I just came back from talking to Max Harms about the Crystal trilogy, which made me think about rationalist fiction, or the concept of hard sci-fi combined with explorations of cognitive science and philosophy of science in general (which is how I conceptualize the idea of rationalist fiction).
I have a general sense that one of the biggest obstacles for making progress on difficult problems is something that I would describe as “focusing attention on the problem”. I feel like after an initial burst of problem-solving activity, most people when working on hard problems, either give up, or start focusing on ways to avoid the problem, or sometimes start building a lot of infrastructure around the problem in a way that doesn’t really try to solve it.
I feel like one of the most important tools/skills that I see top scientist or problem solvers in general use, is utilizing workflows and methods that allow them to focus on a difficult problem for days and months, instead of just hours.
I think at least for me, the case of exam environments displays this effect pretty strongly. I have a sense that in an exam environment, if I am given a question, I successfully focus my full attention on a problem for a full hour, in a way that often easily outperforms me thinking about a problem in a lower key environment for multiple days in a row.
And then, when I am given a problem set with concrete technical problems, my attention is again much better focused than when I am given the same problem but in a much less well-defined way. E.g. thinking about solving some engineering problem, but without thinking about it by trying to create a concrete proof or counterproof.
My guess is that there is a lot of potential value in fiction that helps people focus their attention on a problem in a real way. In fiction you have the ability to create real-feeling stakes that depend on problem solving, and things like the final exam in Methods of Rationality show how that can be translated into large amounts of cognitive labor.
I think my strongest counterargument to this model is something like “sure, it’s easy to make progress on problems when you have someone else give you the correct ontology in which the problem is solvable, but that’s just because 90% of the work of solving problems is coming up with the right ontologies for problems like this”. And I think there is something importantly real about this, but also that it doesn’t fully address the value of exams and fiction and problem sets that I am trying to point to (though I do think it explains a good chunk of their effect).
Going back to the case of fiction, it is clear to me that fiction is as a literary form much more optimized to hold human attention that most non-fiction is. I think first of all that this constraint means that most fiction (and in particular most popular fiction) isn’t about much else than whatever best holds people’s attention, but it also means that if the bottleneck on a lot of problems is just getting people to hold their attention on the problem for a while, then utilizing the methods that fiction-writing has developed seems like an obvious way of making progress on those problems.
I feel like another major effect that explains a lot of the effects that I observe is people believing that a problem is solvable. In a fictional setting, if the author promises you that things have a good explanation, then it’s motivating to figure out why. On an exam you are promised that the problems that you are given are solvable, and solvable within a reasonable amount of time.
I do think this can still be exploited. In the last few chapters of HPMOR, Harry does a mental motion that I would describe as “don’t waste mental energy on asking whether a problem is solvable, just pretend it it, and ask what the solution would be if it was solvable”, in a way that felt to me like it would work on a lot of real-world problems.
I feel like one of the most important tools/skills that I see top scientist or problem solvers in general use, is utilizing workflows and methods that allow them to focus on a difficult problem for days and months, instead of just hours.
This is a really important point, which I kind of understood (“research” means having threads of inquiry that extend into the past and future), but I hadn’t been thinking of it in terms of workflows that facilitate that kind of engagement.
nods I’ve gotten a lot of mileage over the years from thinking about workflows and systems that systematically direct your attention towards various parts of reality.
I suspect that fiction can conveniently ignore the details of real life that could ruin seemingly good plans.
Let’s look at HPMOR.
The general idea of “create a nano-wire, then use it to simultaneously kill/cripple all your opponents” sounds good on paper. Now imagine yourself, at that exact situation, trying to actually do it. What could possibly go wrong?
As a first objection, how would you actually put the nano-wire in the desired position? Especially when you can’t even see it (otherwise the Death Eaters and Voldemort would see it too). One mistake would ruin the entire plan. What if the wind blows and moves your wire? If one of the Death Eaters moves a bit, and feels a weird stinging at the side of their neck?
Another objection, when you pull the wire to kill/cripple your opponents, how far do you actually have to move it? Assuming dozen Death Eaters (I do not remember the exact number in the story), if you need 10 cm for an insta-kill, that’s 1.2 meters you need to do before the last one kills you. Sounds doable, but also like something that could possibly go wrong.
