Spiracular

• Spiracular 6 Oct 2019 22:43 UTC

  2 points

  on: Who lacks the qualia of con­scious­ness?

  In one of the sub-com­ments, I thought about the tests that iden­ti­fied men­tal-imagery, and started think­ing of how you might test for sev­eral var­i­ants of “lack of sense of self” or some re­lated at­tributes.

  Re­lated Tests for in­ner-sense or model of self

  No-qualia seems challeng­ing to test. But “no model of self” (one form of “lack-of-self-aware­ness”) seems halfway-there, or at least in the cor­rect spirit of the ques­tion? And I think that could be tested re­li­ably; just get a group of peo­ple to pre­dict their own be­hav­ior, and watch that sub­set of the group who re­li­ably fail catas­troph­i­cally at this.

  For lack of con­sis­tency and other-aware­ness… There’s a Nazi (ETA: Eich­mann) who seemed likely to be a trou­blingly-vivid ex­am­ple of “no con­sis­tent wor­ld­view or other-aware­ness”; all his words and be­liefs were in­con­sis­tent plat­i­tudes, and he seemed gen­uinely sur­prised when Jewish judges didn’t feel sym­pa­thy for the difficulty in his at­tempts to get pro­moted by do­ing a “good job” op­ti­miz­ing trains for death. Un­for­tu­nately, I can’t track down his name. If some­one knows who I’m talk­ing about, I’d love to be pointed at an ar­ti­cle about his strange psy­chol­ogy again.

  Lack-of-at­tach­ment-to-in­ter­nal-iden­tity seems to be an­other semi-re­lated thing. I feel like there are some things where I care about “iden­tity-al­ign­ment” a great deal, and other mat­ters that oth­ers clearly care about where I just lack any feel­ing of iden­tity eu­pho­ria/​dys­pho­ria around the mat­ter re­gard­less of what I do. I sus­pect there are some peo­ple who lack ei­ther sen­sa­tion al­to­gether. Prob­a­bly some frac­tion of those peo­ple come across as iden­tity chameleons; peo­ple who switch out iden­tities ac­cord­ing to ex­ter­nal in­cen­tives, be­cause they have no in­ter­nal rea­son not to.

  (Per­son­ally, med­i­ta­tion up­dated me con­sid­er­ably to­wards a re­duced at­tach­ment to in­ter­nal-iden­tities, but there are still some I’m at­tached to and care about main­tain­ing.)

  Alex­ithymia is a phe­nomenon where you lack aware­ness of your own emo­tions, some­times even as you are act­ing them out. This seems easy to test in a man­ner similar to green/​red color-blind­ness; have the per­son try to ap­praise what sort of emo­tion they’re feel­ing, and then read their cir­cum­stances or watch their be­hav­ior for a read of which emo­tion it ac­tu­ally is, and see who usu­ally seems to mis­judge it (or be­lieve they’re not feel­ing emo­tions en­tirely).

  Another p-zom­bie variant

  There’s a differ­ent p-zom­bie sub­type I’ve been think­ing about a great deal my­self..

  If you set up a sys­tem where there’s an ob­server, an ac­tor, and an en­vi­ron­ment, then there are 2 kinds of con­scious­ness:

  • The ob­server in­fluences, con­trols, or is the ac­tor who does things in the en­vi­ron­ment

    • A policy feed­back loop

  • The ob­server is just mod­el­ing what the ac­tor will do in the en­vi­ron­ment

    • A one-way mod­el­ing of the agent

  I sus­pect the later can feel “con­scious” even if the ob­server never in­fluences the ac­tor in any way.

  Hu­mans are usu­ally a bit of both, but some­one who only has “con­scious­ness” in the later ca­pac­ity feels a bit like a… “con­scious­ness hitch­hik­ing on a q-zom­bie” to me.

  (Re­lated: The Elephant and The Rider)

• Spiracular 6 Oct 2019 22:36 UTC

  1 point

  in reply to: stochastic_bit's comment on: Who lacks the qualia of con­scious­ness?

