not "of course the people inviting you for interview will reimburse the cost of transport and pay to put you up overnight if necessary" class of job. How the other half lives! 😁
I support further experiments in connecting EA to political structures. For instance, there are many serious ongoing openings for engagement in the U.S. that should be appealing to the EA community, some of these likely among the very highest-leverage near-term opportunities to improve lives across the world at scale. Just some of these include: better vaccine distribution; the opportunity to provide aid to poorer countries at a scale of hundreds of billions of dollars to help them through the pandemic; augmenting sanctions policies so fewer people starve to death; ending various foreign escapades (the U.S. is complicit in a blockade of Yemen that has the potential to cause hundreds of thousands of surplus deaths) and more. And it can be striking how much difference can be made with relatively modest engagement, especially in light of the magnitudes of the potential outputs.
Numeracy, sorting out priorities based on even a vague sense of utilitarian impact, is an underdeveloped skill in DC. And EA community would do well to learn more about politics, legislation, rule-making, etc to come to understand highest-leverage mechanisms for potential influence and how to manipulate them. (I say this as what I suppose one might call an American analog to the Australian grantee.)
Yes, absolutely! It’s politics and nothing is guaranteed, but I consistently find it shocking how much impact just basic engagement with the political process can have.
Have you by chance seen any organizing activity in Australia around the IMF/Special Drawing Rights? That's the most profound opportunity I'm working on here. Conceptually could get on the order of $1 trillion or so out to poorer countries to help contend with pandemic fallout (direct health and economic), with around $200B or so already released.
This is a fantastic initiative and I really hope it becomes a template for others.
Thanks to Scott and all the generous donors on behalf of me, Delia Grace, and the pigs, farmers and pork-lovers in Uganda! Very exciting to be part of such an eclectic and talented group
Hey Scott,
I submitted AI related grant, and I can't be sure if it was rejected, LTFF did not start working on it, or my spam filter ate their mails.
It would be great if you could send status updates about such grants.
Thanks in advance!
I've checked and you are in LTFF's system as a grant that they know they need to review.
Thanks :)
This was an astonishingly fast submission, review and grant allocation process, compared to how this is traditionally done. Impressive, not least taking into account the large number and diversity of the projects seeking a grant. Based only on reading this blog post, it seems that the review process succeeded in selecting exiting projects. Made me think if there are more detailed tips or tricks concerning how Scott organised and implemented the decision-making process, that could be useful for other public or private organisations (or wealthy individuals) funding research. Consider this a suggestion for a possible future blog post.
"Why would somebody working on biochemistry want to talk to someone working on political activism..." Well, why would anyone talk about those topics on the same blog? It is all part of the codex, while all fields of research have their specific tools, they also useful tools of the general craft. As a PhD-student, I really like the events where you get to talk to PhD-students from other fields, because you have a good opportunity to learn about the field of others, and more importantly, learn how your field is seen by others and try to give it a more positive/reasonable image to the person talking to you.
This is awesome
Incredible work, huge praise to the applicants and the work of Scott and the reviewers picking out these exciting projects. Very interested to hear about how the various projects progress.
Great work, and thanks for the effort (except suing chicken factories, come on)!
It was an interesting to see the balance between suing chicken factories and "dedicated mobile slaughterhouses".
Well, I imagine you can't really sue the mobile slaughterhouses out of existence *before* they come into existence. Give the Chicken Farms Suing people time to get round to it!
More seriously, it does raise the moral question: is it better to be a dead pig or a live but sick pig? If the live pigs are spreading disease, is the benefit to them of their lives more than the benefit to the healthy pigs of their death? If you're going to argue that raising and killing chickens for food is wrong, this is another question you have to address. Will the slaughter houses sell the pig carcasses for food and if so, isn't that immoral on animal rights grounds? If they have to kill infected pigs, shouldn't they destroy the carcasses rather than encourage meat eating and so make it profitable for farmers to raise pigs for food?
(Scott is the one who used "moral" in a couple of these awards descriptions, so if you're going to argue morality as a ground, you have to be prepared for those counter-arguing your proposition is immoral).
This whole post makes me almost weep with happiness!
Congratulations to winners! I was asking the PI in my new lab if I can apply as well, but he said this wouldn't be the right time. Maybe next year.
"Other times a reviewer was concerned that if you were successful, your work might be used by terrorists / dictators / AI capabilities researchers / Republicans and cause damage in ways you couldn't foresee. " -- I cannot believe that you actually wrote this. Very disappointed. (Not in the grants, in what how you chose to write the intro.)
Scott's making fun of people who think Republicans and AI capabilities researchers are as bad as dictators and terrorists. It's a classic Scott joke, I laughed.
The AI capabilities isn't a joke on his part. If that's ambiguous then so is the Republican part.
Is there any way to support the next ACX grant with private donations? I am sure many of your readers would love to participate in such a high-impact and diverse portfolio of projects
On a different note, I would like to invite you to visit by blog. I am Dutch and writing in English. It would be great to get feedback, both regarding my English and the general writing style of my blog. I would greatly appreciate it.
henkb.substack.com
I like your writing style - very easy on the brain. Your English is also excellent. I only noticed one idiomatic oddity which was the phrase "3 o clock at night" - most native English speakers would say "3 o clock in the morning. So... your English is even better than the average Dutch person :)
I would prefer longer, more in-depth and more striking articles but each to their own.