In other words, I think that in real life, even Harry Potter’s plan would most likely fail. And if he is smart enough, he would know it.
The implication for real life is that, similarly, smart plans are still likely to fail, and you know it. Which is probably why you are not trying hard enough. You probably already remember situations in your past when something seemed like a great idea, but still failed. Your brain may predict that your new idea would belong to the same reference class.
While I agree that this is right, your two objections are both explicitly addressed within the relevant chapter:
“As a first objection, how would you actually put the nano-wire in the desired position? Especially when you can’t even see it (otherwise the Death Eaters and Voldemort would see it too). One mistake would ruin the entire plan. What if the wind blows and moves your wire? If one of the Death Eaters moves a bit, and feels a weird stinging at the side of their neck?”
Harry first transfigures a much larger spiderweb, which also has the advantage of being much easier to move in place, and to not be noticed by people that are interacting with it.
“Another objection, when you pull the wire to kill/cripple your opponents, how far do you actually have to move it? Assuming dozen Death Eaters (I do not remember the exact number in the story), if you need 10 cm for an insta-kill, that’s 1.2 meters you need to do before the last one kills you. Sounds doable, but also like something that could possibly go wrong.”
Indeed, which is why Harry was waving the web into an intervowen circle that contracts simultaneously in all directions.
Obviously things could have still gone wrong, and Eliezer has explicitly acknowledged that HPMOR is a world in which complicated plans definitely succeed a lot more than they would in the normal world, but he did try to at least cover the obvious ways things could go wrong.
Yes, fiction has a lot of potential to change mindsets. Many Philosophers actually look at the greatest novel writers to infer the motives and the solutions their heroes to come up with general theories that touch the very core of how our society is laid out.
Most of this come from the fact that we are already immersed in a meta-story, externally and internally. Much of our efforts are focused on internal rationalizations to gain something where a final outcome has been already thought out, this being consciously known to us or not.
I think that in fiction this is laid out perfectly. So analyzing fiction is rewarding in a sense. Specially when realizing that when we go to exams or interviews we’re rapidly immersing ourselves in an isolated story with motives and objectives (what we expect to happen), we create our own little world, our own little stories.
I just came back from talking to Max Harms about the Crystal trilogy, which made me think about rationalist fiction, or the concept of hard sci-fi combined with explorations of cognitive science and philosophy of science in general (which is how I conceptualize the idea of rationalist fiction).
I have a general sense that one of the biggest obstacles for making progress on difficult problems is something that I would describe as “focusing attention on the problem”. I feel like after an initial burst of problem-solving activity, most people when working on hard problems, either give up, or start focusing on ways to avoid the problem, or sometimes start building a lot of infrastructure around the problem in a way that doesn’t really try to solve it.
I feel like one of the most important tools/skills that I see top scientist or problem solvers in general use, is utilizing workflows and methods that allow them to focus on a difficult problem for days and months, instead of just hours.
I think at least for me, the case of exam environments displays this effect pretty strongly. I have a sense that in an exam environment, if I am given a question, I successfully focus my full attention on a problem for a full hour, in a way that often easily outperforms me thinking about a problem in a lower key environment for multiple days in a row.
And then, when I am given a problem set with concrete technical problems, my attention is again much better focused than when I am given the same problem but in a much less well-defined way. E.g. thinking about solving some engineering problem, but without thinking about it by trying to create a concrete proof or counterproof.
My guess is that there is a lot of potential value in fiction that helps people focus their attention on a problem in a real way. In fiction you have the ability to create real-feeling stakes that depend on problem solving, and things like the final exam in Methods of Rationality show how that can be translated into large amounts of cognitive labor.
I think my strongest counterargument to this model is something like “sure, it’s easy to make progress on problems when you have someone else give you the correct ontology in which the problem is solvable, but that’s just because 90% of the work of solving problems is coming up with the right ontologies for problems like this”. And I think there is something importantly real about this, but also that it doesn’t fully address the value of exams and fiction and problem sets that I am trying to point to (though I do think it explains a good chunk of their effect).