  Men­tal imagery: Draw­ing seems to re­li­ably dis­t­in­guish be­tween co­her­ent-men­tal-vi­su­al­iz­ers and those who aren’t. Tasks like “count the stripes on the tiger” make sense to a vivid/​de­tailed vi­su­al­izer, but not to some­one who is just hold­ing on to the con­cept “striped big cat.”

  I sus­pect draw­ing-at­tempts would also re­li­ably iden­tify peo­ple like “The Man who Mis­took His Wife for a Hat”, who viewed peo­ple as a “di­s­or­ga­nized bag of fa­cial-fea­tures” and had to rely on a sin­gle dis­tinc­tive trait to iden­tify even peo­ple he knew well (ex: Albert Ein­stein & his ec­cen­tric hairstyle), and who de­scribed things that weren’t there when try­ing to in­ter­pret a low-fea­ture image like a pic­ture of the dunes of the Sa­hara.

• Spiracular 25 Sep 2019 13:46 UTC

  18 points

  in reply to: An1lam's comment on: Are there tech­ni­cal/​ob­ject-level fields that make sense to re­cruit to LessWrong?

  Biol­ogy-nerd LWer here (or ex-biol­ogy-nerd? I do pro­gram­ming as a job now, but still talk and think about bio as a fairly-high-in­vest­ment hobby). BS in en­to­mol­ogy. Dis­claimer that I haven’t done grad school or much re­search; I have just thought about do­ing it and talked with peo­ple who have.

  I sus­pect one thing that might ap­peal to these sorts of peo­ple, which we have a chance of be­ing able to provide, is an in­ter­est­ing ap­plied-re­searcher-tar­geted semi-plain-lan­guage (or highly-vi­sual, or flow-chart/​check­list, or oth­er­wise ac­cessibly pre­sented) ex­pla­na­tion of cer­tain as­pects of statis­tics that are par­tic­u­larly likely to be rele­vant to these fields.

  ETA: A few things I can think of as places to find these peo­ple are “re­search” and “con­fer­ences.” There are google terms they’re go­ing to use a lot (due to re­search), and also a lot of them are go­ing to be in­ter­ested in pub­lish­ing and con­fer­ences as a way to fa­mil­iarize them­selves with new re­search in their fields and fur­ther their ca­reers.

  Lean­ing to­wards the re­search fun­nel… here’s some things I un­der­stand now that I did not un­der­stand when I grad­u­ated, many of which I got from talk­ing/​read­ing in this com­mu­nity, which I think a “coun­ter­fac­tual re­searcher me” would have benefited from a lu­cid ex­pla­na­tion of:

  • Trans­fer­ring in­tu­itions around nor­mal­iza­tion

    • how to do it, why to do it

    • see how it elimi­nates spu­ri­ous leads in data like god­damned magic

  • How to Han­dle Mul­ti­ple-Hy­poth­e­sis-Testing

  • A re­ally good ex­pla­na­tion of MCMC

  • Im­plied “pri­ors,” model as­sump­tions, and how to guess which ones to reach for and screen out the ones that are wrong

  • When is try­ing to add ML to your re­search a good or bad idea

  Things I think we’ve done that seem ap­peal­ing from a re­searcher per­spec­tive in­clude...

  • Some of the stats stuff (I may not re­mem­ber the pre­cise sources, but this is where I picked up an un­der­stand­ing of most of what I listed above)

  • Thought­ful sum­maries/​cri­tiques of cer­tain papers

  • Scott Alexan­der’s stuff on how to thought­fully com­bine mul­ti­ple semi-similar stud­ies (so very much of it, but maybe one in par­tic­u­lar that stood out for me was his bit on the fun­nel graph)

  • I vaguely re­mem­ber see­ing a good ex­pla­na­tion of “how and why to use effect sizes” somewhere

  (...damn, is Scott re­ally car­ry­ing the team here, or is this a per­cep­tion filter and I just re­ally like his blog?)

