In my mind, this is more of a communication issue and a self-knowledge issue than an “actual” problem?
Seems like it depends on what you want. When I see discussions of community here on LessWrong (and especially comparisons to dojos) I imagine the implicit claim is “and this is how we build a rationality community, which is able to get its members to accomplish their goals”. In this circumstance, you may say “well many LessWrongians actually mostly just want community for community’s sake, so they lie in the second case”.
However for those of us who do actually want a rationalist community (that is, the former case), this indeed does seem like a problem. If you offer the trade “you can add more grayspaces meaning you will build a stronger community, but get worse at accomplishing your explicit goals”, this does not seem so nice, even if there are some who would eat that trade up.
Edit: I do also think its worthwhile thinking about how this grayspaces concept could be modified to mitigate the fault Cole sees. Even in the typical case, usually people in a dojo are there for some mix of the two reasons.
When I see discussions of community here on LessWrong (and especially comparisons to dojos) I imagine the implicit claim is “and this is how we build a rationality community, which is able to get its members to accomplish their goals”. In this circumstance, you may say “well many LessWrongians actually mostly just want community for community’s sake, so they lie in the second case”.
Yeah, my biggest issue with LessWrong (and the broader rationality community generally) is that there seems to me to be much more of a focus on growth and inclusion and togetherness (at the expense of quality) than, like, Actual Standards (that would require telling some people that they’re not pulling their weight and can’t be in the garden anymore).
^ That’s an exaggeration; one person can think the boundary ought to be at level X of effort/intensity/prowess/commitment/whatever, and another person can think it ought to be at 1.5X, and it’s not immediately obvious which of these people is correct. But I at least have found that engaging with nominal “rationalists” tends to drag me down rather than helping me accelerate up/forward.
In that case I think this should have been your response to Cole.
I’d be curious to hear what particular ways you think the rationalist community is prioritizing “growth and inclusion and togetherness (at the expense of quality)”, in part because I wonder whether you aren’t just looking at the parts of the rationality community which are already actually grayspaces. Eg public meetups, or LessWrong itself. And perhaps what you may actually want is more black-belt spaces around here.
Eg if your problem is with the userbase of LessWrong in particular, I think its hard to make a public forum have the level of isolation you’d need to make the user-base black-belt quality. Especially if you don’t have any objective (or trusted subjective) measures of what “black-belt quality” means. Therefore LessWrong can only ever be a grayspace.
I don’t think this argument is exactly true – we review and reject like 30 people a day from LW and accept 1-3, and most of the time we aren’t that optimistic about the 1-3, and it’s not that crazy that we switch to the world where we’re just actually pretty selective.
(I think you are nonetheless pointing at an important thing where, when you factor in a variety of goals / resources available, it probably makes more sense to think of LessWrong as a grayspace. Although I think Duncan also thinks, if it were trying on purpose to be a grayspace, there would be more active effort guiding people towards some particular way of thinking/conversing)
Also, Duncan’s written a fair amount about this both in blogposts and comment-back-and-forths and I’m feeling a bit of sadness of “this convo feels like by default it’s going to rehash the Duncan LW Concerns 101 conversation” instead of saying something new.
Seems false for LessWrong 1.0, I’ve read much of the comments on the old Eliezer posts, and they are far far worse than the present forum. I have less to say on early LessWrong 2.0.
I know you said you don’t have a strong view on the actual history, but the actual history does seem pretty central to your argument here. You are free to choose some other argument for your conclusion however.
No—I really did just mean it as an example scenario, and you can plug in ‘x forum’ for ‘lw’. It was a mistake to use something in my example so close to the object level thing being discussed, but I think the argument ‘[forums founded by small communities] can benefit from selection effects that make them something other than gray spaces’ goes through.
I’m objecting to the ‘only ever’ in your initial statement as too strong.
Maybe SL4, Extropians, or something else is better here (like my former-favorite internet music space, llllllll.co, where all the senior members have now moved to Discord).
I think if I’d used a different example you’d be (rightly) dinging me for relevance rather than accuracy, but I really did mean to say the maybe-irrelevant thing and then reached too far for an existence proof (or not far enough, depending on how you see it).
For my two cents, I enjoy dinner parties full of rationalists where we talk about fanfiction and D&D and bread recipes. I would also enjoy having a dojo full of rationalists where I trained how to think faster and make fewer obvious mistakes. Trying to do that with roughly the same people seems likely to work just fine. Trying to do that at the same time (that is, train thinking faster while eating dinner and talking about D&D) sounds like it would not work.
My guess is I’m not disagreeing with Duncan or Garrett here.
I don’t know if I’m disagreeing with Cole. I actually have an idiosyncratic beef with MMA specifically around its very specific set of rules (at least in the MMA scenes I’ve run into) which made it worse at teaching me to fight than some other arts, though I do think MMA is probably more useful than a randomly selected more ritualistic martial art.
Probably this can be fixed by evaluations- I do like the idea of tossing a few masters from different arts into a ring and seeing what happens, I just think in practice a lot of reasonable-at-first-blush-sensible rules around how that gets done wound up distorting things.
