Thoughts on the REACH Patreon

I moved from NY to the Bay sev­eral months ago.

In many senses, the Berkeley com­mu­nity is much big­ger than NYC. There’s a few hun­dred mem­bers in­stead of around 30. I had sev­eral friends in the Bay be­fore mov­ing here, and have made more since ar­riv­ing.

But, it didn’t ac­tu­ally feel like home to me un­til a cou­ple weeks ago, at one of the weekly mee­tups at the new com­mu­nity cen­ter.

Th­ese are my thoughts about the REACH (Ra­tion­al­ity and Effec­tive Altru­ism Com­mu­nity Hub) space, and what con­sid­er­a­tions I think are rele­vant for fund­ing it. It has a Pa­treon which is cur­rently hit­ting the “ju­u­ust enough money that it might pos­si­bly work so long as other things go right” thresh­old. But it doesn’t re­ally have enough fund­ing to re­li­ably break even, let alone thrive.

tldr: I think there are lot of rea­sons you might want to help fund REACH, both from an effec­tive al­tru­ism per­spec­tive, and from a “just buy some nice things for your­self” per­spec­tive.


Epistemic Sta­tus: I’m pretty bi­ased in fa­vor of REACH, and much of my rea­son­ing was at least a bit mo­ti­vated. I ex­pect to make a lot of use of the in­fras­truc­ture there, so my thoughts here are a bit self-serv­ing.

I’ve at­tempted to ac­count for this in my writ­ing of this, and am still fairly con­fi­dent that there’s some­thing im­por­tant go­ing on here.


Table of Contents

  1. Ray’s Opinionated Con­cep­tion of Meetups

  2. Me­chan­ics of Berkeley Meetup Brain Drain

  3. En­ter REACH

  4. Frame­works of Funding

    1. Buy­ing Nice Things vs Effec­tive Altruism

    2. Nice Things and Homemade Prices

    3. Okay, but is this the right nice thing?

    4. The Case for Impact

  5. Mea­sur­ing Intangibles

    1. Agency Ladder

    2. Water Cool­ers and Campuses

  6. Match­ing Funds

  7. In Clos­ing...

Ray’s Opinionated Con­cep­tion of Meetups

In NYC, meetup means “some­one runs an hour long pre­sen­ta­tion, mod­er­ated dis­cus­sion or work­shop.” Every week, you can show up to a pub­lic meetup. There will be con­tent to learn or prac­tice, and peo­ple you can hang­out with. It’s struc­tured such that if you’re shy, or if not as many good con­ver­sa­tion­al­ists show up that week, or if you just pre­fer struc­ture, there’s some­thing to learn and en­gage you.

Mean­while, if you care more about hang­ing out with the reg­u­lars than listen­ing to a talk, you’re free to just ar­rive later in the evening.

More­over, there’s a pal­pable sense that this is a com­mu­nity, more than a cir­cle of friends. There’s a so­cial en­tity greater than the sum of its parts, and there’s a way for new­com­ers to get in­volved.

I think the nearby San Fran­cisco and South Bay Com­mu­ni­ties have mee­tups closer to that for­mat (I haven’t been yet), but in Berkeley, mee­tups gen­er­ally take the form of freeform so­cial­iza­tion.

And on one hand, in NYC the freeform so­cial­iz­ing is very much the point. Hang­ing out and mak­ing friends is the valuable part, moreso than the pre­sen­ta­tion or work­shop. But the pre­sen­ta­tion/​work­shop is what gives that con­ver­sa­tion en­ergy and a sense of cul­ture/​pur­pose.

In Berkeley, there’s a weird com­bi­na­tion of cir­cum­stances where there’s lots of con­tent-driven so­cial events hap­pen­ing, but much of that con­tent is hap­pen­ing in hard-to-find silos. Oc­ca­sional big tent­pole events like Sols­tice or EA Global hap­pen, you might come to Berkeley at that time and get ex­cited but then a few weeks later you’re look­ing around and think­ing “okay… what next…?”.

NY LessWrong is a small com­mu­nity. Berkeley Ra­tion­al­ity/​EA is like a village. You’ll walk to the deli and run into a ra­tio­nal­ist on the way. There’s roughly a dun­bar num­ber of peo­ple in­ter­act­ing, and if you get your­self so­cially net­worked in it feels quite thriv­ing. But if you’re not net­worked in, you have this weird sense that some­thing is hap­pen­ing but you can’t tell where.

If you’re on out­side, this may look like weird so­cial games and pop­u­lar­ity con­tests. Not gonna lie – I think it’s at least a bit of that. But I think a lot of it is just be­ing stuck in a par­tic­u­lar equil­ibrium, and if we all co­or­di­nated to build some in­fras­truc­ture (phys­i­cal and so­cial) we could get to a much bet­ter one.

Me­chan­ics of Berkeley Meetup Brain Drain

Com­pet­ing Opportunities

In NYC, if you’re an agenty ra­tio­nal­ist who wants to con­tribute to the com­mu­nity, there’s only one ob­vi­ous place to put your en­ergy: the meetup it­self, much of which is pub­lic fac­ing.

