This is a mean vs median or Mediocristan vs Extremistan issue. Most people cannot do lone wolf, but if you can do lone wolf, you will probably be much more successful than the average person.
Think of it like this. Say you wanted to become a great writer. You could go to university and plod through a major in English literature. That will reliably give you a middling good skill at writing. Or you could drop out and spend all your time reading sci-fi novels, watching anime, and writing fan fiction. Now most people who do that will end up terrible writers. But when someone like Eliezer does it, the results are spectacular.
Furthermore, because of the Power Law and the “Average is Over” idea, most of the impact will come from the standout successes.
This is a mean vs median or Mediocristan vs Extremistan issue. Most people cannot do lone wolf, but if you can do lone wolf, you will probably be much more successful than the average person.
I cannot disagree with this more strongly. I am serial entrepreneur, and a somewhat successful one. Still chasing the big exit, but I’ve built successful companies that are still private. Besides myself I’ve met many other people in this industry which you’d be excused for thinking are lone wolfs. But the truth is the lone wolf’s don’t make it as they build things that fail to have product/market fit, fail to listen to feedback if and when it is even made available to them (since they don’t seek it), and usually fail to raise or maintain funding from lack of communication and organizational skill.
The successful entrepreneurs, hedge funders, etc. are not afraid of thinking that conventional wisdom is wrong. The success they have is not from trailblazing a new path—that just goes with doing something new—but from having the tenacity to ask “but why is that so?” of conventional wisdom. Every now and then you find something that just shouldn’t be so—it has no good justification except historical accident—and then you execute. And a very important part of execution is building a team that can work together to avoid the heuristics and biases that follow lone wolfs around.
Don’t be a lone wolf. Be a social rationalist willing to question everything and go where that takes you. It’s not the same thing.
I agree with you in the context of entrepreneurship, but the OP was talking about self improvement. The best strategy for learning or self-improving may be very different from the best strategy for building a company.
Most people cannot do lone wolf, but if you can do lone wolf, you will probably be much more successful than the average person.
Maybe we disagree on what it means to “lone wolf.” If I try to steel-man your position, I can come up with a weak and a strong interpretation:
The weak interpretation is that being a autodidact (capable of learning things on your own) will bring you higher chances of success. Being an autodidact myself, I agree from anecdotal experience. Also just being an expert in your field means developing autodidact skills at some point because eventually you surpass the level of all available classes and have to learn from the latest research journals and technical reports. However I would argue that this should still remain a social activity where you continue to interact with collaborators and bounce ideas off of trusted colleagues in order to avoid many of the pitfalls that come from truly working alone. This isn’t a lone wolf so much as a free-thinking pack wolf, to carry the metaphor, that enjoys the best of both worlds.
The strong interpretation is that you will or even can be successful by truly embarking on a lone quest all by yourself. It is this interpretation that I disagree with so strongly for the reasons given. In my experience smart people who go the “lone wolf” route inevitably end up in crackpot / crank territory as they accumulate bad ideas in their personal blind spots, assuming they don’t fall prey to akrasia in the first place. In this sense I agree with the OP: glorifying the “lone wolf” path has done a lot of harm to a lot of LW’ers.
Taking classes is a relatively Mediocristan-style way to work with others, but there are other ways that get you Extremistan-style upside.
One way is to find a close collaborator or two. Amos Tversky and Daniel Kahneman had an extremely close collaboration, doing most of their thinking in conversation as they were developing the field of heuristics and biases research (as described in The Undoing Project). It’s standard startup advice to have more than one founder so that you’ll have someone “to brainstorm with, to talk you out of stupid decisions, and to cheer you up when things go wrong.” Etc.
Another way is to have a group of several people who are all heavily into the same thing. If the documentary Dogtown and Z-Boys is accurate, many innovations in skateboarding came from a group of teenage skateboarders who hung out together in Southern California in the 1970s. People who are trying to understand the state of the art in an intellectual field often put together a reading group to discuss the latest essays in that field and spin off their own ideas (e.g., John Stuart Mill talks about studying political economy and syllogistic logic in this way, which led to new ideas & publications). Etc.