I wish you well in your writing, drawing, dancing, playing and singing!
As an Australian, happy to see the grant to "NA", especially after browsing AshLael's comments history.
> they're smart enough to know about XYZ Grants which is better suited for them, which means they're mostly banking on XYZ funding and using you as a backup
Um, doesn't this confuse "smart enough" with "experienced with applying for grants"?
Sure, if it's considered a norm to apply for N grants in parallel, this means an experienced grant reviewer would encounter N times more applications like that, and it's reasonable to assume this as the default, and to try to adjust for this kind of selection bias.
But as a person who had applied for the first time, this sounds... surprising, to say the least.
Intuitively, it sounds intrusive and rude to just spam every grant program out there. Is this really expected and OK?
Would really appreciate comments from an experienced grant reviewer.
Still, thanks a lot for highlighting this point! Will keep it in mind.
(I've never previously considered grants as a realistic source of funding -- coming from the industry background, and given anecdotal evidence about the amount of bureacracy from friends who successfully received ones. Decided to apply only because the initial post was so much more welcoming and really different from what I've heard about grants previously)
On a second thought, if what this really means is "since the applications are going to have a second chance at ACXG+ anyway, let's focus on the ones that are unlikely get funded via ACXG+" then it's totally reasonable, of course.
I suppose there's two points there:
(1) If you're looking for funding, don't put all your eggs in one basket. Apply everywhere, you'll probably get some positive reply, rather than just applying to one source.
(2) This supposes that people will know that XYZ Funding exists. They may not do! It wasn't until I started working in one area that I became aware of (a) exactly how much the organisation covered and (b) how many people had no idea they could apply for grants from that same organisation.
Someone smart but inexperienced or just not plugged into the "grants and funding application network" may not be aware at all that they could apply for and be awarded XYZ Funding.
Point 2 is really important. I think it'd be a great service (though possibly also a lot of work) if Scott or associates were to go through every candidate who got that comment and still ahs a funding gap and let them know that the reviewers thought they should be applying for XYZ Funding.
It wouldn't be a lot of work if it's boilerplate.
Now this is the timeline that I want to live in
It seems like the system you used to publish the rest of book reviews worked pretty well. Perhaps something similar would work with the grant applications.
Amazing, excited to see what will come out of all these projects. Great Job Scott, truly inspiring
I am very surprised that experiment.com was not mentioned. I am constantly frustrated that there are not more medical science projects there to support.
And the ones that I _do_ want to support get funded almost instantly. Come on, researchers, take my money!
I'm very excited to see how these turn out, especially the one that you admit is very likely to end up being a disaster if not overtly evil lol
"Alice Evans, $60,000, for sabbatical and travel to fund her research and associated book on "the Great Gender Divergence", ie why some countries developed gender equality norms while others didn't."
God, no.
Someone on Reddit looked at some of her other work. https://www.reddit.com/r/CultureWarRoundup/comments/rpgdcg/december_27_2021_weekly_offtopic_and_loweffort_cw/hqeq5pg/
The chickens one is also pretty bad (and I'm not going to claim originality, this was pointed out on Reddit too, I didn't even notice the clause): to help kickstart their project of suing factory farms that violate animal cruelty laws *or otherwise expose themselves to legal action.* Translation: If they don't violate animal cruelty laws, we'll use the $72000 to find something else to get them on.
That some of the funded projects would be trash like this is hardly surprising. Always gonna be some duds in a program like this. Luckily blowing 60k on 1 really dumb thing is actually not even close to a worst case scenario. It also doesn't say who got their money from Scott vs others. Maybe this funding came from someone who agrees with the other stuff.
Basically this. There were a few in the list that looked like wastes of money to me, but out of this many projects it really wasn't that many and they tended to be on the smaller side of funds spent.
It does tar the whole things a bit though. It's not like either were particularly hard to weed out.
I think the chickens one is bad, too, because it's basically funding nuisance lawsuits. The group think factory farming is bad, eating chickens for food is bad, and want to get commercial and private 'chickens for food' ended. Going after factory farms is low-hanging fruit; I'm all for prosecuting abusive industry, but "or otherwise" means "if we can get them on having 0.99 instead of 1.0 hectare per 100 chickens, we'll sue them and with any luck force them to close down". No.
I don't know how a lawsuit like that would even succeed. Chickens don't have any constitutional rights, and if they did, only chickens could sue for their violation, since you can't sue on some other...entity's...behalf unless you are its parent or something. So what would be the basis? Violation of assorted local regulations on cage size, et cetera, one assumes -- but it's unlikely any such regulations allow a private lawsuit as a method of enforcement (unlike the novel and much maligned recent Texas anti-abortion law). Normally the only plaintiff with a cause of action for the violation of regulations set by the state is the state itself (meaning the state can sue, but not any individual). I guess you could try animal cruelty laws, but, again, the usual cause of action lies with the state. The only thing that comes to mind is suing on the basis of your plaintiff's mental distress -- OMG! I ate a chicken that had been cruelly raised! I feel so guilty now I'm going to hang myself -- or a much more difficult to prove claim of some actual physical damages (e.g. the chickens from Factory Farm X turned out to contain high levels of Compound Y because they were under constant stress and Y caused your asthma).