In Berkeley, there’s a host of or­ga­ni­za­tions you can vol­un­teer at, fo­cused on ra­tio­nal­ity train­ing, x-risk, effec­tive al­tru­ism re­search, star­tups, and all kinds of small-to-medium pro­jects.

Or­ga­niz­ing mee­tups is hard, skil­led la­bor. If you’re the sort of per­son who’s will­ing to put in the effort, there’s a long list of com­pet­ing things, each of which make a cred­ible case for be­ing im­por­tant and re­ward­ing.

On the flip­side: if you’ve just ar­rived and you’re not so­cially con­nected, it may take longer to find those com­pet­ing op­por­tu­ni­ties. The pub­lic-fac­ing com­mu­nity dearth is an ob­vi­ous thing to fo­cus on. But if you do, you’ll likely find in the pro­cess of do­ing so that you’ll gain con­fi­dence and skills, and you’ll meet peo­ple work­ing on other pro­jects.

Soon af­ter, you may have a cluster of friends who you know well, who are in­ter­ested in the things you’re in­ter­ested in. I know a cou­ple peo­ple who briefly tried or­ga­niz­ing mee­tups and then found there were other things that felt more ex­cit­ing to them.

Public mee­tups tend to be a grab bag of peo­ple with vary­ing in­ter­ests. Even just show­ing up once a week is some­thing that com­petes with hav­ing con­ver­sa­tions with room­mates and friends and cowork­ers, whose in­ter­ests may more di­rectly re­late to your own.

Mo­ti­va­tions and So­cial Pressures

It seems like the only peo­ple who fo­cus on it are those in­trin­si­cally mo­ti­vated to wel­come new­com­ers into the fold. And not only is that rare, but var­i­ous things sub­tly pun­ish that. The so­cial fabric re­wards peo­ple who can tell an ex­cit­ing story about how what they’re do­ing is Sav­ing The World some­how, and “I’m mak­ing this a nice home to help new­com­ers” is a bit harder to do that with.

Mean­while the ra­tio­nal­ity com­mu­nity at­tracts the sort of per­son who is… well… slightly so­cially and en­vi­ron­men­tally oblivi­ous some­times, which makes it less in­trin­si­cally re­ward­ing to provide a com­mu­nity hub. The peo­ple who show up are smart and friendly but per­haps less likely to help take out the trash, which can add an­other layer of frus­tra­tion.

So, year by year, you have a com­mu­nity of 300 peo­ple that some­how has no high qual­ity pub­lic-fac­ing way to get in­volved.

I think this is a prob­lem worth co­or­di­nat­ing on, but it’s a prob­lem that has to ac­tu­ally be solved while re­al­is­ti­cally ac­cept­ing the peo­ple’s in­cen­tives and goals.

Right now we have a rare con­fluence of per­sonal mo­ti­va­tion, real es­tate and… for lack of a bet­ter word “com­mu­nity magic”, that makes me feel more op­ti­mistic that this prob­lem is solv­able.

En­ter REACH

For the past year or so, Sarah has been ex­plor­ing op­tions for a com­mu­nity cen­ter in Berkeley.

This is some­thing I’ve heard var­i­ous peo­ple talk about for while. Some group houses as­pire to be com­mu­nity cen­ter-es­que things, but run into is­sues like “the peo­ple in the house ac­tu­ally want to live their lives which some­times means they don’t feel like host­ing trav­el­ers or mee­tups.”

The CFAR office has made some effort to be this. But it’s not re­ally op­ti­miz­ing for it, and the build­ing’s se­cu­rity sys­tem cre­ates a bizarre set of triv­ial in­con­ve­niences you have to over­come to get in. (literal bar­ri­ers to en­try :P)

(This seems like it shouldn’t mat­ter, but to­tally mat­ters). It’s also a bit out of the way if you live in south­ern Berkeley.

The prob­lem is, com­mu­nity cen­ters re­quire lots of startup cap­i­tal, and it’s a vague en­deavor with fuzzy benefits that aren’t ob­vi­ously worth $60,000+ a year.

But Sarah said “this seems im­por­tant enough to just do it,” spent a bunch of her own money rent­ing out a cafe near an ex­ist­ing ra­tio­nal­ist house, and hoped the pro­ject would prove it’s value. The Ra­tion­al­ity and Effec­tive Altru­ism Com­mu­nity Hub (REACH) was born.

There’s a list of offi­cial things REACH is aiming to provide a space for:

  • Weekly meetups

  • Var­i­ous events that range from “so­cial/​com­mu­nity” to “in­tel­lec­tual/​growth”

  • Coworking

  • Cheap beds for out-of-town ra­tio­nal­ist/​ea trav­el­ers vis­it­ing berkeley

  • Classes or ac­tivi­ties for kids to help the grow­ing num­ber of par­ents in the community

  • Space and sup­port for helping peo­ple in the com­mu­nity brain­storm about and start new endeavors

On pa­per, these look po­ten­tially valuable, but I think a rea­son­able per­son might be skep­ti­cal you can ac­tu­ally achieve them.

I’m ex­cited about REACH be­cause I’ve been to the weekly mee­tups, and… got a deep sense that the place felt like home.

