“genius” doesn’t seem usefully precise to me? (Is a genius even still a genius once they’ve found their way into a part of the world where their level of pragmatic creativity is ordinary?) I’m looking for a sort of… ability to go for extended, rapid, complicated traversals of broad unfamiliar territories in your head, alone, without getting lost, and to find something of demonstrable value that no one has ever seen before. That kind of thing.
That list might be a good start, but I don’t know. Can you show us examples of divergent, multi-stage, needle in a haystack breakthroughs that those people made while they were years into a vegan diet? I haven’t looked closely at really any of these peoples’ work, and there’s a kind of relevant reason for that. A lot of them (Singer, MacAskill) are mostly apologists. They work mainly with familiar premises in highly legible ways. The reason most people read them is the reason I don’t read them, and the reason I am concerned that a vegan diet tends to limit the ways people can think.
Tomasik is an interesting example though, I’ve gotten the sense that he has that character, but haven’t seen any intense output from him. Recommend some?
Examples of EA-adjacent people who I’d consider to have this quality include Yudkowsky, Wei Dai, Vanessa Kosoy, Robin Hanson (none were vegan during their breakthroughs afaict?).
It might be worth asking which way the causality’s running here. A very EA-charitable answer might say something like: “Being humble and accountable (which leads to doing less risky, more legible, and more approachable work) probably raises a person’s inclination to become vegan.” (It’s kind of interesting that, as far as I can tell, long time vegans, the Brahmins, would argue the opposite causality: “Being vegan decreases the mode of passion, but that’s good for your spiritual path.”)
Maybe Derek Parfit (vegetarian), Chris Olah (vegan), Mark Xu (vegan until this year https://markxu.com/transitioning-vegan), Rohin Shah (~vegan) are other examples?
I think there’s been some impressive technical work out of GPI, and generally in population ethics and decision theory, and I have specific authors in mind, but I can’t tell if they’ve been vegetarian through Google. If you’re really invested in this, I can share names and papers, and you can ask them directly if they’ve been veg.
I’d say people working in population ethics are reasonably likely to be veg.
Do you have specific works by EAs or EA-adjacent people involving “divergent, multi-stage, needle in a haystack breakthroughs” in mind? And are multi-stage (sequentially dependant?) breakthroughs more impressive than a similar number of breakthroughs that aren’t sequentially dependant or that happen far apart in time from each other? Or are you thinking of something where a single breakthrough isn’t enough on its own for useful or interesting conclusions, and more are needed until something valuable can be produced produced?
are multi-stage (sequentially dependant?) breakthroughs more impressive than a similar number of breakthroughs that aren’t sequentially dependant or that happen far apart in time from each other?
Yes, because… it means they couldn’t have been finding low-hanging fruit. When one problem leads to another, you don’t get to wander off and look for easier ones, you have to keep going down one of these few avenues of this particular cave system. So if someone solved a contiguous chain of problems you can be sure that some of those were probably genuinely really hard. It also requires them to develop their own understanding of something that nobody could help them with, and to internalize that deeply enough to keep going.
Sequences like this occur naturally in real-world projects, so if they’re avoiding them it’s kinda telling.
more are needed until something valuable can be produced produced?
?. More are needed before we can make a judgement. I’d believe that lots of value can be produced without any of these big leaps.
But you can also just judge each breakthrough separately, conditional on what they had access to. If they’re deep into a problem past where anyone has been and then go further, that might be more impressive, but it may not be, in case it’s easy to identify the next (possibly hard) subproblem after solving the last subproblem. So you can approach it locally/greedily, without thinking ahead much to where you need to go, only about where you are now and the next step. I think upper-year and grad-level pure math and theoretical computer science problems can be like this, although maybe not as hard as you’re asking for.
Something harder I have in mind would be something like having a non-local/non-greedy approach to solving a problem, where you have a major breakthrough just to get a sketch of a proof or to come up with possible lemmas, and then it takes further breakthroughs to close things up. If your sketch is wrong, then all the work can become basically useless, and you don’t progress things for the next people to try (except by ruling out a dead end).
It’s like thinking more moves ahead in chess with multiple hard moves to identify, compared to just making the same number of hard moves to identify in the same game, but never part of the same sequence simulation in your head. Both are multi-stage, but the second one is local/greedy and isn’t more impressive than making the same number of hard to identify moves across games, fixing the total number of games played.
