Habryka suspects that the rationalist community, writ large, basically wouldn’t exist without the Berkeley hub.
In the counterfactual where MIRI had been in Oxford this whole time, or wherever, I think there still would have been Bay Area, NYC, Boston, and Seattle communities. Maybe the Bay Area community would have been centered in SF instead, or something. Basically, I agree with your footnote, but some additional details:
It’s less likely that we would have had the sense that there was one ‘in-person primary hub’, but I’m less convinced that the counterfactual on mission impact is huge here. Like, perhaps each of those communities would be better-developed and healthier without the significant brain drain? Maybe people would have still switched jobs between orgs, but also switched cities, so many more people would have lived in Boston and Seattle and the Bay Area, instead of lots of people who have lived in one other city and the Bay Area?
I think the rationality community is in a weird place where we get significant returns to concentration when considering getting work done (i.e. it’d be great to have every EA/rationalist org in the same spot) and also significant returns to dispersion when considering recruitment (having a healthy EA/rationalist community near every major university and in every tech hub seems like it would be great, and having the orgs communicating in the open with durable artifacts seems like it would be great for bringing people up to speed).
Habryka points out that it’s very unlikely that he would have started LessWrong 2.0 if he hadn’t known Vaniver socially. Both Habryka and Vaniver(?) initially moved to the Bay because that’s where MIRI/CFAR was.
I laid the groundwork for reviving LW while living in Austin. [I suspect this was a core component of me still caring strongly about LW at the time, instead of going “yeah it’s sad, but what are you going to do?” while going to in-person events.] In-person interactions were still an important part of it, but they happened at CFAR alumni reunions and Solstices, which I flew to.
Oli deciding to work on the project solidified at a… 80k New Year’s house party, I think? I am pretty sure that was end of 2016-start of 2017, and I had just moved to the Bay Area in October 2016 to work for MIRI (and wouldn’t have moved to the Bay except to work at an org like MIRI). That was made much easier by both of us living in the same city, and especially the same city as lots of other stakeholders (which made it low-cost to meet with them), but I’m not sure about the counterfactual of them being video calls. [Like, my sense is it’s harder to schedule meetings / stay on people’s radar without incidental contact, but our meetings with Eliezer were scheduled mostly without incidental contact, I think.]
Sorry for the US-centric viewpoint; I know it’s hard to immigrate here, it’s just that so many of us are already here.
This also feels like a hard-to-evaluate counterfactual. I have a sense that “most of the good people are already in the US or could get here”, but I don’t have a great sense of how much that’s filtered by only having a good read of the people who are already in the US. If you take the LW core team as an example, I think 2 out of 6 are Americans, and the rest are from various parts of the world here on visas; I don’t know whether to say “this is evidence that it’s not that hard to get people into the US” or whether to say “in the world where MIRI had always been in Oxford, it would have been even easier for everyone in that reference class to congregate in Oxford, and we would have even more promising people to work on LW.”
I think, all things considered, I have over 50% probability that MIRI’s decision will be to stay in the Bay, and for more than three quarters of ‘in-person Bay Area community members’ to stay in or return to the Bay. [I don’t work for MIRI at the moment, doing nondisclosed-by-default research made possible by the concentration of people in the Bay, and so even if MIRI moves it’s not obvious that I’d follow in less than a year or two.]
The thing I’m most worried about, tho, is something like “the Bay Area growing crazier and more hostile month over month, and us having squandered our obvious chance to do a coordinated move, in ways that make it harder to coordinate future moves.” If the ‘in-person community’ has survived COVID quarantines and people moving away temporarily (which I think it mostly has, but maybe other people’s sense of this is very different?), it seems likely to me that it would also survive MIRI moving to Boston (or wherever) and lots of other orgs staying in the Bay. [Like, OpenAI isn’t going to move to Boston, OpenPhil probably won’t, and so on.]