I ad­mit I’m a bit bi­ased here – there was a bunch of nos­tal­gia feed­ing into my ex­pe­rience and I’m not sure how uni­ver­sal it is. But… fi­nally, there was an ac­tual god damn meetup – a work­shop run by a com­mu­nity mem­ber (we were prac­tic­ing Gendlin’s Fo­cus­ing), a dozen peo­ple helping each other learn a tech­nique and nav­i­gat­ing our in­ter­nal emo­tional blocks. And af­ter­wards, freeform dis­cus­sion, peo­ple bounc­ing ideas around.

A cou­ple weeks later I went to a sec­ond meetup, just as vibrant, twice as many peo­ple, run by a differ­ent com­mu­nity mem­ber. And I got a strong sense that there’s a sur­plus of pent-up or­ga­ni­za­tional en­ergy in Berkeley – lots of peo­ple who would to­tally step up to run events if they were given the af­for­dance to.

Mean­while, per­haps most im­por­tantly, were peo­ple in the af­ter­math bounc­ing ideas around, dis­cussing their pro­jects, think­ing about how to col­lab­o­rate.

I think REACH has the po­ten­tial to solve the Public Meetup Brain Drain prob­lem, via:

  • Be­ing a ded­i­cated space for it, lo­cated within walk­ing dis­tance of a large num­ber of ra­tio­nal­ists, and a cou­ple blocks from the Ashby BART sta­tion.

  • Hav­ing a ded­i­cated per­son work­ing at least half-time on main­tain­ing the in­fras­truc­ture (both phys­i­cal, or­ga­ni­za­tional and so­cial) for mee­tups and other events to happen

  • Since the in­fras­truc­ture is already there, it’s much eas­ier for new­com­ers to get in­volved, run a cou­ple mee­tups, maybe help im­prove some of the un­der­ly­ing in­fras­truc­ture, and then (most likely) end up mov­ing on as they get in­volved with other or­ga­ni­za­tions in the Bay.

  • Mean­while, old-timers with ex­per­i­men­tal ideas have a venue to try them out, in a way that con­tributes to the pub­lic-fac­ing com­mons.

Frame­works of Funding

Rent­ing the REACH space is fairly ex­pen­sive. $60k+ is noth­ing to sneeze at. Should it be funded? Who should fund it?

I think this is a le­gi­t­i­mate ques­tion. What are rea­son­able frame­works for de­cid­ing whether and how to fund a com­mu­nity space? I’m bi­ased in fa­vor of the REACH cen­ter, but think it’s im­por­tant to get this ques­tion right.

Ben Hoff­man noted the haz­ards of forc­ing peo­ple to sell a nar­ra­tive around im­pact in or­der to get funded. It forces peo­ple to lie, or warp their vi­sion to satisfy a fun­der’s goals (or worse, warp it into an abom­i­na­tion try­ing to satisfy mul­ti­ple fun­ders goals, and in all like­li­hood failing to satisfy any­one).

Ben also ar­gued that if you ad­vo­cate a strat­egy that in­cludes re­cruit­ing peo­ple to the Bay, you have an obli­ga­tion to ac­tu­ally take care of those peo­ple once they get here, and part of that in­cludes giv­ing hu­mans a space to be.

I think both these ar­gu­ments are im­por­tant. But aren’t quite the frame I’d ap­proach this with.

Peo­ple spend money for differ­ent rea­sons. I de­cided to frame this post as ad­vice I’d give to al­ter­nate ver­sions of my­self. Peo­ple who shared my gen­eral val­ues, but might vary along axes like:

How much money do you have?

How well do you get along with the so­cial clusters that are most in­volved with REACH? (al­though note that differ­ent events and cowork­ing days-of-week tend to at­tract differ­ent peo­ple)

How of­ten do you visit Berkeley?

How close do you live to REACH?

Would you benefit from a cowork­ing en­vi­ron­ment?

Effec­tive Altru­ism vs Pay­ing for Nice Things

There’s ba­si­cally two rea­sons I’d per­son­ally give peo­ple money:

One rea­son is effec­tive al­tru­ism – helping peo­ple as much as I can with the re­sources available to me (this can in­clude in­vest­ing in my own fu­ture ca­pa­bil­ities, or fund­ing pro­jects that seem like they will have good im­pacts even if it’s not nec­es­sar­ily their mis­sion state­ment).

That’s great and all. But, much more com­mon than al­tru­ism is that I like hav­ing nice things.

Some­times I just want to go see a movie. Or have a nice place to live. Or art sup­plies or video games. Get­ting these things in­volves pay­ing peo­ple.

Notably, hav­ing Nice Things in­cludes buy­ing things for my friends be­cause they make me happy. It also in­cludes en­gag­ing in pos­i­tive sum trades, and one box­ing in New­comb-like-prob­lems so that peo­ple will re­li­ably model me as the sort of per­son they can trust with good op­por­tu­ni­ties.

Both EA and NiceThings per­spec­tives some­times in­volve sym­bolic value – do­ing small to­ken things to re­mind peo­ple (or my­self!) what I care about. If I’m do­ing it right (which I don’t always), the to­ken is ob­vi­ously a to­ken (i.e. not de­ceiv­ing my­self or oth­ers), but just as ob­vi­ously rep­re­sen­ta­tive of some­thing real that will pay off later.