Going back to the case of fiction, it is clear to me that fiction is as a literary form much more optimized to hold human attention that most non-fiction is. I think first of all that this constraint means that most fiction (and in particular most popular fiction) isn’t about much else than whatever best holds people’s attention, but it also means that if the bottleneck on a lot of problems is just getting people to hold their attention on the problem for a while, then utilizing the methods that fiction-writing has developed seems like an obvious way of making progress on those problems.
I feel like another major effect that explains a lot of the effects that I observe is people believing that a problem is solvable. In a fictional setting, if the author promises you that things have a good explanation, then it’s motivating to figure out why. On an exam you are promised that the problems that you are given are solvable, and solvable within a reasonable amount of time.
I do think this can still be exploited. In the last few chapters of HPMOR, Harry does a mental motion that I would describe as “don’t waste mental energy on asking whether a problem is solvable, just pretend it it, and ask what the solution would be if it was solvable”, in a way that felt to me like it would work on a lot of real-world problems.
This is a really important point, which I kind of understood (“research” means having threads of inquiry that extend into the past and future), but I hadn’t been thinking of it in terms of workflows that facilitate that kind of engagement.
nods I’ve gotten a lot of mileage over the years from thinking about workflows and systems that systematically direct your attention towards various parts of reality.
Warning: HPMOR spoilers!
I suspect that fiction can conveniently ignore the details of real life that could ruin seemingly good plans.
Let’s look at HPMOR.
The general idea of “create a nano-wire, then use it to simultaneously kill/cripple all your opponents” sounds good on paper. Now imagine yourself, at that exact situation, trying to actually do it. What could possibly go wrong?
As a first objection, how would you actually put the nano-wire in the desired position? Especially when you can’t even see it (otherwise the Death Eaters and Voldemort would see it too). One mistake would ruin the entire plan. What if the wind blows and moves your wire? If one of the Death Eaters moves a bit, and feels a weird stinging at the side of their neck?
Another objection, when you pull the wire to kill/cripple your opponents, how far do you actually have to move it? Assuming dozen Death Eaters (I do not remember the exact number in the story), if you need 10 cm for an insta-kill, that’s 1.2 meters you need to do before the last one kills you. Sounds doable, but also like something that could possibly go wrong.
In other words, I think that in real life, even Harry Potter’s plan would most likely fail. And if he is smart enough, he would know it.
The implication for real life is that, similarly, smart plans are still likely to fail, and you know it. Which is probably why you are not trying hard enough. You probably already remember situations in your past when something seemed like a great idea, but still failed. Your brain may predict that your new idea would belong to the same reference class.
While I agree that this is right, your two objections are both explicitly addressed within the relevant chapter:
“As a first objection, how would you actually put the nano-wire in the desired position? Especially when you can’t even see it (otherwise the Death Eaters and Voldemort would see it too). One mistake would ruin the entire plan. What if the wind blows and moves your wire? If one of the Death Eaters moves a bit, and feels a weird stinging at the side of their neck?”
Harry first transfigures a much larger spiderweb, which also has the advantage of being much easier to move in place, and to not be noticed by people that are interacting with it.
“Another objection, when you pull the wire to kill/cripple your opponents, how far do you actually have to move it? Assuming dozen Death Eaters (I do not remember the exact number in the story), if you need 10 cm for an insta-kill, that’s 1.2 meters you need to do before the last one kills you. Sounds doable, but also like something that could possibly go wrong.”
Indeed, which is why Harry was waving the web into an intervowen circle that contracts simultaneously in all directions.
Obviously things could have still gone wrong, and Eliezer has explicitly acknowledged that HPMOR is a world in which complicated plans definitely succeed a lot more than they would in the normal world, but he did try to at least cover the obvious ways things could go wrong.
I have covered both of your spoilers in spoiler tags (“>!”).
Yes, fiction has a lot of potential to change mindsets. Many Philosophers actually look at the greatest novel writers to infer the motives and the solutions their heroes to come up with general theories that touch the very core of how our society is laid out.
Most of this come from the fact that we are already immersed in a meta-story, externally and internally. Much of our efforts are focused on internal rationalizations to gain something where a final outcome has been already thought out, this being consciously known to us or not.
I think that in fiction this is laid out perfectly. So analyzing fiction is rewarding in a sense. Specially when realizing that when we go to exams or interviews we’re rapidly immersing ourselves in an isolated story with motives and objectives (what we expect to happen), we create our own little world, our own little stories.