If I had to take a wild guess, I'd say nuisance lawsuits -- with no real hope of success -- is the actual game plan. You sue for any wild reason you can get a lawyer to commit to paper without fearing for his license -- and hope to get into the news, embarass the defendant, get a 6-figure settlement which you use to fund the next round, et cetera.
I suspect Scott's reasoning below is correct (that actual full on abuses are sufficiently common to avoid needing to sue on technicalities).
However, even if that's wrong, if you think factory farming is basically immoral (I do), then making the industry act overly cautiously to avoid lawsuits seems great. If Tyson Chickens ends up giving all its chickens 1.2 hectares because it's worried it might slip under 1.0 and get sued, that seems like good news. If Tyson Chickens ends up going bankrupt, all the better.
I'm against nuisance lawsuits in medicine etc because I think they have negative externalities and reduce the quality and quantity of a good with positive externalities and high social value. I'm at least queasy about Peter Thiel suing Gawker into non-existence because it impinges on a fundamental political freedom (freedom of the press). None of that applies to factory farming.
If the critique of this charity is that they might shut down factory farming for $72,000, that is probably the best thing Scott will ever do.
I think enough factory farms genuinely violate animal cruelty laws (and whatever your common sense notion of animal cruelty is) to burn through $72,000 (or $72 million) without getting to edge cases.
Dr. Evans has some beliefs I find annoying, but her core work is about things of the level of forced marriage or women being banned from owning property and so on. I feel like people on every side of first-world debates about social justice can agree those are bad and be interested in changing them, and I'm told Dr. Evans' work in those fields is rigorous and important. I think if I were to think less of her based on some of her political comments I find annoying, and refuse to back someone's rigorous / important / potentially beneficial work on those grounds, that would be akin to cancelling / witch hunting / whatever you want to call the thing where you blacklist someone because of their politics.
I agree with Deiseach (below) about the chicken people. If they were motivated to keep factory farms to the spirit, or even letter of the law I'd say fine, why not. But they aren't - they expressly want to shut down law-abiding businesses for ideological reasons. However your defence of them is fair, and pragmatically they will merely harass the worst actors so their influence will (in my view) be mildly positive.
I'm more impressed with your ability to put aside the annoyingness of Dr Evans. I couldn't do it, despite wishing that I had the courage and charitableness. Bravo!
I think your sociological reasoning is in error. The most likely targets for harassment are actually the *best* businesses, those most concerned for their public image, and those which are more profitable through efficient and effective operations -- but who have made a mistake here and there (and what big operation has not?) You don't sue people who are barely making it, financially, because they won't be able to pay out if you win. You don't sue people who are already shady operators, because ipso facto they don't care as much for public opinion. To have the most impact (to your accounts receivable), you sue the people who have a high public profile, who are big and wealthy and successful, and who appear in general to be nice clean folks -- and who prize that reputation. Those are exactly the kind of people who will give you $1.5 million to just go away as soon as you serve them.
I mean, people sue brain surgeons and obstetricians at famous clinics, not used-car salesmen from the wrong side of the tracks who chisel them on a beater. As Willie Sutton said about why he robs banks, because that's where the money is.
Yes, I agree with you about the targets. However I don't think the motivation is money and the likelihood of payouts - Tyson foods gets sued fairly regularly and pays out a few million here and there for various breaches of regulations. But they're not paying out to the activists who want them completely shut down.
I think the objective - at least the proximal one - is simply to get negative images (of suffering/malformed/generally oppressed chickens) into the public eye as often and as viscerally as possible. To tie those things into "industrial farming" would obviously be a bonus. Because on the slow march to a virtuously vegan world there will probably have to be some other destinations along the way including generally higher welfare standards, free-range and organic.
My objection to the support of this project is the dishonesty involved, and I admit that I may just be naive on this point. But the project isn't to genuinely keep the industry sticking to the regulations - or even to change the regulations themselves. It's to harass people because of an ideology which, if admitted more honestly, would result in them receiving a great deal less support.
>I'm more impressed with your ability to put aside the annoyingness of Dr Evans. I couldn't do it, despite wishing that I had the courage and charitableness. Bravo!
It's admirable in one sense, though I think extremely naive to think that these two areas are two totally independent things and that good insights can be expected to come from somebody so absolutely wrong about a related area.
I'd expect her beliefs about structural racism to be associated with particular beliefs about gender equality. It's not as if she's the person working on autoimmune disorders or neutrino physics--it's not "cancellation" because her beliefs are relevant to what you wanted to fund her for. Everyone here can agree that forced marriages are bad, but she's not just going to be saying "these are forced marriages. Stop them." She's going to have recommendations and analysis and I don't see someone into social justice as doing well with those.
(And if you say "someone can have bad beliefs about racism but be perfectly sane when it comes to gender relations"--that's logically possible, but my priors are strongly against that.)