I think this discussion is somewhat confused by the elision of the difference between ‘autodidact’ and ‘lone wolf’. ‘Autodidact’, in internet circles, is generally used to mean ‘anyone who learns things primarily outside a formalized educational environment’; it’s possible to be an autodidact while still being heavily engaged with communities and taking learning things as a social endeavor and so on, and in fact Eliezer was active in communities related to LW’s subject matter for a long time before he started LW. By the same token, one of the main things I took from reading Ben Franklin’s autobiography was that, despite having little formal schooling and being solely credited for many of his innovations, he didn’t actually do it alone. I doubt he would’ve been even a tenth as successful as he was without something like his Junto.
Some people will get more out of formal education than others, although getting things out of formal education is itself a skill that can be learned. (It seems to require an ability to buy into institutions on an emotional level that many of us lack. I saw college as an obnoxious necessity rather than a set of opportunities, and as a result got much less out of it than I could have. This seems to be a common mistake.) But I just don’t think it’s possible to become a spectacular writer, or even a middling one, as a lone wolf. If nothing else, you need feedback from a community in order to improve. Look at lone-wolf outsider art—it’s frequently unusual, but how much of it is good?
Now most people who do that will end up terrible writers. But when someone like Eliezer does it, the results are spectacular.
You could have found a more convincing example.
The objective metrics of quality of literature are hard to come by, but HPMOR does suffer from quite many, many stereotypical sins of fanfic / bad genre writing and makes a tiresome read. (One that I found especially grating and made me finally drop the story altogether could be described as “my main character is not OP because I set up these plainly arbitrary obstacles as a ‘balance’ ”. Please, no. There’s more to writing enjoyable, interesting characters in meaningful stories than such naive “balancing”. The preferable end result is a piece of fiction that has something more going in them than surface-level entertainment plot which is amenable to measured in terms such as “is my character OP”.)
However, I did not register an account just to lambaste Eliezer’s fiction. Here’s a couple of points that hopefully tie this comment to the main thread of discussion (so that this contribution provides some signal instead of pure noise):
Taking a year-long course in lit or at least some input from the tradition of literature might have improved Eliezer’s writing. A class isn’t the only way to attain that input, but it certainly helps in finding out if you have missed something vital in your self-study. (After finding out those pieces of information you are free to judge and dismiss them, too, if you want, but you are now dismissing that information with the knowledge that it exists, which is prone to make your act of dismissal more intelligent and productive.)
I don’t have exhaustive collection of biographies at hand, but I believe the general trend in “successful writing” is that the significant portion (probably majority) of successful writers (including, but not limited to, authors included in the Western canon as creators of “good literature”), read a lot, wrote a lot, and had a lot of corrective input to improve their writing during their writing careers. Actually, I wouldn’t be surprised if prior to starting a writing career, they already had lots of corrective input to improve their general storytelling skills before they even started considering writing. The input came in the form of teachers, friends who reviewed their manuscripts in process (or commented on their stories), how popular their first published drabbles were, and (often most importantly) the input given by the publishing house editors whose job is to improve writing of authors who submit their fiction for publishing.
Also the first few dozen chapters of HPMoR are terribly written. It is rather horrid, strained, constipated writing. Particularly if you view the early releases of the text, not the revised text that is currently available. The writing got decently good towards the middle, and was top notch by the end. But that was after thousands of pages written and lots of feedback on every chapter. No surprise, lots of writing practice and (critically, to the point of this thread:) feedback leads to becoming a better writer.
Thanks for pointing that out; maybe one important part that I left implicit is the feedback coming from domain experts (such as the staff of a major publishing house).
Why not both? The English literature lessons and sci-fi novels / anime / fan fiction.
I don’t know much about writing, but e.g. studying computer science at universities does not seem to prevent people from creating open source software.
You can also have impact by being okay at several things. For example, I’m okay at both algorithms and UI, so I can get hired easily and then have my pick of projects (because most programmers prefer backend work). If I was also okay at writing or management, I’d probably be rich by now. I think Eliezer’s success was also due to being good at several things, not being the best at one thing.
