[ Question ]

One hub, or many?

by abramdemski1 min read4th Oct 202012 comments


The SF Bay AreaCommunity

I was reading the comments to The Rationalist Community's Location Problem and especially this comment by Damien Tatum, and thinking about some classic questions of LessWrong/EA/fully-general physical community.

I think a guiding idea for years has been that if we get as many good people and organizations to have close proximity in the bay area, good things will happen. But I don't know whether that's been positive or negative overall.

For years I've heard about brain-drain problems in other would-be hubs. The best people move to the bay area, and the local hub either gets much worse or disappears. (I'm talking especially about meetup groups, but in the hypothetical where this sort of think didn't happen so much, we might also see more serious organizations in those other hubs.)

In a significant sense, this makes for a higher bar to getting involved. If what you have to do in order to get good in-person discussions with rationalists/EAs is move to the bay area rather than get to the closest decent meetup in your city/state/country, well, that puts a big damper on things.

If there have been discussions/modeling about this trade-off before, I'd especially appreciate links.

Edit: The Berkeley Community & The Rest Of Us discusses the brain-drain dynamic I'm discussing.

New Answer
Ask Related Question
New Comment

4 Answers

Was being part of the Bay area rationalist community a primary motivating factor for moving to the area for most rationalists who live there?

Maybe this community just attracts a lot of techies. Or maybe there's more AI alignment jobs there.

Personally, I've participated in relatively little in-person rationalist community. The online forum meets my needs pretty decently. But this is partly because I'm building a career in scientific research, and anticipate that I'll be around smart people wherever I go. I'm averse to moving to the Bay area for all the usual reasons: the high cost of living, fires, gender skew, and state legal system. May wind up getting sucked down there anyway.

Is the Bay area rationalist community really all that great as a social group compared to just meeting the smart nice sane people who live in every town?

I think many hubs is probably the right answer, but it depends on the goal.

If the goal is to feed people into MIRI, or generally advance some single community organization as far and as quickly as possible, then the benefits of centralization are well known and hard to overstate.  Similarly, if the goal is to have a social environment for rationalists, where one's friendships and broader social circles mostly involve other rationalists, then one hub is also clearly the best goal.

However, if the goal is to spread awareness of rationality, raise the sanity waterline generally, grow the movement, or find new ideas and fruitful applications, then many smaller hubs are a vastly better option.  All of these things are encouraged by putting rationalists in differing environments, where they contact a wider variety of people, ideas, and problems.

It seems to me that the stated goals of the community are the latter, and so the "right" answer is obvious.  We should be encouraging people to find and befriend like-minded people in their local area, and introduce the ideas to their potential rationalists.  Have many friends, rationalist and not.  Figure out how to talk about rationality's greatest insights quickly and concisely, so you can give people that ah-ha moment and get them hooked.  Etc.

To be clear, Berkeley people have a community too, and I don't think we should tell them they have an obligation to move out and evangelize, like early Christian apostles.  But I think we should stop encouraging people to move there and if someone is fed up and wants to leave the Bay, they should be bid a happy farewell.

It might be worth to think more about what we mean with the term hub. The distinguishing feature about the Bay Area isn't that it has LessWrong or SlateStarCodex meetups that you can't find anywhere else. 

People that moved to the Bay Area frequently reported afterwards that finding connection to other people in the Bay Area is a lot harder then they expected because a lot of the events where rationalist meet each other aren't open to the general public. 

The distinguishing feature of the Bay Area is that it employs plenty of people in jobs that are adjacent to the community through orgs. If a person wants to work in a community adjacent org, applying to one of those jobs makes sense and that does cause brain drain. It turn out that the kind of people who want to invest energy into community building and run meetups outside of the Bay Area are also the kind of people who would like to work in adjacent orgs. 

It's good when many cities have meetups but that has little to do with the question about hubs. 

Currently, it seems like there are a bunch of people who are looking to relocate and who would feel like it would be good to relocate to the same place as other rationalists. 

I believe there should be at least a few, as soon as the population can support it. As I recall the course of conversations, the initial challenge was achieving a critical mass of intellectual development in the first place; this is when the decision about the Bay Area was made.

I do not live in the Bay Area and do not have this complaint myself, but lately I am reading a lot of complaints about the Bay becoming an area with a lot of conformity pressure. This highlights what I think are the two primary advantages of multiple hubs:

  1. On the productivity side, multiple hubs will make it easier to pursue multiple specializations; our work still needs development in many directions simultaneously, and multiple hubs will allow more space for this to develop, as well as make it easier to capitalize on different regional specializations (for example, the Bay Area is particularly good for AI and CS; other areas might be better for other subjects like philosophy or finance or an EA focus).
  2. On the risk reduction side, consider (epistemic) disaster recovery: there are lots of reasons a given hub might no longer suit, whether it is conformist pressures, or being priced out, or economic collapse, etc. If the Bay becomes unsuitable, currently the likely outcome is everyone to scatters to the four winds. With other hubs, there would be fallback options.