If I’m poor right now, I might donate small bits of money, to re­mind my­self that I’m the sort of per­son who will put his money where his mouth is, so that later on when I can af­ford it, I’m in the habit of ac­tu­ally do­ing that.

I think there’s an Effec­tive Altru­ist case to be made for REACH. (I think you can make that case with­out con­tort­ing peo­ple into weird com­pro­mises over their vi­sion). But be­fore we get into the realm of EA, let’s just ask straight­for­wardly:

Does a com­mu­nity cen­ter sound like some­thing you’d benefit from?

This is not a rhetor­i­cal ques­tion. Maybe the an­swer is no. But if you would benefit from mee­tups, or cowork­ing, or pe­ri­od­i­cally get crash space in Berkeley for cheap… maybe you and oth­ers should just pay for that be­cause it’s nice and you’d benefit from it.

This’d be harder (though not im­pos­si­ble) in most cities – even other places in the Bay, be­cause there are only so many ra­tio­nal­ists and com­mu­nity cen­ters are ex­pen­sive. But one of the uniquely promis­ing things about REACH is that it’s lo­cated within walk­ing dis­tance of a village-worth of ra­tio­nal­ists and EA folk.

If you find your­self in a village, it’s quite rea­son­able to ask your­self “do I want my village to be a place with nice things?”. What sort of nice things would you like? How much are you will­ing to pay for them?

Nice Things and Homemade Prices

Quoth Zvi:

What is the price of nice things? The first price, at­ten­tion to de­tail. The quest for nice things is a sa­cred quest. Creation of them, even more so. It re­quires effort, fo­cus, sac­ri­fice. You have to care.
Other­wise, we can’t have nice things. Be­cause you didn’t make them.

Quoth me, in more de­tail, in Melt­ing Gold:

It costs more to build some­thing your­self than to buy it fac­tory made. Things you make your­self are of­ten able to be more unique and spe­cial than things mass-pro­duced by cap­i­tal­ism. They can cater to niche in­ter­ests with­out enough de­mand to de­velop mass pro­duc­tion.
“Homemade” may trig­ger bad as­so­ci­a­tions, be­cause there was a weird fol­lowup step where Cap­i­tal­ism no­ticed that peo­ple had no­ticed that home­made things took more time and were worth more. And en­ter­pris­ing en­trepreneurs saw free money and learned to slap a “home­made” la­bel on prod­ucts for a quick buck.
Is an ar­ti­sanal hand-crafted coffee mug re­ally worth more than a mass pro­duced ver­sion on Ama­zon?
But… when the home­made thing is unique, when you liter­ally can’t get it any­where else, and you are get­ting im­por­tant so­cial or cul­tural value from it… then… well, if you want that thing, the only way to get it is to pay home­made prices for it.
You may not be able to pay for them with money. They are usu­ally labors of love. If there was enough de­mand for them for some­one to do them full-time, you’d prob­a­bly be able to mass pro­duce them more cheaply any­way.
It’s un­likely the peo­ple mak­ing them could ac­tu­ally more eas­ily pro­duce them if they were paid more. Or, the amount of money would be dra­mat­i­cally more than what seems ob­vi­ous. It’s not enough to cover costs. It has to be enough to quit your day job, and then not worry about quit­ting your day job turn­ing out to be a hor­rible idea.
This means if you want to pay for a rare, pre­cious thing that you want to keep ex­ist­ing, it is quite likely that the only ways to guaran­tee its con­tinued ex­is­tence is to put in sweat and sac­ri­fice. If things are well or­ga­nized it shouldn’t need to be a ma­jor sac­ri­fice, but it may mean se­ri­ous time and at­ten­tion that you were spend­ing on other things you cared about too.
I don’t mean to say any of this in a mor­al­iz­ing way. This is not an es­say about what you “should” do. This is just a de­scrip­tion of what is in fact nec­es­sary for cer­tain things to hap­pen, if they are things that mat­ter to you.

Or, some­times, you just plain need both a lot of sweat and effort and literal dol­lars.

Lo­cal Charity

There’s a say­ing: “Char­ity starts at home.” Most peo­ple’s de­fault ori­en­ta­tion to al­tru­ism is “find nice pro­jects nearby that make me feel good and donate there.”

I don’t think this mode of al­tru­ism is wrong. But I think it’s been warped by the rise of moder­nity. A mod­ern city has mil­lions of peo­ple nearby, and you don’t ac­tu­ally have more con­nec­tion to them than you do to drown­ing strangers in Africa or far-fu­ture-civ­i­liza­tions. Whereas in The Be­fore Times, the im­pulse to help peo­ple around you di­rectly led to a world where you had more op­por­tu­ni­ties, were more trusted or re­spected by your peers, and felt more fulfilled.

Naive ap­pli­ca­tion of tra­di­tional char­ity re­sults in a worst-of-both-wor­lds, where you aren’t re­ally get­ting much out of it, and you aren’t helping peo­ple very well.