This is a mean vs median or Mediocristan vs Extremistan issue. Most people cannot do lone wolf, but if you can do lone wolf, you will probably be much more successful than the average person.
Think of it like this. Say you wanted to become a great writer. You could go to university and plod through a major in English literature. That will reliably give you a middling good skill at writing. Or you could drop out and spend all your time reading sci-fi novels, watching anime, and writing fan fiction. Now most people who do that will end up terrible writers. But when someone like Eliezer does it, the results are spectacular.
Furthermore, because of the Power Law and the “Average is Over” idea, most of the impact will come from the standout successes.
I cannot disagree with this more strongly. I am serial entrepreneur, and a somewhat successful one. Still chasing the big exit, but I’ve built successful companies that are still private. Besides myself I’ve met many other people in this industry which you’d be excused for thinking are lone wolfs. But the truth is the lone wolf’s don’t make it as they build things that fail to have product/market fit, fail to listen to feedback if and when it is even made available to them (since they don’t seek it), and usually fail to raise or maintain funding from lack of communication and organizational skill.
The successful entrepreneurs, hedge funders, etc. are not afraid of thinking that conventional wisdom is wrong. The success they have is not from trailblazing a new path—that just goes with doing something new—but from having the tenacity to ask “but why is that so?” of conventional wisdom. Every now and then you find something that just shouldn’t be so—it has no good justification except historical accident—and then you execute. And a very important part of execution is building a team that can work together to avoid the heuristics and biases that follow lone wolfs around.
Don’t be a lone wolf. Be a social rationalist willing to question everything and go where that takes you. It’s not the same thing.
I agree with you in the context of entrepreneurship, but the OP was talking about self improvement. The best strategy for learning or self-improving may be very different from the best strategy for building a company.
Your post said:
Maybe we disagree on what it means to “lone wolf.” If I try to steel-man your position, I can come up with a weak and a strong interpretation:
The weak interpretation is that being a autodidact (capable of learning things on your own) will bring you higher chances of success. Being an autodidact myself, I agree from anecdotal experience. Also just being an expert in your field means developing autodidact skills at some point because eventually you surpass the level of all available classes and have to learn from the latest research journals and technical reports. However I would argue that this should still remain a social activity where you continue to interact with collaborators and bounce ideas off of trusted colleagues in order to avoid many of the pitfalls that come from truly working alone. This isn’t a lone wolf so much as a free-thinking pack wolf, to carry the metaphor, that enjoys the best of both worlds.
The strong interpretation is that you will or even can be successful by truly embarking on a lone quest all by yourself. It is this interpretation that I disagree with so strongly for the reasons given. In my experience smart people who go the “lone wolf” route inevitably end up in crackpot / crank territory as they accumulate bad ideas in their personal blind spots, assuming they don’t fall prey to akrasia in the first place. In this sense I agree with the OP: glorifying the “lone wolf” path has done a lot of harm to a lot of LW’ers.
Taking classes is a relatively Mediocristan-style way to work with others, but there are other ways that get you Extremistan-style upside.
One way is to find a close collaborator or two. Amos Tversky and Daniel Kahneman had an extremely close collaboration, doing most of their thinking in conversation as they were developing the field of heuristics and biases research (as described in The Undoing Project). It’s standard startup advice to have more than one founder so that you’ll have someone “to brainstorm with, to talk you out of stupid decisions, and to cheer you up when things go wrong.” Etc.
Another way is to have a group of several people who are all heavily into the same thing. If the documentary Dogtown and Z-Boys is accurate, many innovations in skateboarding came from a group of teenage skateboarders who hung out together in Southern California in the 1970s. People who are trying to understand the state of the art in an intellectual field often put together a reading group to discuss the latest essays in that field and spin off their own ideas (e.g., John Stuart Mill talks about studying political economy and syllogistic logic in this way, which led to new ideas & publications). Etc.
