Greg_Colbourn

• Greg_Colbourn 11 Apr 2019 9:39 UTC

  1 point

  in reply to: Milan_Griffes's comment on: Can we ap­ply start-up in­vest­ing prin­ci­ples to non-prof­its?

  I think that’s kind of the point: that it should be thought of as ap­ples-to-ap­ples—there are these huge ar­bi­trage op­por­tu­ni­ties when it comes to non-profit in­vest­ing (or EA) that aren’t available with a di­rect for-profit in­cen­tive (if you make some ba­sic as­sump­tions like hu­man lives be­ing of equal value re­gard­less of na­tion­al­ity).

  Imag­ine if you could buy 200kg of gold for $3,500 in Malawi!

• Greg_Colbourn 9 Apr 2019 10:03 UTC

  15 points

  in reply to: Jonas Vollmer's comment on: $100 Prize to Best Ar­gu­ment Against Donat­ing to the EA Hotel

  Hu­mans in gen­eral of­ten don’t take crit­i­cal feed­back nearly as se­ri­ously as they should, and of­ten don’t ad­just their think­ing/​ac­tions due to sunk costs, want­ing to save face in their peer group, grandiose per­son­al­ity, etc. This also ap­plies to EAs (maybe some­what but not vastly less so).

  Agree that this can be tough (from ex­pe­rience). I would add that it can be emo­tion­ally drain­ing, es­pe­cially if the feed­back is un­char­i­ta­ble or has some ba­sis of mi­s­un­der­stand­ing or fac­tual er­ror. Also, it can be fur­ther com­pli­cated if af­ter re­flec­tion one still doesn’t fully agree with the feed­back and there is a gen­uine philo­soph­i­cal dis­agree­ment. (NB I’m happy to have re­ceived feed­back from Jonas Vol­lmer and think it has/​will make the EA Ho­tel pro­ject stronger; “un­char­i­ta­ble or has some ba­sis of mi­s­un­der­stand­ing or fac­tual er­ror” does not ap­ply to his feed­back).

  Aaron Gertler’s ap­proach of “look­ing at the Ho­tel’s guest list, pick­ing the best-sound­ing pro­ject, and offer­ing money di­rectly to the per­son be­hind it.”

  This does re­quire the ho­tel to ex­ist though (or some­thing like it). See my com­ment here.

  I have also shared some thoughts for how to de­sign the ad­mis­sion pro­cess with EA Ho­tel staff.

  Based on some ini­tial ideas from Jonas, we are work­ing on a rat­ing sys­tem for ap­pli­cants and on­go­ing hosted pro­jects. Ten­ta­tively it might be some­thing like a log­a­r­ith­mic scale of EV, {-5,+5} with +1 = giv­ing the money to GiveDirectly*. Trus­tees/​Man­ager in one anony­mous pool, Ad­vi­sors in an­other. Bayesian pri­ors stated in words. 95% con­fi­dence in­ter­vals given. Another round of scor­ing af­ter see­ing oth­ers’ in­put and dis­cus­sion (spe­cial care taken to dis­cuss when rat­ings <=-1 are given). Fi­nal scores ag­gre­gated. Guests ac­cepted if clear­ing a bar of +1 (to in­crease with diminish­ing ca­pac­ity). If fal­ling be­low, guests have 3 months to pivot/​im­prove.

  *Would be in­ter­ested in com­par­ing with any nu­mer­i­cal schemes other EA grant­mak­ers are us­ing.

• Greg_Colbourn 9 Apr 2019 9:02 UTC

  13 points

  in reply to: Milan_Griffes's comment on: Long Term Fu­ture Fund: April 2019 grant decisions

  We ap­plied. Judg­ing by the email I re­ceived, I think we are also part of the small group of ~7 men­tioned here. Await­ing the fol­low up email.

• Greg_Colbourn 8 Apr 2019 22:43 UTC

  5 points

  in reply to: Milan_Griffes's comment on: The Case for the EA Hotel

  I haven’t yet, but fundrais­ing is prov­ing to be a lot more stress­ful than I ex­pected :)

• Greg_Colbourn 8 Apr 2019 22:39 UTC

  3 points

  in reply to: Peter_Hurford's comment on: The Case for the EA Hotel

  To avoid peo­ple be­com­ing overly de­pen­dent, we do have a hard limit of 2 years for stays at the ho­tel (most peo­ple seem to be go­ing for some­where in the range of 3-12 months).