I have never read that book that she reviewed (nor will I ever want to - based off its scoldy title) but its actual contents seems a lot more anodyne than something like White Fragility: https://en.wikipedia.org/wiki/Why_I'm_No_Longer_Talking_to_White_People_About_Race
Her actual work does not seem to be 'structural racism but gender', especially the stuff on South Asian history. I ironically came across her via Razib Khan of all people: https://www.draliceevans.com/blog
>I feel like people on every side of first-world debates about social justice can agree those are bad and be interested in changing them, and I'm told Dr. Evans' work in those fields is rigorous and important.
Okay? She's an academic employed by two elite universities. She doesn't need funding.
SURELY there are people doing rigorous and important work like this who actually need funding and don't have completely hateful and wrong-headed views like Evans, so?
And I sincerely doubt that her work can possibly even be rigorous when her view of the world in this other, related domain (race) is so completely false. Do you really think she's going to produce some stunning insights on "gender equality" when she is totally, 180 degrees wrong (and extremely obnoxious) about other social issues?
I'm all for funding work on reducing gender equality, but I struggle to believe that this was even the tenth best option available for doing so, even if you had to ask to get people to go looking for those opportunities for you.
>I think if I were to think less of her based on some of her political comments I find annoying, and refuse to back someone's rigorous / important / potentially beneficial work on those grounds, that would be akin to cancelling / witch hunting / whatever you want to call the thing where you blacklist someone because of their politics.
It certainly doesn't help that you say this after essentially saying "we don't want to give money to those evil republicans haha just a joke but not really".
Good for you, Sir. I am such a fan.
If Scott couldn't fund your project, but it might be a good fit for a Focused Research Organization, please get in touch with us at https://convergentresearch.org/.
so the grant for "D" is really a blinded grant for a team studying psychological torture and an addiction to pain, right
Thanks for doing this in such short a time! This is really amazing and I am very excited for many of these projects!
Hi, I'm would like to make up some of the $150,000 (I can't give the entire thing but I am willing to give what I can) that you say Kartik Akileswaran and Jonathan Mazumdar are looking for but you didn't list a way to get in tough with them. I'm hesitant to use a contact I found on Google for fear of being scammed by the wrong Kartik Akileswaran or Jonathan Mazumdar. I also can't seam to find, Dr. Alexander's e-mail listed on ACX or Substack anywhere. I'm probably just missing it.
I was raised to be quite paranoid and I'm trying to figure out how to verifiability give money to the right people. Are there any suggestions on how I should do this? If Dr. Alexander can see the e-mail associated with this comment then feel free to e-mail me with contact information for the pair. If it is known that Scott Alexander checks his Twitter DMs then I can probably contact him through a friend of mine.
His email is in the blog post above, under "How to Get Your Money" -- it is scott[at]slatestarcodex[dot]com
Thank you so much. I suspected I just was missing it somewhere. I appreciate it.
Hi Benjamin, wow, what a great surprise! Thanks very much for your interest and generosity.
You can reach me at kartik.akileswaran@post.harvard.edu, and Jonathan at jonathan.mazumdar@gmail.com. (We're in the process of sorting out our organization emails.)
Looking forward to connecting on this. Happy new year!
Cheers,
Kartik
Okay, here's where I need some clarification. Take this:
"Still other times, my grant reviewers tied themselves up in knots with 4D chess logic like "if they're smart enough to attempt this project, they're smart enough to know about XYZ Grants which is better suited for them, which means they're mostly banking on XYZ funding and using you as a backup, but if XYZ doesn't fund these people then that's strong evidence that they shouldn't be funded, so even though everything about them looks amazing, please reject them."
Then take this:
"Alice Evans, $60,000, for sabbatical and travel to fund her research and associated book on "the Great Gender Divergence", ie why some countries developed gender equality norms while others didn't. A large body of research shows that gender equality, aside from its moral benefits, is also deeply important for economic development. Dr. Evans is an expert on the interaction of gender, history, and economics, whose work has been cited on BBC, Al Jazeera, and Sky News. She blogs here and podcasts here."
That is a project that should be funded either by her publisher for the forthcoming book on it, or her university department. Are Princeton University Press suddenly so penniless they no longer pay out advances? https://press.princeton.edu/news/the-great-gender-divergence
Anyway, there a couple of things I would like to know further:
(1) How do you define "gender equality"?
(2) There are plenty who complain that we don't have gender equality even now, today, in the developed Western world. Are we going to see talk of the pay gap, or will all this be subsumed into "women can be prime ministers, too!"
(3) I know it's a blurb, but there seems to be already a structure in place for why this is so:
"Inspired by research on The Great Divergence, Evans asks why, as societies around the world have become more gender-equal, some regions have become more equal than others. Something radical happened over the twentieth century: women entered the workforce and became political leaders across the world. This had never happened before, not in the entirety of human history. What changed? And why is progress so uneven?
The Great Gender Divergence will explain the causes of Europe’s “precocious equality,” how East Asia and Latin America caught up, why gender equity in the Middle East, South Asia, and Sub-Saharan Africa has lagged behind, and why South-East Asia was always ahead. Evans offers a comparative history of how societies come to support gender equality and why this varies around the world, telling a story about geography, economic growth, strong states, and militant activism."
So it would seem that Dr. Evans has already answered the question posed and it is about "geography, economic growth, strong states, and militant activism".
