JoshYou

• JoshYou 3 Aug 2019 22:55 UTC

  4 points

  in reply to: John_Maxwell_IV's comment on: Four prac­tices where EAs ought to course-correct

  Ab­solutely, EAs shouldn’t be toxic, in­ac­cu­rate, or un­char­i­ta­ble on Twit­ter or any­where else. But I’ve seen a few ex­am­ples of peo­ple effec­tively com­mu­ni­cat­ing about EA is­sues on Twit­ter, such as Ju­lia Galef and Kel­sey Piper, at a level of fidelity and nice­ness far above the av­er­age for that web­site. On the other hand they are briefer, more flip­pant, and spend more time re­spond­ing to crit­ics out­side the com­mu­nity than they would on other plat­forms.

• JoshYou 30 Jul 2019 23:36 UTC

  4 points

  in reply to: kbog's comment on: Four prac­tices where EAs ought to course-correct

  Yep, though I think it takes a while to learn how to tweet, whom to fol­low, and whom to tweet at be­fore you can get a con­sis­tently good ex­pe­rience on Twit­ter and avoid the nas­ti­ness and mi­s­un­der­stand­ings it’s in­fa­mous for.

  There’s a bit of an ex­tended uni­verse of Vox writ­ers, economists, and “ne­oliber­als” that are in­ter­ested in EA and some­times tweet about it, and I think it would be po­ten­tially valuable to add some peo­ple who are more knowl­edge­able about EA into the mix.

• JoshYou 30 Jul 2019 22:50 UTC

  1 point

  on: Four prac­tices where EAs ought to course-correct

  On point 4, I won­der if more EAs should use Twit­ter. There are cer­tainly many op­tions to do more “ruth­less” com­mu­ni­ca­tion there, and it might be a good way to spread and pop­u­larize ideas. In any case it’s a pretty con­crete ex­am­ple of where fidelity vs. pop­u­lar­ity and nice­ness vs. ag­gres­sive pro­mo­tion trade off.

• JoshYou 15 Jun 2019 15:52 UTC

  2 points

  on: What Do Un­con­scious Pro­cesses in Hu­mans Tell Us About Sen­tience?

  This all seems to as­sume that there is only one “ob­server” in the hu­man mind, so that if you don’t feel or per­ceive a pro­cess, then that pro­cess is not felt or per­ceived by any­one. Have you ruled out the pos­si­bil­ity of sen­tient sub­rou­tines within hu­man minds?

• JoshYou 6 May 2019 5:09 UTC

  11 points

  in reply to: michaelchen's comment on: Is pre­vent­ing child abuse a plau­si­ble Cause X?

  Sadly, Ji­woon passed away last year.

• JoshYou 22 Apr 2019 21:16 UTC

  1 point

  on: [Ques­tion] Pros/​Cons of Donor-Ad­vised Fund

  Some links if you haven’t seen them yet:

  https://​​re­duc­ing-suffer­ing.org/​​ad­vanced-tips-on-per­sonal-fi­nance/​​

  https://​​80000hours.org/​​2013/​​06/​​how-to-cre­ate-a-donor-ad­vised-fund/​​

  I don’t use a DAF but I’ve con­sid­ered it in the past. In my view, the chief ad­van­tage is that they al­low you to claim the tax de­duc­tion when you de­posit money into the DAF, be­fore you ac­tu­ally make the dona­tion. They’re also ex­empt from cap­i­tal gains taxes, al­though you can also avoid cap­i­tal gains taxes by donat­ing ap­pre­ci­ated as­sets di­rectly to the char­ity, but that de­pends on whether the or­ga­ni­za­tion will ac­cept them (not sure how uni­ver­sal this is). They also charge fees, which can be fairly ex­pen­sive but are cheaper than cap­i­tal gains taxes on ex­pec­ta­tion.

• JoshYou 18 Apr 2019 2:47 UTC

  13 points

  in reply to: Milan_Griffes's comment on: Should EA grant­mak­ing be sub­ject to in­de­pen­dent au­dit?

  Open Phil would be a good can­di­date for this, though that’s a difficult propo­si­tion due to its sheer size. It is a some­what odd situ­a­tion that Open Phil throws huge amounts of money around, much of which hap­pens with­out any com­ment from the EA com­mu­nity.

• JoshYou 26 Mar 2019 23:42 UTC

  15 points

  on: Why is the EA Ho­tel hav­ing trou­ble fundrais­ing?