  Small sam­ple sizes, but I think in the biol­ogy refer­ence class, I’ve seen more peo­ple bounce off of Eliezer’s writ­ing style than the pro­gram­ming refer­ence class does (fairly typ­i­cal “reads-as-ar­ro­gant” stuff; I didn’t per­son­ally bounce off it, so I’m trans­mit­ting this sec­ond­hand). I don’t think there’s any­thing to be done about this; just shar­ing the im­pres­sion. Per­son­ally, I’ve felt mo­ments of an­noy­ance with ran­dom LWers who re­ally don’t have an in­tu­itive feel for the nu­ances for evolu­tion, but Eliezer is ac­tu­ally one of the peo­ple who seems to have a re­ally solid grasp on this par­tic­u­lar topic.

  (I’ve tended to like Elizer’s stuff on statis­tics, and I re­spected him pretty early on be­cause he’s one of the (minor­ity of) peo­ple on here who have a re­ally solid grasp of what evolu­tion is/​isn’t, and what it does/​doesn’t do. Re­spect for his un­der­stand­ing of a field-of-study I did un­der­stand, rubbed off as re­spect­ing him in fields of study he un­der­stood bet­ter than I did (ex: ML) by de­fault, at least un­til my knowl­edge caught up enough that I could rea­son about it on my own.)

  ((FWIW; I sus­pect peo­ple in fi­nance might feel similarly about “Inad­e­quate Equil­ibria,” and I sus­pect they wouldn’t be as turned off by the writ­ing style. They are likely to be de­sir­able re­cruits for other rea­sons: fi­nance at its best is fast-turnaround and ruth­lessly em­piri­cal, it’s of­ten pro­gram­ming or pro­gram­ming-ad­ja­cent, EA is es­sen­tially “char­ity for quan­ti­ta­tively-minded peo­ple who think about black swans,” plus there’s some­thing of a cul­tural fit there.))

  Net­work­ing and ca­reer-de­vel­op­ment-wise… quite frankly, I think we have some, but not a ton to offer biol­o­gists di­rectly. Maybe some EA grants for aca­demics and fu­ture aca­demics that are good at self-ad­vo­cacy and open to mov­ing. I’ve met maybe a dozen ra­tio­nal­ists I could talk heavy bio with, over half of which are pri­mar­ily in some other field at this point. Whereas we have a ton to offer pro­gram­mers, and at ear­lier stages of their ca­reers.

  (I say this par­tially from per­sonal ex­pe­rience, al­though it’s slightly out-of-date: I started my stay in the Berkeley ra­tio­nal­ist com­mu­nity ~4 years ago with a biol­ogy-type de­gree. I had a strong in­ter­est in biorisk, and vi­rol­ogy in par­tic­u­lar. I still switched into pro­gram­ming. There weren’t many re­sources pointed to­wards early-ca­reer peo­ple in bio at the time (this may have changed; a group of bio-minded peo­ple in­clud­ing my­self got a grant to host a group giv­ing pre­sen­ta­tions on this topic, and were re­cently able to get a grant to host a con­fer­ence), and any that ex­isted was pointed at get­ting peo­ple to go to grad school. Given that I had a dis­taste for academia and no in­ten­tion of go­ing to grad school, I even­tu­ally re­al­ized the level of re­sources or sup­port that I could ac­cess around this at the time was effec­tively zero, so I did the ra­tio­nal thing and switched to some­thing that pays well and plugged in with a mas­sive net­work of com­mu­nity sup­port. And yes, I’m a tad bit­ter about this. But that’s par­tially be­cause I just had mis­cal­ibrated ex­pec­ta­tions, which I’m try­ing to help some­one else avoid.)

• Spiracular 22 Sep 2019 1:30 UTC

  6 points

  on: What are some of your “Crazy Ideas” that you’re cur­rently think­ing about?

  Many of the ideas that most alienated me from nor­mal peo­ple are pretty mun­dane here.

  (Years ago, some nor­mal per­son asked me what I thought about what would hap­pen in the fu­ture, in light of over­pop­u­la­tion and the cli­mate crisis. When my re­sponse in­volved “AI-based catas­tro­phe,” “prob­lems cap­i­tal­ism is or isn’t ad­e­quate to solv­ing,” and geo­eng­ineer­ing, they straight up checked out of that con­ver­sa­tion and asked some­body else.)

  So what am I think­ing about that might seem a lit­tle strange even here...