Myself, I would put the difference in one word: "education". Can I now hold my hand out for a share of $60,000 for a pleasant junket jetting around internationally while I 'research' a topic I have already made my mind up on?
However, let me quote one grant that I do think is a good and useful idea:
"Michael Todhunter, $40,000, to continue work on automating testing cell culture media."
Yep, badly needed, and I wish him good luck on this.
Good job raising all that money. Congratulations to the donors and grant winners.
> illegal fuel enrichment produces neutrinos which could theoretically be detected from thousands of miles away
To nitpick: I do not understand why fuel enrichment (as in U-235 enrichment, e.g. in a centrifuge, which involves neither nuclear reactions nor the weak interaction in particular) would produce neutrinos. Are you sure you are not talking about monitoring the anti-neutrinos from reactors used to breed Plutonium?
Fuel enrichment obviously does not create neutrinos itself, but if you had a super-duper sensitive neutrino telescope, I imagine you could probably tell that the concentration of U-235 was going up in some location from the higher level of spontaneous fission reactions there. Although the level of sensitivity required is incredible. Normally neutrino detectors the size of swimming pools are needed to detect the enormous blast of neutrinos put out by a supernova, or by the Sun's fusion at its core. (Which also points to the incredible level of noise-discrimination you'd need.)
What sort of follow-ups can we expect on the status of these projects, if any?
No firm plans, but I'd like to ask them all how they're doing a year from now and post their answers.
I look forward to it!
"Still other times, my grant reviewers tied themselves up in knots with 4D chess logic like "if they're smart enough to attempt this project, they're smart enough to know about XYZ Grants which is better suited for them, which means they're mostly banking on XYZ funding and using you as a backup, but if XYZ doesn't fund these people then that's strong evidence that they shouldn't be funded, so even though everything about them looks amazing, please reject them."
Completely understandable, but there are many reasons that a given philanthropy might not donate to a seemingly obvious recipient. Only one of the reasons (and, I suspect, not the most common one) is that they made a full, fair, and objective investigation into the recipient, and decided to reject it for reasons that everyone else would agree with.
If Foundation A gives a lot of money to Issue B, and if a proposal relates to Issue B but got turned down by Foundation A, there could be lots of reasons for that:
1. Foundation A took a solid look at the proposal, and decided for objective reasons that it wasn't worth funding compared to everything else related to Issue B. (Fair enough).
2. It's December, and Foundation A didn't look at the proposal at all yet, because their internal process requires the proposal to have been submitted by June 1 for a decision to be made by the Board on Nov. 1, and there's now a new funding cycle that won't make decisions until March of next year.
3. Foundation A already exhausted its internal budget for Issue B for now.
4. Foundation A hasn't quite exhausted its internal budget for Issue B, but staff are saving the rest while they argue about an expansion into Issue C.
5. Foundation A has a new President with a different vision for the organization, and unbeknownst to anyone has decided not to invest in Issue B going forward.
6. Foundation A has decided to revisit its entire strategy on Issue B, and the strategy planning and discussion will take a year before they make any more grants.
7. Foundation A's staff had already put a lot of time and internal capital into arguing for a particular approach to Issue B, and this new proposal takes a slightly different approach, and they are embarrassed to change direction so quickly.
8. Foundation A's staff have been involved in an internal squabble for reasons unrelated to Issue B, but which make it impossible for now to get the necessary agreement.
9. Foundation A hasn't funded it *yet* . . . but might soon!
10. Foundation A is itself playing a strategic game where they are trying to anticipate what other funders might do, and that is delaying everything.
11. Prominent grantees of Foundation A are skeptical or insecure about the newcomer proposal, and Foundation A wants to preserve good relationships with them.
12. The decisionmakers at Foundation A were somehow rubbed the wrong way by the proposal and/or accompanying PowerPoint.
13. The decisionmakers at Foundation A just don't have a good feeling about the grant, for reasons that they can't articulate.
None of these rationales are only hypothetical--they do happen. And I could keep going. There are many reasons why we wouldn't expect the efficient market hypothesis to apply in philanthropy. And it doesn't. Due to the lack of market feedback and the prevalence of idiosyncratic philanthropist/staff preferences, there are many $20 bills lying on the ground for anyone to pick up. This is 1) exciting for potential philanthropists and 2) cause for concern about the overall incentive structure in philanthropy.
I'm really happy to see the endohazard project here. I've been finding the research into (and huge public blind spot around) EDCs and POPs so alarming that it's basically displaced all of my concerns about climate change. I think there's a good chance that we're basically manufacturing another Great Oxygenation Event.
This is well beyond the concerns I've heard about endohazards - can you explain?
There are three properties that worry me: (1) they are dangerous to human and ecological health, (2) they are basically never broken down and so accumulate in human bodies and the environment, and (3) the rate at which we're inventing and manufacturing them is growing.
It's especially number 3 that I think has the biggest tail risk. In the time it took to discover that just BPA was harmful, who knows how many similarly dangerous compounds had entered the supply chain. As more of these chemicals are invented, they'll have more and more useful properties and start replacing ever more previously-harmless materials in new applications. We already store, transport, and prepare/serve food and water in plastic. Medical equipment, toiletries and textiles are often made with PFAS.