• Greg_Colbourn 8 Apr 2019 22:35 UTC

  2 points

  in reply to: ofer's comment on: The Case for the EA Hotel

  I’m guess­ing it was prob­a­bly this com­ment.

• Greg_Colbourn 8 Apr 2019 22:22 UTC

  10 points

  in reply to: aarongertler's comment on: $100 Prize to Best Ar­gu­ment Against Donat­ing to the EA Hotel

  In an­swer to your first point, see my re­ply to Moses here.

  --

  Re­gard­ing your sec­ond point—com­par­ing with other Meta op­por­tu­ni­ties—one might also want to con­sider that many Meta pro­jects have goals of bring­ing more peo­ple into the com­mu­nity. We need to then cater for these peo­ple. The EA Ho­tel can help do that, and fill an im­por­tant gap.

• Greg_Colbourn 8 Apr 2019 21:56 UTC

  7 points

  in reply to: casebash's comment on: $100 Prize to Best Ar­gu­ment Against Donat­ing to the EA Hotel

  Per­haps it would have been bet­ter to have started with a smaller scale ex­per­i­ment such as a group house

  I don’t think a small group house would’ve gen­er­ated com­mu­nity (or in­ter­est) on the same level. What we have seems more valuable on ac­count of its scale than, say, 4 sep­a­rate group houses with the same num­ber of to­tal res­i­dents.

  too costly a way of learn­ing about the po­ten­tial of such projects

  In the scheme of things, I don’t think it’s been that costly (~£60k spent to date by donors).

  achieve suffi­cient ex­pected return

  Depends on what your coun­ter­fac­tual is. My ini­tial thoughts were on the lines of “is the EV of fund­ing this per­son’s work likely to be greater than that of donat­ing the money to a GiveWell top char­ity?”. We are cur­rently work­ing on im­ple­ment­ing a rat­ing sys­tem for pro­jects. I have sug­gested that, space-per­mit­ting, we set the bar to clear as “equiv­a­lent to donat­ing the money to Give Directly”. The bar would be raised pro­por­tion­ally to how lit­tle free ca­pac­ity the ho­tel has, al­though in prin­ci­ple the ho­tel has the po­ten­tial to ex­pand given available build­ings on our street. Ob­vi­ously mak­ing such judge­ments comes with large er­ror bars and a heavy weight­ing of pri­ors. Also, given this is Hits-based Giv­ing, I’m hope­ful that the long-term mean value of pro­jects will be sig­nifi­cantly above the en­try bar.

  peo­ple might not want to donate be­cause they don’t be­lieve that other peo­ple will donate

  We have con­sid­ered that per­haps a Kick­starter-like mechanism is needed here. How­ever, given the re­cent in­ter­est of a few peo­ple in donat­ing at the 4-figure level, I’m more op­ti­mistic that we can get by with­out it (al­though it might be use­ful for other new pro­jects in the EA space that re­quire a sig­nifi­cant ini­tial out­lay, the added bonus be­ing the so­cial proof).

• Greg_Colbourn 8 Apr 2019 20:03 UTC

  8 points

  in reply to: Anoneacommenter's comment on: $100 Prize to Best Ar­gu­ment Against Donat­ing to the EA Hotel

  For more on our res­i­dents to date, and how the ho­tel has helped them, here are some case stud­ies. See also: out­puts.

  fun­ders now are look­ing at the first “low­est qual­ity” res­i­dents and not find­ing them attractive