I think it’s very good that the ra­tio­nal/​EA-sphere spends at­ten­tion on helping far away or fu­ture peo­ple, even if they won’t re­turn the fa­vor. Part of liv­ing in the pre­sent era can and should in­clude notic­ing that you have a lot of power, and the op­por­tu­nity to use that power to help peo­ple at scale.

But, while money is the unit of car­ing, there’s plenty of things to care about other than far away peo­ple.

Free­thinker-es­que com­mu­ni­ties don’t just have trou­ble co­op­er­at­ing for grand al­tru­is­tic pro­jects. They strug­gle to co­op­er­ate to just buy them­selves some god damn nice things.

I’m of the opinion peo­ple do not spend nearly enough money in­vest­ing in their own com­mu­ni­ties.

How much is a com­mu­nity cen­ter worth?

If I didn’t have a job, or were oth­er­wise strug­gling, I wouldn’t donate more than a to­ken amount. I think it is far more im­por­tant to get your­self to a po­si­tion of abun­dance and strength, so that you can help peo­ple for real.

If I had money but didn’t live close to REACH, I’d prob­a­bly di­rect my Nice Thing bud­get to­wards more lo­cal events. (Oddly enough, I sus­pect it makes more sense to fund REACH if you live in South Bay than in San Fran­cisco, since in South Bay you get the benefit of re­li­able crash space when you visit).

But given that I live nearby, and I’m in a fi­nan­cial state where I’d pay $15 for oc­ca­sional movies, din­ners or out­ings, it seems to me that the lower bound for the value of a good meetup is some­thing like $10.

If 20 peo­ple are com­ing to a meetup each week, and 15 of them can af­ford small lux­u­ries, it seems like a rea­son­able lower bound on the meetup’s value is (15 peo­ple x 4 weeks x $10 = $600/​month), with each per­son chip­ping in $40.

If you’re get­ting value from con­nec­tions, op­por­tu­ni­ties, and fulfill­ment, then I think a good meetup can eas­ily be worth more than a $15 movie.

The func­tion of “how valuable is it?” might not line up with “how much can I re­al­is­ti­cally pay?”. But this brings up an ad­di­tional ques­tion:

If you’re some­one who did get a lot of value from mee­tups (when you were new to a city, didn’t have a job and were strug­gling to make ends meet) but now are a pro­gram­mer mak­ing a 6 figure salary… I think it is quite rea­son­able to put in some­thing ex­tra from a “pay it for­ward” stand­point, even if by now you’re spend­ing more of your time with par­tic­u­lar friends rather than the pub­lic-fac­ing mee­tups.

Check that it’s worth it

There’s an im­por­tant counter-view­point here:

If you’re not get­ting value out of mee­tups – if you showed up and hoped some magic would hap­pen but it se­ri­ously just didn’t – then please do not let your­self feel pres­sured to give out of a vague pro-so­cial-guilt. I am ar­gu­ing about why you should be will­ing to pay for nice things for your­self, not nice things for other peo­ple.

Maybe you don’t get along with the or­ga­niz­ers. Maybe the sort of peo­ple a space at­tracts aren’t that in­ter­est­ing to you.

Only you know what things are ac­tu­ally nice.

Dangers of overcommitment

Fun fact: I ini­tially pledged $100/​month to REACH. Then I walked that pledge back to $50.

REACH is eas­ily worth $100/​month to me. But. There’s a bunch of other things in the same refer­ence class that are also worth­while.

Sum­mer Sols­tice, EA Global, the CFAR Re­u­nion and other ma­jor events are com­ing up. Th­ese events tend to barely scrape by with fund­ing (and of­ten, the or­ga­niz­ers just lose thou­sands of dol­lars). I want to make sure I can af­ford to sup­port a healthy, di­verse so­cial land­scape.

There are also other mee­tups nearby, which you might want to sup­port do­ing more am­bi­tious pro­jects.

I want to en­gage in fundrais­ing strate­gies that will work for the long term, al­low­ing other com­mu­nity mem­bers to get new pro­jects funded – next year’s ver­sion of some­thing-weird-the-way-Sols­tice-2011-was-weird.

Part of this means be­ing care­ful with the pub­lic re­source of “dis­course around fundrais­ing”, and not get­ting peo­ple to over­com­mit so that they can’t af­ford the next as­piring pro­ject.

(Also, since I spend a lot of time at REACH, I wanted to re­serve some money to just spon­ta­neously buy spe­cific nice things for it that I par­tic­u­larly want, like the coffee table I just or­dered. Spon­ta­neously buy­ing pre­sents for my­self and my friends is nice. See Robby Bens­inger’s Chaos Altru­ism.)

Bring­ing the Party

One source of value that REACH brings is mak­ing it salient that you can hold more events, or bring your own value to the weekly meetup.

Of­ten­times, I want to try some­thing out – a new ex­er­cise, a weird event, a new way of look­ing at the world that I think is in­ter­est­ing but want to san­ity check. Some­times these seem valuable from an EA lens (more on that later), and some­times they’re just weird and fun.

Hav­ing an ex­ist­ing com­mu­nity in­fras­truc­ture dras­ti­cally low­ers the ac­ti­va­tion-en­ergy needed to try some­thing like this out.