If POPs/EDCs are partly responsible for obesity, which seems likely, then obesity will only get worse (and the increasing rate of use of these compounds means they'll likely become the leading cause if they're not already). Obesity, at higher and higher levels, will be a catastrophic blow not only to health, but to our economic and cognitive capabilities as well.
The other (possibly silly) thing I worry about is why I referenced the Great Oxygenation Event. When (not if) bacteria/fungi/something adapt to metabolize plastics, several things might happen: more leeching of EDCs/POPs from this metabolism, loss of plastic-based infrastructure, major damage to food/water/medicine safety (eg. plastic food packaging suddenly being useless), and a new family of pollutants from whatever we replace plastic with.
And with climate change, economic forces are at least partly in the right direction, where solar power and batteries are getting cheaper, fast. But I don't see any natural economic incentives pushing against POPs/EDCs.
"The other (possibly silly) thing I worry about is why I referenced the Great Oxygenation Event. When (not if) bacteria/fungi/something adapt to metabolize plastics, several things might happen: more leeching of EDCs/POPs from this metabolism, loss of plastic-based infrastructure, major damage to food/water/medicine safety (eg. plastic food packaging suddenly being useless), and a new fsmily of pollutants from whatever we replace plastic with."
That concern would seem to mitigate against the grant for plastic-eating beetles!
I do still think it's worthwhile research, but I also fear it misses the main problem of plastic pollution. As I understand it, it's the polymers themselves which have the energy that can be metabolized. But the biologically harmful pollutants (BPA probably being the most famous) tend to be additives to the polymers. So breaking down the polymers risks just releasing more toxins from existing pollution.
Thank you so much for your support and encouragement. I have wondered if focusing on POPs rather than EDCs (a rough subset of many POPs) might be a better focus for labelling, as there is longstanding political backing to phase out POPs, and generally strong evidence of their dangers. However, there is considerable overlap between EDCs and POPs, and the general public encounters EDCs every day, whereas POPs are supposed to be rather restricted. Hence, the focus on EDCs themselves.
If you proposed an ACX grant in an area related to alt-ac (alternative models to the standard University system, broadly conceived), please consider dropping me a line. You can find me by e-mail at sdedeo[at]andrew.cmu.edu
This is because, in tandem with a few colleagues within the system, I’ve been thinking about alternatives to how we provide what is currently described as “elite” education (small seminars, research mentorship, residential, etc). It would be helpful to learn what other people are thinking about.
I really want to know what the Australian political lobbyist guy is planning to do with $90,000. Seems like it could take a lot of different directions, depending on what your policy proposals were and which party you were talking to.
I am also very keen to know.
The beetle thing and plastic.
1. this is such a fun idea you can't help loving it.
2. do we need it? we need to stop throwing plastic in the water system. and collect it all in one place. to either bury or burn.
3. I was thinking of a wood/ plastic analogy. beetles are to plastic as termites are to wood.
if I want to get rid of some wood, do I turn to termites or fire?
One thing that would be cool is if there was some feedback for the rejected proposals, on why it was rejected, even if it's very brief. Even if you disagree with it, it would be awesome to know what potential funders have thought of your idea when first hearing about it.
I second this!
For the person titled D receiving a $10,000 grant, I thought this paper about how early death can be selected for by evolution might be relevant: https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.238103
Just saw this now, thanks!
Personally, like many in the field I think the classical evolutionary theory of aging* is more plausible, or more widely relevant: that it's the result of evolutionary neglect, not itself an adaptation [selection cares less about late life than about early life; if there are trade-offs between them selection favours early life**]. However, I'm also glad that there's a minority of researchers pursuing alternative theories, in case they're true or in case e.g. they affect the rate of ageing even if not being the reason it exists. Here's another recent review, this one focusing on colonial organisms, and emphasizing kin selection not group selection***.
*https://www.nature.com/scitable/knowledge/library/the-evolution-of-aging-23651151/
**For more on the formalism, when it comes out hopefully soon see (my first publication!): https://doi.org/10.1098/rspb.2021.1910; see also my lecture for a class I TA'd: https://drive.google.com/file/d/1PLI8fTFQ2jtpGyRw253coaV6HcV3KvV0/view?usp=sharing.
***https://royalsocietypublishing.org/doi/full/10.1098/rstb.2019.0730
>L, $17,000, to breed a line of beetles that can digest plastic. Darkling beetles (and their associated gut microbes) can already do this a little. Maybe if someone selectively bred them for this ability, they could do it better. Plastic is generally considered bad for the environment because it's "not biodegradable", but maybe everything is biodegradable if you have sufficiently advanced beetles. This project will find out!
The contrarian perspective here is that since plastic is made from hydrocarbons, it's actually a good thing that it doesn't degrade. Throwing non-biodegradable hydrocarbons in the dump = carbon sequestration.
I hope these new beetles don't fart out CO2 or methane after eating plastic...
Interesting! A lot depends on where the plastic is. We don’t consider microplastic in the ocean to be sequestration… maybe we need plastic-eating algae. I suppose we would have to balance the badness of landfills against the badness of atmospheric carbon.