I’m kind of skeptical of this claim because in my experience MMA schools are literally mixed and will take stuff from many different systems, and do tend to be mostly focused on winning real fights. But I would be interested in hearing where you think this falls short.
Most people who train MMA are not training to become professional fighters, so I don’t think the rules of the sport tend to effect the training system that much.
Where I trained, we did ground fighting, kickboxing style sparring, takedowns, and then eventually combined everything, and including in live sparring. Also a little weapons defense. That’s been my experience at most MMA gyms I visited, with some minor variation. I don’t know how a system can train more realistically and still be safe—like Krav Maga guys talk about going for the eyes and argue MMA isn’t hardcore enough basically, but we actually also trained that kind of thing a bit and the reality is you just can’t actually go for the eyes in a sparring match so you mostly focus on the other stuff.
Your skepticism seems warranted! I can’t tell how the rules of the sport affect the training system, and I haven’t studied MMA in depth so maybe it has more answers to problems which I just didn’t find.
the reality is you just can’t actually go for the eyes in a sparring match so you mostly focus on the other stuff.
Yeah, that’s the kind of reasonable-at-first-blush rules that come to mind. This is a tangent to the grayspaces post and if we get more than another back and forth into this I might just try to write this up as its own post. I don’t think what I’ve said above or what I’m saying in this comment should be convincing. Training to completion against a resisting opponent usually means training with moves that won’t put someone in the hospital. I don’t know that’s the wrong tradeoff, training against a resisting opponent is pretty useful.
The circumstances I needed martial arts for were mostly across a weight difference, with varying number of opponents, in an environment full of hard abrasive surfaces. There was a Brazilian Jiu Jitsu class that said it was pulling in some MMA techniques (yeah, I know calling MMA a style is a bit weird, pretty sure it was mostly Muay Thai) I tried to learn from, and it proved a trap. Later on I learned a bit more about MMA and the parts that felt like traps seemed to come from assumptions like the ground being okay to land on.
Circling back to the grayspaces-as-worrying-insulation-from-tests; if I imagine a bunch of skilled scientists and a bunch of skilled debaters talking about who has better discussion norms, I could see the scientists saying “look, we’re not going to spar with you by your rules, and we’re not going to let you come in and spar with us either unless you spend some time understanding what we think sucks about your rules.” Maybe the scientists have allowed themselves to become too insulated from someone asking the right questions or from outside perspectives. That does seem like it can happen! But also the debaters are focused on a different purpose and that could be an actual problem.
Seems like it depends on what you want. When I see discussions of community here on LessWrong (and especially comparisons to dojos) I imagine the implicit claim is “and this is how we build a rationality community, which is able to get its members to accomplish their goals”. In this circumstance, you may say “well many LessWrongians actually mostly just want community for community’s sake, so they lie in the second case”.
However for those of us who do actually want a rationalist community (that is, the former case), this indeed does seem like a problem. If you offer the trade “you can add more grayspaces meaning you will build a stronger community, but get worse at accomplishing your explicit goals”, this does not seem so nice, even if there are some who would eat that trade up.
Edit: I do also think its worthwhile thinking about how this grayspaces concept could be modified to mitigate the fault Cole sees. Even in the typical case, usually people in a dojo are there for some mix of the two reasons.
Yeah, my biggest issue with LessWrong (and the broader rationality community generally) is that there seems to me to be much more of a focus on growth and inclusion and togetherness (at the expense of quality) than, like, Actual Standards (that would require telling some people that they’re not pulling their weight and can’t be in the garden anymore).
^ That’s an exaggeration; one person can think the boundary ought to be at level X of effort/intensity/prowess/commitment/whatever, and another person can think it ought to be at 1.5X, and it’s not immediately obvious which of these people is correct. But I at least have found that engaging with nominal “rationalists” tends to drag me down rather than helping me accelerate up/forward.
In that case I think this should have been your response to Cole.
I’d be curious to hear what particular ways you think the rationalist community is prioritizing “growth and inclusion and togetherness (at the expense of quality)”, in part because I wonder whether you aren’t just looking at the parts of the rationality community which are already actually grayspaces. Eg public meetups, or LessWrong itself. And perhaps what you may actually want is more black-belt spaces around here.
Eg if your problem is with the userbase of LessWrong in particular, I think its hard to make a public forum have the level of isolation you’d need to make the user-base black-belt quality. Especially if you don’t have any objective (or trusted subjective) measures of what “black-belt quality” means. Therefore LessWrong can only ever be a grayspace.
I don’t think this argument is exactly true – we review and reject like 30 people a day from LW and accept 1-3, and most of the time we aren’t that optimistic about the 1-3, and it’s not that crazy that we switch to the world where we’re just actually pretty selective.
(I think you are nonetheless pointing at an important thing where, when you factor in a variety of goals / resources available, it probably makes more sense to think of LessWrong as a grayspace. Although I think Duncan also thinks, if it were trying on purpose to be a grayspace, there would be more active effort guiding people towards some particular way of thinking/conversing)
Also, Duncan’s written a fair amount about this both in blogposts and comment-back-and-forths and I’m feeling a bit of sadness of “this convo feels like by default it’s going to rehash the Duncan LW Concerns 101 conversation” instead of saying something new.