Maybe I shouldn’t be very worried about this? 80k left the Bay, after all, and seems to be doing well, and landed in another EA/rationalist hub. If the temperature in the Bay gradually ramps up, maybe at some future point AI Impacts leaves, and then the LW team leaves, and then MIRI leaves, or whatever, and the existence of secondary hubs means this is a more gradual transition than it might seem.
In the counterfactual where MIRI had been in Oxford this whole time, or wherever, I think there still would have been Bay Area, NYC, Boston, and Seattle communities. Maybe the Bay Area community would have been centered in SF instead, or something. Basically, I agree with your footnote, but some additional details:
It’s less likely that we would have had the sense that there was one ‘in-person primary hub’, but I’m less convinced that the counterfactual on mission impact is huge here. Like, perhaps each of those communities would be better-developed and healthier without the significant brain drain? Maybe people would have still switched jobs between orgs, but also switched cities, so many more people would have lived in Boston and Seattle and the Bay Area, instead of lots of people who have lived in one other city and the Bay Area?
I think the rationality community is in a weird place where we get significant returns to concentration when considering getting work done (i.e. it’d be great to have every EA/rationalist org in the same spot) and also significant returns to dispersion when considering recruitment (having a healthy EA/rationalist community near every major university and in every tech hub seems like it would be great, and having the orgs communicating in the open with durable artifacts seems like it would be great for bringing people up to speed).
I laid the groundwork for reviving LW while living in Austin. [I suspect this was a core component of me still caring strongly about LW at the time, instead of going “yeah it’s sad, but what are you going to do?” while going to in-person events.] In-person interactions were still an important part of it, but they happened at CFAR alumni reunions and Solstices, which I flew to.
Oli deciding to work on the project solidified at a… 80k New Year’s house party, I think? I am pretty sure that was end of 2016-start of 2017, and I had just moved to the Bay Area in October 2016 to work for MIRI (and wouldn’t have moved to the Bay except to work at an org like MIRI). That was made much easier by both of us living in the same city, and especially the same city as lots of other stakeholders (which made it low-cost to meet with them), but I’m not sure about the counterfactual of them being video calls. [Like, my sense is it’s harder to schedule meetings / stay on people’s radar without incidental contact, but our meetings with Eliezer were scheduled mostly without incidental contact, I think.]
This also feels like a hard-to-evaluate counterfactual. I have a sense that “most of the good people are already in the US or could get here”, but I don’t have a great sense of how much that’s filtered by only having a good read of the people who are already in the US. If you take the LW core team as an example, I think 2 out of 6 are Americans, and the rest are from various parts of the world here on visas; I don’t know whether to say “this is evidence that it’s not that hard to get people into the US” or whether to say “in the world where MIRI had always been in Oxford, it would have been even easier for everyone in that reference class to congregate in Oxford, and we would have even more promising people to work on LW.”
I think, all things considered, I have over 50% probability that MIRI’s decision will be to stay in the Bay, and for more than three quarters of ‘in-person Bay Area community members’ to stay in or return to the Bay. [I don’t work for MIRI at the moment, doing nondisclosed-by-default research made possible by the concentration of people in the Bay, and so even if MIRI moves it’s not obvious that I’d follow in less than a year or two.]
The thing I’m most worried about, tho, is something like “the Bay Area growing crazier and more hostile month over month, and us having squandered our obvious chance to do a coordinated move, in ways that make it harder to coordinate future moves.” If the ‘in-person community’ has survived COVID quarantines and people moving away temporarily (which I think it mostly has, but maybe other people’s sense of this is very different?), it seems likely to me that it would also survive MIRI moving to Boston (or wherever) and lots of other orgs staying in the Bay. [Like, OpenAI isn’t going to move to Boston, OpenPhil probably won’t, and so on.]
Maybe I shouldn’t be very worried about this? 80k left the Bay, after all, and seems to be doing well, and landed in another EA/rationalist hub. If the temperature in the Bay gradually ramps up, maybe at some future point AI Impacts leaves, and then the LW team leaves, and then MIRI leaves, or whatever, and the existence of secondary hubs means this is a more gradual transition than it might seem.