Thanks for doing this and for this post! I'm one of the guest fund managers on one of the EA Funds (the EA Infrastructure Fund, specifically), and *I would really like many of these people to apply to EA Funds for a top up or a substantially larger grant right now (if they haven't already), and for many others to apply later on for "next phases" of these projects or for new projects*. https://funds.effectivealtruism.org/apply-for-funding
This can pretty easily be worthwhile in expectation because:
1. It should take 0.5-2 hours to write an application (setting aside time actually planning the project)
2. The actual evaluation process is typically pretty quick too, for both the applicant and the grant evaluators
3. It's faster for things that don't end up funded (so it's relatively unlikely for lots of time to be spent without impactful-in-expectation work ending up funded)
4. EA Funds's impact is probably most bottlenecked by number of good applications received (more so than by fund manager time or money available) (I'm most confident of this for the Long-Term Future Fund and the Infrastructure Fund)
Scott, did you or someone else already heavily emphasise roughly that message to the grantees? If not, could you do so? Let me know if there's any way I can help (I can be reached at michaeljamesaird AT gmail DOT com )
Here are two posts that might be helpful:
List of EA funding opportunities: https://forum.effectivealtruism.org/posts/DqwxrdyQxcMQ8P2rD/list-of-ea-funding-opportunities
Things I often tell people about applying to EA Funds: https://forum.effectivealtruism.org/posts/4tsWDEXkhincu7HLb/things-i-often-tell-people-about-applying-to-ea-funds
Also, many readers of this comment should probably consider applying too.
(Caveat: This is mostly a message I spam repeatedly in lots of EA-adjacent contexts, rather than something I'm saying because I think lots of these projects in particular sound extremely impactful and funding constrained. And there are many projects listed here that I don't feel very excited about from an impartially altruistic perspective, even if they sound cool from a vaguely-progress-studies perspective. That said, many do sound either likely to be great or *likely enough* to be great that submitting an application is worthwhile in expectation.)
This is kinda what I imagined early university/bell labs or google/steam research looked like. Eclectic, smart people asking for money to try to solve interesting problems and getting money to do so. This was one of the most inspiring posts you've ever put together, intentionally or not. Makes me more optimistic than I've been in a while about what the future holds. Thank you.
Reading all these projects imbues me with some measure of EAmposter Syndrome - so many amazing ideas that can impact so many lives!
Nevertheless, I'm humbled by the grant and trust it represents, and will do my best to live up to it. Back to the 'scope.
Scott, you are probably already doing this, but if not… My wife is a professional fundraiser for a granting organization. I told her about the ACX grants. She recommends that, for tax reasons alone, you establish yourself as a foundation under the appropriate tax code. Especially if this is not going to be a one-off, make sure the IRS doesn’t come and bite you. Anyway, I hope you have already talked to lawyers about this.
> IRS doesn’t come and bite you
Which part would be the problem of not having foundation?
I've seen some concern about bias against Republicans, so I asked at datasecretlox.com about what Republican or (as it turned out, Republican-adjacent) grants would look like, and got some interesting answers, and I'm asking again here.
https://www.datasecretslox.com/index.php/topic,5490.msg199936.html#msg199936
I would be interested in seeing the version of this post with the unsuccessful candidates, maybe those 500 people can submit their own TLDR blurb with a link, and Scott just post the whole thing?
Or a curated subset.
As for a ‘group’ of ‘forum’: the SSC reddit sounds definitely like a nice place for the winners to share follow up stuff and status updates etc etc back to the community! maybe with subject prefixed by an agreed on [ACX 2021 Grantee] tag or something?
> SSC reddit sounds definitely like a nice place
Reddit is filled with ads for "investment" scams (added by Reddit itself), mostly of crypto variety.
This is amazing. I do worry that Scott is too trusting to transfer millions of dollars around, and may get into sticky legal situations. E.g., researcher Z has spent $20,000 on an approved project, but the funders haven't yet delivered Scott enough money to pay researcher Z $20,000; or a major donor died leaving Scott without a legally-binding promise to deliver the money; and Scott is on the hook for it. The usual practice is to have a legal team and some full-time staff to manage contracts and payments.
How will payments work? Will Scott have the money, or legally-binding promises for them, before the grantees start spending money? Do grantees get the money up-front or after-the fact, is it in installments, do the installments require milestone deliverables? Does each project need to deliver a final report for public distribution before receiving their final payment? What is the status of the intellectual property produced? I have no need to know myself, but hope these things get addressed before it's too late.
In the past, before grants came with excruciatingly-detailed contracts outlining milestones, deliverables, and payments, grantees sometimes spent all of the money doing the research, without leaving enough time or money to organize, summarize, and publish the results. There were archaealogical digs which excavated important sites, yet never published any of their findings; so all we have left are boxes of artifacts without information beyond which site they came from.
Today, grantees must deliver reports, but these reports might not be saved, or even read, let alone be published or publicly available.
I don't know if it's possible to sign a binding contract to make a donation in the US. US contracts must have "consideration" to be enforceable, meaning that a donor signing a contract to hand over money must, by that contract, receive something of approximately equal value to that donor in return. So a contract with a donor to donate some sum of money couldn't be binding unless it's tied to specific research projects, and to the donor receiving some sort of compensation from those projects. Perhaps the compensation can be an emotional value, but I doubt it can be anything as general as "the feeling of satisfaction from making the world a better place".