Some recap:
Concentration of Force
Duncan Sabien on Moderating LessWrong
Nit: selection effects can lead to a de facto black belt setting, as I think some would consider LW 1.0 or early LW 2.0.
As the paths here become broader, more numerous, it becomes a gray space.
[I don’t have a strong view on the Actual History; I just want to tell a plausible story pushing back against ‘can only ever be’.]
Seems false for LessWrong 1.0, I’ve read much of the comments on the old Eliezer posts, and they are far far worse than the present forum. I have less to say on early LessWrong 2.0.
I know you said you don’t have a strong view on the actual history, but the actual history does seem pretty central to your argument here. You are free to choose some other argument for your conclusion however.
No—I really did just mean it as an example scenario, and you can plug in ‘x forum’ for ‘lw’. It was a mistake to use something in my example so close to the object level thing being discussed, but I think the argument ‘[forums founded by small communities] can benefit from selection effects that make them something other than gray spaces’ goes through.
I’m objecting to the ‘only ever’ in your initial statement as too strong.
Maybe SL4, Extropians, or something else is better here (like my former-favorite internet music space, llllllll.co, where all the senior members have now moved to Discord).
I think if I’d used a different example you’d be (rightly) dinging me for relevance rather than accuracy, but I really did mean to say the maybe-irrelevant thing and then reached too far for an existence proof (or not far enough, depending on how you see it).
I agree that if nobody knows about your forum other than the black-belts, then your forum will be black-belt quality.
For my two cents, I enjoy dinner parties full of rationalists where we talk about fanfiction and D&D and bread recipes. I would also enjoy having a dojo full of rationalists where I trained how to think faster and make fewer obvious mistakes. Trying to do that with roughly the same people seems likely to work just fine. Trying to do that at the same time (that is, train thinking faster while eating dinner and talking about D&D) sounds like it would not work.
My guess is I’m not disagreeing with Duncan or Garrett here.
I don’t know if I’m disagreeing with Cole. I actually have an idiosyncratic beef with MMA specifically around its very specific set of rules (at least in the MMA scenes I’ve run into) which made it worse at teaching me to fight than some other arts, though I do think MMA is probably more useful than a randomly selected more ritualistic martial art.
Probably this can be fixed by evaluations- I do like the idea of tossing a few masters from different arts into a ring and seeing what happens, I just think in practice a lot of reasonable-at-first-blush-sensible rules around how that gets done wound up distorting things.
I’m kind of skeptical of this claim because in my experience MMA schools are literally mixed and will take stuff from many different systems, and do tend to be mostly focused on winning real fights. But I would be interested in hearing where you think this falls short.
Most people who train MMA are not training to become professional fighters, so I don’t think the rules of the sport tend to effect the training system that much.
Where I trained, we did ground fighting, kickboxing style sparring, takedowns, and then eventually combined everything, and including in live sparring. Also a little weapons defense. That’s been my experience at most MMA gyms I visited, with some minor variation. I don’t know how a system can train more realistically and still be safe—like Krav Maga guys talk about going for the eyes and argue MMA isn’t hardcore enough basically, but we actually also trained that kind of thing a bit and the reality is you just can’t actually go for the eyes in a sparring match so you mostly focus on the other stuff.
Your skepticism seems warranted! I can’t tell how the rules of the sport affect the training system, and I haven’t studied MMA in depth so maybe it has more answers to problems which I just didn’t find.
Yeah, that’s the kind of reasonable-at-first-blush rules that come to mind. This is a tangent to the grayspaces post and if we get more than another back and forth into this I might just try to write this up as its own post. I don’t think what I’ve said above or what I’m saying in this comment should be convincing. Training to completion against a resisting opponent usually means training with moves that won’t put someone in the hospital. I don’t know that’s the wrong tradeoff, training against a resisting opponent is pretty useful.
The circumstances I needed martial arts for were mostly across a weight difference, with varying number of opponents, in an environment full of hard abrasive surfaces. There was a Brazilian Jiu Jitsu class that said it was pulling in some MMA techniques (yeah, I know calling MMA a style is a bit weird, pretty sure it was mostly Muay Thai) I tried to learn from, and it proved a trap. Later on I learned a bit more about MMA and the parts that felt like traps seemed to come from assumptions like the ground being okay to land on.
Circling back to the grayspaces-as-worrying-insulation-from-tests; if I imagine a bunch of skilled scientists and a bunch of skilled debaters talking about who has better discussion norms, I could see the scientists saying “look, we’re not going to spar with you by your rules, and we’re not going to let you come in and spar with us either unless you spend some time understanding what we think sucks about your rules.” Maybe the scientists have allowed themselves to become too insulated from someone asking the right questions or from outside perspectives. That does seem like it can happen! But also the debaters are focused on a different purpose and that could be an actual problem.
Meh, the debate team and the scientists know they’re aiming at different things, that’s not a great example on my part.