  I think this is un­fair. I have not had any po­ten­tial fun­ders speci­fi­cally say this. I ac­tu­ally think that the qual­ity of res­i­dents is already pretty high (and this is un­der-ap­pre­ci­ated from the out­side). I also dis­agree with the no­tion of the first res­i­dents nec­es­sar­ily be­ing of low qual­ity just by way of not hav­ing strong com­pe­ti­tion to be ac­cepted; there might even be a “de­but album effect” in that there is a glut of good peo­ple (“songs”) ready for the ini­tial re­lease. I know I’ve per­son­ally been pleas­antly sur­prised more than once with guests who I were a bit un­sure of origi­nally but was happy to take a chance on. There is also the con­sid­er­a­tion that it takes a cer­tain amount of pi­o­neer spirit to join a new pro­ject and com­mu­nity early in its life, and that this is pos­i­tively cor­re­lated with agency and via that, value cre­ation.

  se­lect­ing good fund­ing op­por­tu­ni­ties… no ev­i­dence has been given as to why the ho­tel staff are likely to be good at that role

  See our pitch doc (that we’ve re­cently cir­cu­lated to in­ter­ested peo­ple) for more de­tail on our Staff/​Trus­tees. Ul­ti­mately, most of the grant mak­ers in charge of the EA Funds have gained their ex­pe­rience through be­ing ac­tively in­volved in EA and mak­ing fund­ing de­ci­sions (dona­tions) over many years. We are not that differ­ent. See my record of EA dona­tions here*. Also, as Vipul says, we have skin in the game here.

  *“EA Ho­tel” is the money I’ve put in as of 29th March 2019, not in­clud­ing pur­chas­ing the build­ing. Meta note: would be good if it was pos­si­ble to make pages on app.effec­tivealtru­ism.org/​​dash­board/​​pledge/​​dona­tions pub­li­cly vie­w­able. Is this fea­ture planned?

• Greg_Colbourn 8 Apr 2019 18:56 UTC

  12 points

  in reply to: Moses's comment on: $100 Prize to Best Ar­gu­ment Against Donat­ing to the EA Hotel

  I think this view as pre­sented has an overly nar­row fo­cus. In terms of think­ing of the ex­pected value of the ho­tel and whether it’s worth fund­ing on the mar­gin, it’s use­ful to also con­sider:

  • The benefits of the in-per­son com­mu­nity in terms of sup­port, feed­back, mo­ti­va­tion, pro­duc­tivity, col­lab­o­ra­tion, net­work­ing.
    

  • All the po­ten­tial fu­ture value from fu­ture guests and iter­at­ing, ex­pand­ing and fran­chis­ing the model.
    

  • The effect it failing from a lack of fund­ing would have on the like­li­hood of similar ini­ti­a­tives be­ing started in fu­ture.

  • The no­tion of Hits-based Giv­ing.
    

  Also note that the coun­ter­fac­tu­als—EA Grants and EA Meta Fund—have not had as­sess­ments of their out­puts performed, or at least made pub­lic (apart from the larger more es­tab­lished pro­jects they have funded). In­deed—some­one cor­rect me if I’m wrong—we don’t even know who the re­cip­i­ents are of any of the EA Grants made af­ter Fall 2017 (but then again, EA Grants in and of it­self does not pub­li­cly ac­cept dona­tions, so does not need to be so trans­par­ent). Also: our costs per per­son are sig­nifi­cantly lower than those of the av­er­age EA Grants or Meta Fund grantee. You need to fac­tor this mul­ti­plier in when judg­ing the rel­a­tive mer­its of them vs. the EA Ho­tel.

  In terms of pick­ing and choos­ing peo­ple to fund, this is not read­ily pos­si­ble with­out some kind of ag­gre­ga­tion of ap­pli­cants. Such ag­gre­ga­tion pro­jects have been pro­posed here and here, but not with­out con­tro­versy (see com­ments on those posts). The ho­tel has the effect of ag­gre­gat­ing a se­lec­tion of peo­ple and pro­jects start­ing out in EA, and in prin­ci­ple would-be fun­ders can offer to pay for the costs of in­di­vi­d­ual peo­ple or pro­jects hosted at the ho­tel (or even offer to fur­ther fund them at higher rates to ex­pand their pro­jects el­se­where). But this does re­quire the ho­tel to con­tinue to ex­ist (or some other kind of ag­gre­ga­tor to take its place). Another way of look­ing at it would be by way of anal­ogy to ven­ture cap­i­tal. By fund­ing the EA ho­tel you are buy­ing the port­fo­lio of a VC firm (that is tak­ing a hits-based giv­ing ap­proach); by pick­ing in­di­vi­d­ual pro­jects you are do­ing the work of an An­gel In­vestor. The lat­ter takes sig­nifi­cantly more time and work. (NB this is differ­ent to or­di­nary in­vest­ing in that we need to be con­cerned about anti-uni­corns whilst on the look out for uni­corns).