Without contracts for donations, it would be wise to get the donations up-front, so Scott doesn't end up on the hook if some donors fail to pay. Better yet might be for Scott not to be in the loop handling money for all of these grants, but to link each funder with specific grantees and let them work out their own contracts.
My plan is to give everyone their money ASAP. If they take it and run, then they've successfully scammed me and I will feel sad, but I chose people who I thought wouldn't do that, and this is the only option available to me given that I don't have the infrastructure to constantly monitor everyone.
I should have all the money I've been promised within a week or two, after which it will go to the grantees. I've told all of them that until it's in their hands, these aren't legally binding promises. I have a backup plan for who I'm going to ashamedly defund if I get a 95th or 99th percentile bad outcome in terms of people shirking.
Another option (based on my own experience with receiving grants) would be to submit part now and part later, after minimal cursory examination.
It may be a a balance between "infrastructure to constantly monitor everyone" and zero verification.
Don't mean to be rude, but did Alice Evans really need $60k? She's an academic at two elite universities and is doing research which strikes me as broadly indistinguishable from other people working in gender studies. Would this work really not be done without such funding?
And from (an admittedly brief) perusal of her work, it seems she completely shirks consideration of any heritable biological differences between sexes or populations that could at least partly explain the phenomena she investigates, which must surely limit the depth of insight (e.g. concluding that cultural/political factors explain X difference between countries with the implicit idea that these things are just a matter of random chance like which parts of the world have arable land or something).
Can we get a post on why the hell Bryan Caplan never loses a bet??
Jimmy Koppel: Would someone connect me to him?
I'm giving software developers coaching in the Effective Altruism community:
https://forum.effectivealtruism.org/posts/FkWHn6WaFGzrzqb9P/i-m-offering-free-coaching-for-software-developers-in-the-ea
This seems somewhat similar, except (1) most of my discussions are focused on career development, only a few are on technical skills, and (2) I'm approaching the scaling problem differently
I wonder if we'd change each other's minds about something, or collaborate about something, I have no idea really, but it seems worth reaching out
Send me an email so I know your email address and I'll connect you to him.
"A large body of research shows that gender equality, aside from its moral benefits, is also deeply important for economic development." [[Citation Needed]]
I see a large correlation between gdp per capita and gender equality norms, but that could just be that the countries that were smart enough to develop a high GDP per capita were also smart enough to abandon or attenuate their traditional religions.
On the other hand there's a large anticorrelation between gdp per capita and gender equality in choice of profession. So the countries with more gender-egalitarian ideologies have less gender-egalitarian labor allocation. This undermines the OP. My first paragraph seems like a much more likely explanation.
[Edit: maybe they are just talking about the very basic first steps of gender egalitarianism, i.e., giving women access to contraception and control over their own fertility, so that reproduction doesn't outpace the capacity of schools and other infrastructure. That seems very plausibly beneficial to growing GDP per capita in third world countries. But I'm still very doubtful that taking it further than the gender norms of a TFR<2.5 country confers any additional benefit to GDP per capita]
Nuclear tests already generate shockwaves detectable by seismometers thousands of miles away.
The stated plan to detect mere enrichment via neutrinos is impossible. Maybe he means transmutation of U-238 to Pu-239? That involves a beta decay that would release a neutrino.
Solar neutrino flux is circa 10^15 per m^2 per second.
Let's calculate the neutrino flux due to someone producing a critical mass (10kg) of Pu-239 in a year, 1km away.
Total neutrinos = number of plutonium atoms = (10000g/239g)*6.022e23 = 2.5e25
Neutrinos per second = 2.5e25 / (365*24*3600) = 8e17
Surface area of 1km radius sphere = 1.25e7 m^2
Neutrinos per second intersecting 1m^2 target at 1km distance = 6.4e10
So that's four orders of magnitude below the background noise.
But maybe you can very precisely detect the energy of the neutrinos and thus distinguish between plutonium neutrinos and all others. You'd also need to simultaneously get a precise direction vector, because there are hundreds of legitimate plutonium producers around the globe. (e.g., every uranium reactor has a side effect of making some plutonium). I don't know enough physics to say whether it's possible to design a neutrino detector with those parameters. But I know that the largest neutrino detector in the world (https://icecube.wisc.edu/about-us/facts/#:~:text=IceCube%20detects%20275%20atmospheric%20neutrinos%20daily%20and%20about%20100%2C000%20per%20year) gets about 275 neutrinos per day out of a solar flux ~15000 times bigger than the signal we're looking for. So in the case of the guy making 10kg of plutonium in a year, 1km away, it'd receive about 7 neutrinos in a year. So in terms of quantity alone, detection is just barely possible if you're within 1km of the largest neutrino detector in the world. If you're not on the same continent, (it's located at the south pole), forget about it.
What a cool suite of ideas and initiatives. And your insights on grantmaking and the process are just as valuable. Delighted to see the good you are doing (even though my idea didn't get up).
I wonder what the status of these grants is now that more than a year is over and there we can start judging their outcomes