Okay, but is this the right Nice Thing?

I’m highly con­fi­dent that we should be will­ing to spend a bunch on com­mu­nity in­fras­truc­ture. I think it’s rea­son­able to de­bate whether this is the right Nice Thing to spend a bunch on, within the Nice Things paradigm.

[Note: it’s a bit fraught to ask “is this the best thing to spend money on?” That way leads end­less de­ci­sion paral­y­sis. But be­fore we con­verge on “spend a bunch on this par­tic­u­lar thing” it seems rea­son­able to look at some al­ter­na­tives. Is this at least a rea­son­able con­tender for ex­pen­sive thing to co­or­di­nate on fund­ing?]

Is the phys­i­cal space im­por­tant? Is it more im­por­tant than hav­ing a ful­l­time per­son do­ing the meta-or­ga­niz­ing? Is Sarah the right per­son for the job? Is the cur­rent ap­proach the best ap­proach? Is the cur­rent venue the best venue?

In­stead of a venue, you could hire two full-time peo­ple who didn’t man­age a space but who did co­or­di­nate pub­lic events (po­ten­tially in differ­ent cities).

Ben Hoff­man asks that we take care of the peo­ple we re­cruit. From this per­spec­tive, is a com­mu­nity-cen­ter shaped thing more im­por­tant than, say, hiring a full time ther­a­pist, me­di­a­tor, Berkeley-bu­reau­cracy-spe­cial­ist, other sup­port role that the com­mu­nity could use filling?

I think that ques­tion is im­por­tant, and I’d tie that back to the “don’t over­com­mit your re­sources” is­sue. It seems ob­vi­ous to me that some­thing com­mu­nity-cen­ter-shaped would be valuable, and just as ob­vi­ous that other things would be valuable. My sense is that most of the al­ter­na­tives to a com­mu­nity-cen­ter-shaped-thing are harder to ex­e­cute.

I’m not sure how to think about that. Mean­while I’d say some­thing like: If you think you’d com­mit re­sources to other com­mu­nity-fo­cused pro­jects con­di­tional on them look­ing vi­able, maybe...

  1. If mak­ing re­cur­ring dona­tions (a la Pa­treon), maybe donate in pro­por­tion to how much you’d want to sup­port the com­mu­nity cen­ter in par­tic­u­lar, if you were also donat­ing to other com­mu­nity-fo­cused pro­jects that you con­sid­ered higher pri­or­ity.

  2. Mean­while, if you have ad­di­tional in­come now that you’d like to spend on com­mu­nity in­fras­truc­ture (but no cur­rent pro­jects to give it to), maybe make an ad­di­tional one-time dona­tion.

    Since REACH is still in the “get­ting set up” stage, there’s a lot of things they could use an in­flux of cash for, like “get a real shower/​bath set up”.

My over­all thoughts here:

A. There’s been vague ges­tur­ing in the di­rec­tion of REACH for years, with no one ac­tu­ally mak­ing it hap­pen. If you’re wor­ried about the ex­act ex­e­cu­tion, I think it makes more sense to frame ob­jec­tions in the form of “how do we take the mo­men­tum we cur­rently have and op­ti­mize it?” rather than “should we even have this mo­men­tum right now?”

The ra­tio­nal­sphere has hit a point where I think agency and good ideas are more pre­cious than money, and I’m wary of ac­ci­den­tally kil­ling that mo­men­tum.

B. Ul­ti­mately, my rea­son for sup­port­ing the idea ba­si­cally as-is is be­cause it’s work­ing, and I don’t want to mess with it too much.

C. Re: the “if you want to take care of peo­ple, is this the best way?” ques­tion: There are a bunch of other roles and in­fras­truc­ture that the lo­cal com­mu­nity could use, which I’d tie back to my “dan­ger of over­com­mit­ment” point in the pre­vi­ous sec­tion.

I also think hav­ing a cen­tral­ized com­mu­nity cen­ter will be make those things eas­ier to co­or­di­nate on. (I think this is also an ex­plicit part of Sarah’s plan, al­though I think it’s good not to over­promise on a new ini­ti­a­tive SOLVING ALL THE PROBLEMS, and fo­cus ini­tially on just do­ing one thing well).

D. It seems ex­tremely rare (not just among ra­tio­nal­ists but gen­er­ally) to have the op­por­tu­nity to build a good village. The cen­tral Berkeley Ra­tion­al­ist/​EA world pretty closely re­sem­bles the sort of com­mu­nity-shaped-hole that I think much of mod­ern Amer­ica has been miss­ing.

There’s around 50 peo­ple I know and (I think?) an­other 50 that I don’t who live within walk­ing dis­tance of each other. Hav­ing come so close, sort of by-ac­ci­dent, I re­ally want to see what a ra­tio­nal­ist village can do when given the re­sources nec­es­sary to thrive.

In­fras­truc­ture seems valuable. In-per­son com­mu­nity spaces seem valuable for peo­ple to thrive.

Epistemic Sta­tus: Spec­u­la­tive and all that, but it seems real im­por­tant for villages to have a com­mu­nity cen­ter.