  I’ve ap­par­ently been putting a whole lot of thought in the last cou­ple of months into the ex­tent to which ideal­iza­tion (or the pairing of ideal­iza­tion/​de­mo­niza­tion, which are prob­a­bly differ­ent sides of the same coin given how they turn on a dime) is ut­terly ubiquitous and seems to be ex­tremely bad for good gov­er­nance. Indi­rectly, it strongly in­cen­tivizes those in power to de­velop worse epistemics (cover things up, don’t ask ques­tions, be easy for oth­ers to model) no mat­ter how good they origi­nally were. Now that I’ve started look­ing for it, I keep see­ing ev­i­dence ev­ery­where.

  I’ve gen­tly-but-se­ri­ously con­sid­ered try­ing out a pro­cess loosely based on the one de­scribed in this crazy note­book­ing write-up.

  One be­lief I’ve had for a while, which might be slightly strange in this group, is that I’m not both­er­ing with cry­on­ics. It seems to break down into 2 fac­tors… 1) I be­lieve al­most any post-Sin­gu­lar­ity “hu­mans” will be rad­i­cally differ­ent to be point of not iden­ti­fy­ing or be­ing iden­ti­fi­able as the same thing, and even if “I” do make it to the end, I’ll quickly mod­ify my­self into some­thing neu­roti­cism-free that I can’t in­ter­nally iden­tify with (whether by means of my­self, or AI-im­posed val­ues, the out­come is the same). There­fore, hav­ing my ex­act men­tal con­figu­ra­tion prob­a­bly doesn’t mat­ter too much. 2) Weird “bug” of pri­ori­tiz­ing con­ti­nu­ity only of “pre­dictable ex­ter­nal iden­tity,” while largely de­pri­ori­tiz­ing or treat­ing as free-vari­ables most of my in­ter­nal con­ti­nu­ity that’s in­de­pen­dent from that (in other words… I al­low my­self to make rad­i­cal men­tal shifts, so long as I ex­pect to be able to be­have similarly, con­tinue to serve my fu­ture self, keep my al­lies, and keep my word).

  Prob­a­bly the sin­gle high­est “craz­i­ness in­dex” idea I’ve mul­led on in the past year is whether I wanted to track and see if there’s cor­re­spon­dence be­tween med­i­ta­tional vibra­tions (and where I “feel” them to be; I have a sense of “lo­ca­tion” with them) and Brod­mann Areas (or a similar lo­ca­tion-num­ber­ing sys­tem, so that I can leave my­self par­tially-blind on ini­tially as­sign­ing them). It’s the kind of thing where I ex­pect the origi­nal fram­ing to fail, but I also ex­pect to learn some­thing in­ter­est­ing in the pro­cess. Set­tled on “not worth the effort,” though.

  Most of what I’m think­ing about is prob­a­bly merely ec­cen­tric/​spe­cial-in­ter­est… biol­ogy stuff, metaphor­i­cal cor­re­spon­dences be­tween fi­nan­cial data and ideas from evolu­tion or en­tropy, how I’m us­ing in­tu­itions/​per­cep­tions and get­ting bet­ter at com­mu­ni­cat­ing them clearly...

  (Plus a fairly typ­i­cal hu­man baseline: so­cial, emo­tional, pro­duc­tivity, self-im­prove­ment, iden­tity, fu­ture plan­ning)

• Spiracular 17 Sep 2019 3:03 UTC

  7 points

  on: Spirac­u­lar’s Short­form Feed

  One of my fa­vorite lit­tle tid­bits from work­ing on this post: re­al­iz­ing that idea in­noc­u­la­tion and the Streisand effect are op­po­site sides of the same heuris­tic.

• Spiracular 17 Sep 2019 2:49 UTC

  3 points

  on: Spirac­u­lar’s Short­form Feed

  Bub­bles in Thingspace

  It oc­curred to me re­cently that, by anal­ogy with ML, defi­ni­tions might oc­ca­sion­ally be more like “bound­aries and scor­ing-al­gorithms in thingspace” than clusters per-say (messier! no cen­tral ex­am­ple! no guaran­teed con­ti­guity!). Given the need to co­or­di­nate around defi­ni­tions, most of them are go­ing to have a sim­ple and some­what-mean­ingful cen­ter… but for some words, I sus­pect there are dis­lo­cated “bub­bles” and oddly-shaped “smears” that use the same word for a com­pletely differ­ent con­cept.