  I won­der if the lack of tax de­ductibil­ity and the non-con­ven­tional fundrais­ing plat­form (GoFundMe) nudge peo­ple into not donat­ing or donat­ing less than they would to a more re­spectable-seem­ing char­ity.

  (As a tan­gent, there’s a dona­tion swap op­por­tu­nity for the EA Ho­tel that most peo­ple are prob­a­bly not aware of).

• JoshYou 13 Mar 2019 0:07 UTC

  9 points

  on: EA Ho­tel Fundraiser 3: Es­ti­mat­ing the rel­a­tive Ex­pected Value of the EA Ho­tel (Part 1)

  Speak­ing as some­one with a un­der­grad de­gree in math, I would have found a non-tech­ni­cal sum­mary for this post to be helpful. So I ex­pect this would ap­ply much more to many other fo­rum read­ers.

• JoshYou 28 Feb 2019 20:14 UTC

  2 points

  in reply to: Jonas Vollmer's comment on: After one year of ap­ply­ing for EA jobs: It is re­ally, re­ally hard to get hired by an EA organisation

  For one of the work tests I did for Open Phil, the in­struc­tion sheet speci­fi­cally asked that the work test not be shared with any­one. That might have been in­tended as a tem­po­rary re­stric­tion, I’m not sure, but I’m not plan­ning on shar­ing it un­less I hear oth­er­wise.

• JoshYou 26 Jan 2019 15:23 UTC

  8 points

  in reply to: casebash's comment on: Vox’s “Fu­ture Perfect” column fre­quently has flawed journalism

  Agreed. I don’t see any “poor jour­nal­ism” in any of the pieces men­tioned. A few of them would be “poor in­ter­ven­tion re­ports” if we chose to judge them by that stan­dard.

• JoshYou 12 Jan 2019 3:30 UTC

  9 points

  on: Cli­mate Change Is, In Gen­eral, Not An Ex­is­ten­tial Risk

  It’s clear that cli­mate change has at best a small prob­a­bil­ity (well un­der 10%) of caus­ing hu­man ex­tinc­tion, but many pro­po­nents of work­ing on other x-risks like nu­clear war and AI safety would prob­a­bly give low prob­a­bil­ities of hu­man ex­tinc­tion for those risks as well. I think the pos­i­tive feed­back sce­nar­ios you men­tion (per­mafrost, wet­lands, and ocean hy­drates) de­serve some at­ten­tion from an x-risk per­spec­tive be­cause they seem to be poorly un­der­stood, so the up­per bound on how se­vere they might be may be very high. You cite one simu­la­tion that burn­ing all available fos­sil fuels would in­crease tem­per­a­tures by 10 °C, but that isn’t nec­es­sar­ily an up­per bound be­cause there are non-fos­sil fuel sources car­bon on Earth that could be re­leased to the at­mo­sphere. It would of course also be nec­es­sary to es­ti­mate how high the ex­tinc­tion risk con­di­tional on var­i­ous lev­els of ex­treme warm­ing (8°C, 10°C, 15°C, 20°C?) would be.

  Re­gard­less, it’s a good idea to have a clear view of how big the risk is. You’re right that the ca­sual claims about ex­tinc­tion or plane­tary un­in­hab­it­a­bil­ity I hear from many peo­ple who are con­cerned about cli­mate change are not jus­tified, and they seem a bit ir­re­spon­si­ble.

• JoshYou 10 Jan 2019 3:48 UTC

  6 points

  on: How should large donors co­or­di­nate with small donors?

  Holden also wrote (by the way, I think your link is bro­ken):

  We fully funded things we thought were much bet­ter than the “last dol­lar” (in­clud­ing cer­tain top char­i­ties grants) but not things we thought were rel­a­tively close when they also posed co­or­di­na­tion is­sues. For this case, fully fund­ing top char­i­ties would have had pros and cons rel­a­tive to split­ting: we think the dol­lars we spent would’ve done slightly more good, but the dol­lars spent by oth­ers would’ve done less good (and we think we have a good sense of the coun­ter­fac­tual for most of those dol­lars). We guessed that the lat­ter out­weighed the former.

  So an im­por­tant crux here is the pro­por­tion of small-donor money to e.g. GiveWell char­i­ties that would be crowded out into much less effec­tive char­i­ties or to new pro­jects with high ex­pected value. For refer­ence, GiveWell has moved about $30-40 mil­lion a year in small dona­tions. I am not sure what pro­por­tion of that comes from peo­ple who are not closely al­igned/​af­fili­ated with the EA com­mu­nity, but I would guess it’s the ma­jor­ity.