The Case for Impact

Peo­ple buy­ing them­selves nice things only gets you so far. Many of the peo­ple who would most benefit from the nice things are peo­ple new to the area who haven’t got­ten jobs yet.

There are liter­ally billions of dol­lars float­ing around in the EA sphere.

Is REACH a rea­son­able dona­tion tar­get from an EA paradigm?

This re­quires a more care­ful an­swer than the “buy our­selves nice things” per­spec­tive. When buy­ing nice things, it’s im­por­tant not to over­think it too much. Choices are bad and can in­terfere with the hav­ing-of-your-nice things.

When try­ing to save lives and bring about the best pos­si­ble longterm fu­ture, it mat­ters a lot more whether your strat­egy is good. You should be at least a lit­tle sus­pi­cious if the ques­tion “what’s the most good you can do?” out­puts “buy your com­mu­nity some nice things.”

But un­der­con­fi­dence is just as real a sin as over­con­fi­dence. And by 2018, the sheer track record of mee­tups clearly in­di­cates that com­mu­nity in­fras­truc­ture is a con­tender for “se­ri­ous EA cause”.

Quoth Mingyuan in What Are Mee­tups Ac­tu­ally Try­ing To Ac­com­plish:

> I don’t work on mee­tups just be­cause I want peo­ple to have friends (al­though that’s definitely a nice side-effect); I work on them be­cause it seems that they have his­tor­i­cally been able to pro­duce peo­ple and out­puts that have maybe marginally con­tributed to us be­ing less likely to go ex­tinct within the next cou­ple of decades. So let’s try to figure that out.

At the very least, mee­tups are a weird black-box that out­puts agenty peo­ple and pro­jects. And just like you might want to keep fund­ing ba­sic-physics-re­search be­cause it seems to out­put cool shit with­out a clear case for the im­pact, I think we want to keep pour­ing effort into lo­cal com­mu­ni­ties.

But I do think we can get more spe­cific than this.

Mea­sur­ing Intangibles

The most saliently high-im­pact out­put of mee­tups I’m aware of:

  • The Bos­ton meetup pro­vided a lot of en­thu­si­asm and vol­un­teer in­fras­truc­ture that al­lowed Max Teg­mark to launch the Fu­ture of Life In­sti­tute, which in turn got Elon Musk in­volved with AI. (There’s room to de­bate if this was net-pos­i­tive or not. But my cur­rent guess is yes, and the mag­ni­tude of the im­pact was both un­mis­tak­ably high and un­mis­tak­ably causal with mee­tups)

  • I’m not sure how rele­vant the mee­tups were, but my im­pres­sion is that years ago in NYC, the ex­is­tence of the ra­tio­nal­ity com­mu­nity low­ered the ac­ti­va­tion en­ergy for Michael Vas­sar in­tro­duc­ing Holden Karnos­fky to Carl Shul­man, most likely sub­stan­tially chang­ing Givewell’s di­rec­tion.

More gen­er­ally, mee­tups seem to :

  • Foster the growth of ra­tio­nal­ists and EAs, many of whom then go on to work on im­por­tant pro­jects. Some­times this effect is im­me­di­ate, some­times it hap­pens over the course of years

  • In­tro­duce peo­ple to each other (and to the broader ecosys­tem of or­ga­ni­za­tions and thinkers) that in­creases the over­all “luck sur­face area” of both in­di­vi­d­u­als, and the col­lec­tive ra­tio­nal­sphere.

  • Provide a change-in-so­cial-en­vi­ron­ment that al­low im­por­tant ideas from the se­quences to ac­tu­ally take root, lead­ing even­tu­ally to more ca­pa­ble in­di­vi­d­u­als and ideas.

  • In­cu­bate pro­jects (I think FLI counts. Me­taMed didn’t work out in the end but I think from an ex­pected-value and learn­ing frame­work it counted. I think many or­ga­ni­za­tions have their roots in peo­ple bounc­ing into each other at mee­tups)

There are two par­tic­u­lar as­pects of this I’d like to dive into:

The Agency Ladder

Agency – the abil­ity to look at a situ­a­tion, no­tice things that could be im­proved, and that proac­tively set out to do those things on pur­pose – is a mus­cle that can be trained. And it’s eas­ier to train with a smooth difficulty curve, with stakes just high enough to be mean­ingful but not so high as to be par­a­lyz­ingly in­timi­dat­ing.

Weird fact: a lot of peo­ple I know (my­self in­cluded) gained a bunch of agency from run­ning mee­tups.

When I ar­rived in the NYC com­mu­nity, I no­ticed an op­por­tu­nity for some kind of win­ter holi­day. I held the first Sols­tice. The only stakes were 20 peo­ple pos­si­bly hav­ing a bad time. The next year, I planned a larger event that peo­ple trav­eled from nearby cities to at­tend, which re­quired me to learn some lo­gis­tics as well as to im­prove at rit­ual de­sign. The third year I was able to run a ma­jor event with a cou­ple hun­dred at­ten­dees. At each point I felt challenged but not over­whelmed. I made mis­takes, but not ones that ru­ined any­thing longterm or im­por­tant.