  The con­tro­versy around ag­gre­ga­tors re­volves around in­creas­ing the risk from con­sid­er­a­tions re­lat­ing to the “unilat­er­al­ists curse” i.e. po­ten­tially net nega­tive pro­jects would have more of a chance of be­ing funded if given a wider au­di­ence of po­ten­tial fun­ders. I think this is one rea­son why EA Grants don’t widely share their ap­pli­ca­tions. The ho­tel guards against this some­what by hav­ing a com­mit­tee of trustees (and soon to add—ex­ter­nal ad­vi­sors) in­volved with over­see­ing the vet­ting pro­cess. Also there is the filter for com­mit­ment that is hav­ing peo­ple phys­i­cally move to Black­pool, rather than just tak­ing the money. And the on hand in-per­son com­mu­nity to get ad­vice and feed­back from.

• Greg_Colbourn 8 Apr 2019 13:37 UTC

  3 points

  in reply to: JoshYou's comment on: Why is the EA Ho­tel hav­ing trou­ble fundrais­ing?

  We now also have a PayPal MoneyPool.

• Greg_Colbourn 8 Apr 2019 13:36 UTC

  1 point

  in reply to: toonalfrink's comment on: Why is the EA Ho­tel hav­ing trou­ble fundrais­ing?

  See here and here.

• Greg_Colbourn 7 Apr 2019 15:45 UTC

  1 point

  in reply to: Brendon_Wong's comment on: $100 Prize to Best Ar­gu­ment Against Donat­ing to the EA Hotel

  1. The same work, yes, but at a much slower rate (fac­tor­ing in the need to work out­side jobs to sup­port them­selves; the haste con­sid­er­a­tion might also come into play), or burn­ing run­way at a faster rate than the EA Ho­tel is spend­ing on them (if you take the view that con­serv­ing EA money in gen­eral is a plus, then this is a win). Re­gard­ing your point (3) as stated in the origi­nal blog post, we are open to fund­ing peo­ple’s liv­ing ex­penses in other places if they are com­pa­rable to what they are for ho­tel guests. No one has yet taken us up on this though. And it also re­quires a high level of trust that peo­ple won’t just spend more money some­where else and so there­fore work for a shorter amount of time for the cost.
  
  As you men­tion, the ad­di­tional com­mu­nity benefits are likely to be con­sid­er­able. Co-work­ing out of REACH might give some of them, but is also likely to be a lot more ex­pen­sive, and there isn’t that much space at REACH for peo­ple to do this. Also liv­ing to­gether gen­er­ally al­lows for a much greater de­gree of so­cial in­ter­ac­tion and a more close-knit sup­port­ive com­mu­nity.
  
  2. This cost sav­ing for all in­volved (EA Grants, EA Gran­tees) re­quires the ho­tel to ex­ist though. And it won’t if it doesn’t bring in enough early stage fund­ing to be­come more es­tab­lished. Also, it’s worth con­sid­er­ing that fact that it’s un­likely the ho­tel would’ve got off the ground at all had we charged ev­ery­one from the out­set.
  
  Note that we re­quest peo­ple who have salaries or >2 years of run­way to pay cost price.
  
  3. What pro­ject ex­penses are you imag­in­ing that don’t in­clude salary? (I say this as a salary is usu­ally spent on liv­ing costs and if liv­ing costs are pro­vided the need for it is a lot lower). The ho­tel al­lows for the cheap grant­ing—and pur­chas­ing—of run­way for EA pro­jects.
  