I think this discussion is somewhat confused by the elision of the difference between ‘autodidact’ and ‘lone wolf’. ‘Autodidact’, in internet circles, is generally used to mean ‘anyone who learns things primarily outside a formalized educational environment’; it’s possible to be an autodidact while still being heavily engaged with communities and taking learning things as a social endeavor and so on, and in fact Eliezer was active in communities related to LW’s subject matter for a long time before he started LW. By the same token, one of the main things I took from reading Ben Franklin’s autobiography was that, despite having little formal schooling and being solely credited for many of his innovations, he didn’t actually do it alone. I doubt he would’ve been even a tenth as successful as he was without something like his Junto.
Some people will get more out of formal education than others, although getting things out of formal education is itself a skill that can be learned. (It seems to require an ability to buy into institutions on an emotional level that many of us lack. I saw college as an obnoxious necessity rather than a set of opportunities, and as a result got much less out of it than I could have. This seems to be a common mistake.) But I just don’t think it’s possible to become a spectacular writer, or even a middling one, as a lone wolf. If nothing else, you need feedback from a community in order to improve. Look at lone-wolf outsider art—it’s frequently unusual, but how much of it is good?
You could have found a more convincing example.
The objective metrics of quality of literature are hard to come by, but HPMOR does suffer from quite many, many stereotypical sins of fanfic / bad genre writing and makes a tiresome read. (One that I found especially grating and made me finally drop the story altogether could be described as “my main character is not OP because I set up these plainly arbitrary obstacles as a ‘balance’ ”. Please, no. There’s more to writing enjoyable, interesting characters in meaningful stories than such naive “balancing”. The preferable end result is a piece of fiction that has something more going in them than surface-level entertainment plot which is amenable to measured in terms such as “is my character OP”.)
However, I did not register an account just to lambaste Eliezer’s fiction. Here’s a couple of points that hopefully tie this comment to the main thread of discussion (so that this contribution provides some signal instead of pure noise):
Taking a year-long course in lit or at least some input from the tradition of literature might have improved Eliezer’s writing. A class isn’t the only way to attain that input, but it certainly helps in finding out if you have missed something vital in your self-study. (After finding out those pieces of information you are free to judge and dismiss them, too, if you want, but you are now dismissing that information with the knowledge that it exists, which is prone to make your act of dismissal more intelligent and productive.)
I don’t have exhaustive collection of biographies at hand, but I believe the general trend in “successful writing” is that the significant portion (probably majority) of successful writers (including, but not limited to, authors included in the Western canon as creators of “good literature”), read a lot, wrote a lot, and had a lot of corrective input to improve their writing during their writing careers. Actually, I wouldn’t be surprised if prior to starting a writing career, they already had lots of corrective input to improve their general storytelling skills before they even started considering writing. The input came in the form of teachers, friends who reviewed their manuscripts in process (or commented on their stories), how popular their first published drabbles were, and (often most importantly) the input given by the publishing house editors whose job is to improve writing of authors who submit their fiction for publishing.
Eliezer did have a lot of writing practice and feedback since childhood, but with a peculiar audience (transhumanists on the net).
Also the first few dozen chapters of HPMoR are terribly written. It is rather horrid, strained, constipated writing. Particularly if you view the early releases of the text, not the revised text that is currently available. The writing got decently good towards the middle, and was top notch by the end. But that was after thousands of pages written and lots of feedback on every chapter. No surprise, lots of writing practice and (critically, to the point of this thread:) feedback leads to becoming a better writer.
I feel uncomfortable criticizing HPMoR for its writing when it clearly succeeded at its job beyond all expectation.
I feel no qualms for calling a spade a spade.
Thanks for pointing that out; maybe one important part that I left implicit is the feedback coming from domain experts (such as the staff of a major publishing house).
Why not both? The English literature lessons and sci-fi novels / anime / fan fiction.
I don’t know much about writing, but e.g. studying computer science at universities does not seem to prevent people from creating open source software.
You can also have impact by being okay at several things. For example, I’m okay at both algorithms and UI, so I can get hired easily and then have my pick of projects (because most programmers prefer backend work). If I was also okay at writing or management, I’d probably be rich by now. I think Eliezer’s success was also due to being good at several things, not being the best at one thing.