Also, breakthroughs across very different areas rather than all concentrated in the same area demonstrates greater flexibility and generalizability of their strengths.
“genius” doesn’t seem usefully precise to me? (Is a genius even still a genius once they’ve found their way into a part of the world where their level of pragmatic creativity is ordinary?)
I’m looking for a sort of… ability to go for extended, rapid, complicated traversals of broad unfamiliar territories in your head, alone, without getting lost, and to find something of demonstrable value that no one has ever seen before. That kind of thing.
That list might be a good start, but I don’t know. Can you show us examples of divergent, multi-stage, needle in a haystack breakthroughs that those people made while they were years into a vegan diet? I haven’t looked closely at really any of these peoples’ work, and there’s a kind of relevant reason for that. A lot of them (Singer, MacAskill) are mostly apologists. They work mainly with familiar premises in highly legible ways. The reason most people read them is the reason I don’t read them, and the reason I am concerned that a vegan diet tends to limit the ways people can think.
Tomasik is an interesting example though, I’ve gotten the sense that he has that character, but haven’t seen any intense output from him. Recommend some?
Examples of EA-adjacent people who I’d consider to have this quality include Yudkowsky, Wei Dai, Vanessa Kosoy, Robin Hanson (none were vegan during their breakthroughs afaict?).
It might be worth asking which way the causality’s running here. A very EA-charitable answer might say something like: “Being humble and accountable (which leads to doing less risky, more legible, and more approachable work) probably raises a person’s inclination to become vegan.” (It’s kind of interesting that, as far as I can tell, long time vegans, the Brahmins, would argue the opposite causality: “Being vegan decreases the mode of passion, but that’s good for your spiritual path.”)
Maybe Derek Parfit (vegetarian), Chris Olah (vegan), Mark Xu (vegan until this year https://markxu.com/transitioning-vegan), Rohin Shah (~vegan) are other examples?
I think there’s been some impressive technical work out of GPI, and generally in population ethics and decision theory, and I have specific authors in mind, but I can’t tell if they’ve been vegetarian through Google. If you’re really invested in this, I can share names and papers, and you can ask them directly if they’ve been veg.
I’d say people working in population ethics are reasonably likely to be veg.
Do you have specific works by EAs or EA-adjacent people involving “divergent, multi-stage, needle in a haystack breakthroughs” in mind? And are multi-stage (sequentially dependant?) breakthroughs more impressive than a similar number of breakthroughs that aren’t sequentially dependant or that happen far apart in time from each other? Or are you thinking of something where a single breakthrough isn’t enough on its own for useful or interesting conclusions, and more are needed until something valuable can be produced produced?
Yes, because… it means they couldn’t have been finding low-hanging fruit. When one problem leads to another, you don’t get to wander off and look for easier ones, you have to keep going down one of these few avenues of this particular cave system. So if someone solved a contiguous chain of problems you can be sure that some of those were probably genuinely really hard. It also requires them to develop their own understanding of something that nobody could help them with, and to internalize that deeply enough to keep going.
Sequences like this occur naturally in real-world projects, so if they’re avoiding them it’s kinda telling.
?. More are needed before we can make a judgement. I’d believe that lots of value can be produced without any of these big leaps.
But you can also just judge each breakthrough separately, conditional on what they had access to. If they’re deep into a problem past where anyone has been and then go further, that might be more impressive, but it may not be, in case it’s easy to identify the next (possibly hard) subproblem after solving the last subproblem. So you can approach it locally/greedily, without thinking ahead much to where you need to go, only about where you are now and the next step. I think upper-year and grad-level pure math and theoretical computer science problems can be like this, although maybe not as hard as you’re asking for.
Something harder I have in mind would be something like having a non-local/non-greedy approach to solving a problem, where you have a major breakthrough just to get a sketch of a proof or to come up with possible lemmas, and then it takes further breakthroughs to close things up. If your sketch is wrong, then all the work can become basically useless, and you don’t progress things for the next people to try (except by ruling out a dead end).
It’s like thinking more moves ahead in chess with multiple hard moves to identify, compared to just making the same number of hard moves to identify in the same game, but never part of the same sequence simulation in your head. Both are multi-stage, but the second one is local/greedy and isn’t more impressive than making the same number of hard to identify moves across games, fixing the total number of games played.
Also, breakthroughs across very different areas rather than all concentrated in the same area demonstrates greater flexibility and generalizability of their strengths.