  4. See here and here for a start.
  
  The EV post might look like a lot of scary maths on the face of it, but it’s re­ally not that com­pli­cated. It’s just an at­tempt to ex­plic­itly fac­tor out the differ­ent con­sid­er­a­tions that go into es­ti­mat­ing the EV of the EA Ho­tel. How­ever, I un­der­stand that gut feel­ings tend to dom­i­nate such es­ti­mates. It can be use­ful to san­ity check your gut in­stincts by do­ing such an ex­plicit calcu­la­tion, but equally it’s the math­e­mat­i­cal model that of­ten needs ad­just­ing (or throw­ing out) if it doesn’t give you an an­swer close to your in­tu­itive es­ti­mate.

• Greg_Colbourn 7 Apr 2019 15:14 UTC

  1 point

  in reply to: Anoneacommenter's comment on: Why is the EA Ho­tel hav­ing trou­ble fundrais­ing?

  Here’s the list of out­puts. At the time of the first fundrais­ing post, we’d only been at ca­pac­ity for a cou­ple of months, and not many peo­ple had stayed longer than that (with only ap­prox. half the to­tal per­son-days spent at the ho­tel). Back then, I was hop­ing the ap­peal of “Hits-based Giv­ing”, and “EA com­mu­nity hub” would be stronger than they have been.

• Greg_Colbourn 1 Apr 2019 15:11 UTC

  25 points

  on: My Q1 2019 EA Ho­tel donation

  Thanks so much Vipul! :)

  With this model of spend­ing, I can con­cretely think of my dona­tion of 3200 GBP as hav­ing added about 13 days to the ho­tel’s run­way.

  Note that costs have been lower re­cently (due to rely­ing more on vol­un­teer labour, and an in­crease in con­tri­bu­tions from guests) - for an av­er­age over Jan­uary-March your dona­tion cov­ers 20 days of costs, and for March alone it cov­ers 27 days (and will likely be similar for April).

• Greg_Colbourn 29 Mar 2019 12:34 UTC

  1 point

  in reply to: Raemon's comment on: EA Ho­tel Fundraiser 3: Es­ti­mat­ing the rel­a­tive Ex­pected Value of the EA Ho­tel (Part 1)

  Thanks for the dona­tion! :)

• Greg_Colbourn 29 Mar 2019 11:41 UTC

  1 point

  in reply to: ofer's comment on: $100 Prize to Best Ar­gu­ment Against Donat­ing to the EA Hotel

  I’m guess­ing it’s be­cause it links to a post in the EA Ho­tel Face­book group, which is a closed group. You can join here.

• Greg_Colbourn 28 Mar 2019 17:18 UTC

  3 points

  in reply to: Raemon's comment on: EA Ho­tel Fundraiser 3: Es­ti­mat­ing the rel­a­tive Ex­pected Value of the EA Ho­tel (Part 1)

  Ok, PayPal money pool set up here: https://​​www.pay­pal.com/​​pools/​​c/​​8dptKvmvxO

  Also linked to on our fundrais­ing page: ea­ho­tel.org/​fundraiser.

• Greg_Colbourn 27 Mar 2019 23:34 UTC

  17 points

  in reply to: Liam_Donovan's comment on: Why is the EA Ho­tel hav­ing trou­ble fundrais­ing?

  2. RAISE very much does en­dorse the ho­tel (es­pe­cially given that the founder works for and lives at the ho­tel, and the ho­tel was in­te­gral to their progress over the last 6 months). See e.g. here and here. We have no for­mal re­la­tion­ship with Re­think Char­ity (or Re­think Pri­ori­ties in par­tic­u­lar) - in­di­vi­d­u­als at the ho­tel have ap­plied for and got work from them in­de­pen­dently.

  3. The marginal cost of adding a new res­i­dent when already at ~75% ca­pac­ity is ~£4k/​yr.

  6. I won­der too. I won­der also how differ­ent it would be if it was done af­ter an­other 6-12 months of get­ting es­tab­lished.

• Greg_Colbourn 27 Mar 2019 23:00 UTC

  3 points

  in reply to: Raemon's comment on: EA Ho­tel Fundraiser 3: Es­ti­mat­ing the rel­a­tive Ex­pected Value of the EA Ho­tel (Part 1)

  PayPal won’t let me make one un­for­tu­nately. Will look into it fur­ther to­mor­row.