In the Bay, there are many or­ga­ni­za­tions, some of which are look­ing for vol­un­teers. But there are only so many vol­un­teers an or­ga­ni­za­tion can pro­duc­tively make use of (and of­ten, even those po­si­tions re­quire skills that a new­comer may lack, es­pe­cially if they’re young).

There isn’t enough sur­face area for new­comer-molecules to bind with and start build­ing their skills and net­work. One of the most im­por­tant ser­vices I think REACH can provide is be­ing op­ti­mized for pro­vid­ing agency-lad­der op­por­tu­ni­ties to new­com­ers, in a scal­able way.

Water Cool­ers and Campuses

I’ve talked a lot about the meetup-as­pect of REACH, be­cause his­tor­i­cally I’ve been a meetup-ori­ented guy and I have a lot of opinions about it. I also hadn’t re­ally seen the cowork­ing as­pect of REACH start to bear fruit un­til re­cently.

But I’ve lately come to be­lieve this is ex­tremely im­por­tant.

There’s a cam­pus based model of in­no­va­tion I’ve been chat­ting about with Oliver and oth­ers. There’s a rea­son Google tries to get all its en­g­ineers to spend a lot of time at Google Cam­pus, pro­vid­ing for all kinds of phys­i­cal and so­cial needs. It keeps en­g­ineers from differ­ent de­part­ments bump­ing into each other. It makes for an en­vi­ron­ment where peo­ple can slowly come to know each other in an or­ganic fash­ion, bounce ideas off each other (ca­su­ally at first, then per­haps more se­ri­ously).

It means that if you want to know some­thing about some­one’s pro­ject, you can just ask some­one the next time you see them at lunch, with­out the vaguely obli­ga­tion-shaped var­i­ant where you send them an email and they have to get around to re­spond­ing.

Water Cooler talk is an im­por­tant as­pect of in­no­va­tion.

I think the cowork­ing as­pect of REACH can be very valuable for this. In the­ory, CFAR offers this, but in prac­tice CFAR is far enough away (and again with the triv­ial in­con­ve­niences of their se­cu­rity sys­tem), that I wouldn’t visit the office nearly as of­ten as I’d visit REACH.

Re­cently, Katja Grace started ex­per­i­ment­ing with an “Peo­ple work­ing on in­de­pen­dent AI pro­jects get to­gether to co-work” thing some­times at REACH, in­spired by a similar thought pro­cess.

This is the sort of thing a com­mu­nity cen­ter can fa­cil­i­tate.

Match­ing Funds

From an EA grant­mak­ing per­spec­tive, I don’t think REACH should be at the top of the list of things you’d want to give money to. But I do think it should be some­where on the list. This is a case where the rele­vant ques­tion is “are you more money con­strained, or shovel-ready-pro­ject con­strained?”

If you’re an Earn­ing-to-Give per­son who has non-triv­ial but limited amounts of money, an­other im­por­tant con­sid­er­a­tion is that, with OpenPhil and BERI fund­ing many of the most ob­vi­ous, cred­ibly high-im­pact-pro­jects, most of the value of be­ing a mid-level donor is in us­ing your lo­cal knowl­edge to seed-fund early stage, fuzzier pro­jects.

From both per­spec­tives, I think it may have been fair to not want to fund the pro­ject be­fore it had any track record. I think I prob­a­bly roughly agree with CEA’s de­ci­sion (at the time) to not give a grant – at that point it was hard to tell what was go­ing to hap­pen, and it was fairly ex­pen­sive as far as com­mu­nity pro­jects go.

But I think a rea­son­able way to eval­u­ate this sort of pro­ject is a match­ing-grants paradigm. The fact that in­di­vi­d­u­als have put for­ward $2500/​month on Pa­treon by now is pretty strong “put money where your mouth is” ev­i­dence that REACH is pro­vid­ing the value it as­pired to.

So… In Clos­ing…

I think REACH is pretty good.

If you’re con­sid­er­ing donat­ing from the “buy our­selves nice things” per­spec­tive but aren’t yet sold… well, just come on by and check it out. Weekly mee­tups are on Wed­nes­day at 7pm. Cowork­ing is ev­ery week­day (and hope­fully soon on week­ends once we work out some kinks).

If you get value out of it, I think it makes sense to pay for a nice thing.

If not, no wor­ries.

If you’re con­sid­er­ing it from an EA/​im­pact per­spec­tive and think it’s promis­ing but have some spe­cific con­cerns, it’s prob­a­bly at least worth a chat with Sarah to see if those con­cerns can be ad­dressed. As noted ear­lier, I think it’s im­por­tant for pro­jects not to be too be­holden to fun­ders, but I do think there’s room for an earnest ex­change of ideas.

If you’re liv­ing in some other ge­o­graphic area, think REACH sounds great and wish you had one in your area… you might want to think about how to make that hap­pen. Differ­ent cities may have differ­ent con­straints, but I sus­pect it will of­ten be achiev­able. There’s a range of pos­si­bil­ities be­tween “group house func­tion­ing as de-facto com­mu­nity hub” and “full fledged cen­ter.”

The Berkeley REACH Pa­treon Page is here.

If you’d like to make a one-time dona­tion, you can do so here.