  I would ques­tion whether Holden is cor­rect though. Global health/​de­vel­op­ment is a big space, so if Good Ven­tures in­creased fund­ing to GiveWell top char­i­ties by a lot, GiveWell would still ex­ist and would move their recom­men­da­tions over to in­ter­ven­tions that aren’t fully funded yet. For ex­am­ple, cash trans­fers seem­ingly could ab­sorb a lot of money, and the Gates Foun­da­tion prob­a­bly moves more to global poverty causes ev­ery year than GoodVen­tures will spend per year at its peak. The claim seems to de­pend on small GiveWell donors be­ing ex­cited by GiveWell’s spe­cific top char­i­ties right now, such that they would not give to GiveWell top char­i­ties if the cur­rent top char­i­ties were fully funded and GiveWell is­sued new recom­men­da­tions, and would in­stead give to char­i­ties even less effec­tive than these new top char­i­ties. That might be true if donors are re­ally mo­ti­vated by the head­line cost-per-life-saved num­ber rather than be­ing at­tracted by GiveWell’s re­search and method­ol­ogy. I don’t have a very strong in­tu­ition ei­ther way, so I’d be cu­ri­ous if some­one more knowl­edge­able could shed some light.

• JoshYou 6 Jan 2019 14:56 UTC

  32 points

  on: EA orgs are try­ing to fundraise ~$10m - $16m

  If we’re us­ing these num­bers to in­form whether EA is fund­ing con­strained, it would be good if some­one fol­lowed up and figured out how much these or­ga­ni­za­tions ac­tu­ally ended up rais­ing.

• JoshYou 21 Dec 2018 23:32 UTC

  3 points

  on: Challenges in Scal­ing EA Organizations

  One thing I’ve won­dered about is what the op­ti­mal rate at which new EA or­ga­ni­za­tions should be founded, and whether that’s an effec­tive way around growth bot­tle­necks. For ex­am­ple, Re­think Pri­ori­ties has grown rapidly this year, and it doesn’t seem likely that that growth would have hap­pened any­way within pre­vi­ously ex­ist­ing or­ga­ni­za­tions had Re­think Pri­ori­ties not been founded.

• JoshYou 20 Dec 2018 1:29 UTC

  6 points

  on: An­i­mal Welfare Fund AMA

  This fund has seem­ingly taken a very “hits-based” ap­proach to fund­ing small, in­ter­na­tional grass­roots or­ga­ni­za­tions. How do you plan on eval­u­at­ing and learn­ing from these grants?

• JoshYou 20 Dec 2018 1:14 UTC

  15 points

  on: Long-Term Fu­ture Fund AMA

  This post con­tains an ex­ten­sive dis­cus­sion on the difficulty of eval­u­at­ing AI char­i­ties be­cause they do not share all of their work due to info haz­ards (in the “Open­ness” sec­tion as well as the MIRI re­view). Will you have ac­cess to work that is not shared with the gen­eral pub­lic, and how will you ap­proach eval­u­at­ing re­search that is not shared with you or not shared with the pub­lic?

• JoshYou 20 Dec 2018 1:11 UTC

  14 points

  on: Long-Term Fu­ture Fund AMA

  Un­der what con­di­tions would you con­sider mak­ing a grant di­rected to­wards catas­trophic risks other than ar­tifi­cial in­tel­li­gence?

• JoshYou 17 Dec 2018 5:16 UTC

  1 point

  in reply to: Davidmanheim's comment on: New web app for cal­ibra­tion train­ing funded by the Open Philan­thropy Project

  Vague or con­text-less ques­tions might help you cal­ibrate your views on top­ics you know very lit­tle about?

  I am now some­what bet­ter cal­ibrated at claims about Euro­pean foot­ball than I was be­fore, I guess.

• JoshYou 16 Nov 2018 0:30 UTC

  9 points

  on: 2017 Donor Lot­tery Report

  This is a great writeup and also a good demon­stra­tion of the value of donor lot­ter­ies. Is CEA plan­ning on run­ning an­other one any­time soon? Their lot­ter­ies page just says “There are cur­rently no ac­tive lot­ter­ies”. I think the lot­tery ex­per­i­ments have gone well and this should be a reg­u­lar thing, un­less run­ning a lot­tery con­sumes a lot of staff time or has some other large cost.