  Ho­mo­phones are one of the clear­est ex­am­ples; to­tally dis­con­nected bub­bles of sub­stance.

  Another ex­am­ple is when a word cov­ers all cases ex­cept those where a differ­ent word ap­plies bet­ter; in that case, you can ex­pect to see a “bite” taken out of its space, or even a mul­ti­di­men­sional empty bub­ble, or a dough­nut-like gap in the defi­ni­tion. If the hole is cen­tered (“the strongest cases go by a differ­ent term” ac­tu­ally seems like a very com­mon phe­nomenon), it even makes the idea of a “cen­tral” defi­ni­tion rather mean­ingless, un­less you’re will­ing to fuse or switch terms.

Bioinfohazards

• Spiracular 15 Sep 2019 21:09 UTC

  4 points

  in reply to: Spiracular's comment on: Look­ing for an­swers about quan­tum im­mor­tal­ity.

  Re­lat­edly: I would bet some­one money that Greg Egan does some­thing in­sight-med­i­ta­tion-ad­ja­cent.

  I started read­ing his work af­ter some­one noted my com­men­tary on “the un­sharable­ness of per­sonal qualia” bore a con­sid­er­able re­sem­blance to Closer. And since then, when­ever I read his stuff, I keep see­ing him giv­ing in­tel­li­gent com­men­tary and elab­o­ra­tion on things I had per­ceived and as­so­ci­ated with deep med­i­ta­tion or LSD (the effects are some­times similar for me). He’s ob­vi­ously a big physics fan, but I sus­pect in­sight med­i­ta­tion is an­other one of his big “cre­ativity” gen­er­a­tors. (Be­fore some­one in­evitably asks: No, I don’t say that about ev­ery­thing.)

  To me, Egan’s view­point reads as very athe­ist, but also very Bud­dhist. If you shear off all the woo and dis­till the re­main­der, Bud­dhism is very into see­ing through “illu­sions” (even re­as­sur­ing ones), and he seems to have a par­tic­u­lar in­ter­est in this.

  I can make up a plau­si­ble story that de­vel­op­ing an ob­ses­sion with how we co­or­di­nate-and-man­i­fest the illu­sion of con­ti­nu­ity from dis­parate brain-parts… could be a pretty nat­u­ral side-effect of some­times watch­ing the men­tal sub-pro­cesses that gen­er­ate the illu­sion of “a sin­gle, con­scious, con­tin­u­ous self” fall apart from one an­other? (Med­i­ta­tion can do that, and it’s very un­set­tling the first time you see it.).

• Spiracular 15 Sep 2019 21:07 UTC

  4 points

  in reply to: Spiracular's comment on: Look­ing for an­swers about quan­tum im­mor­tal­ity.

  So, here’s the spe­cific thing I can think of that seems like it might be helpful...

  I try to be cau­tious about us­ing med­i­ta­tion-based wire-head­ing or emo­tional-dul­ling, but at min­i­mum, there’s a state one step down from en­light­en­ment (equa­nim­ity) that per­ceives suffer­ing as merely “dis­so­nance” in vibra­tions. The judg­ing/​nega­tive-con­no­ta­tion gets dropped, and in­ter­nal-per­cep­tion of emo­tional af­fect is pretty flat (Note of cau­tion: the emo­tions prob­a­bly aren’t gone, it’s more like you per­ceive them differ­ently. I’m not 100% sure how it works, my­self. While it might sound similar, it’s not quite the same as dis­so­ci­a­tion; the move­ment is more like you lean into your ex­pe­rience rather than out of it. Also, I read in a pa­per that its painkil­ler prop­er­ties are ap­par­ently not based on opi­ods? Weird, right? So neu­rolog­i­cally, I don’t re­ally know how it works, al­though I might de­velop the­o­ries if I re­searched it a bit harder.).

  En­light­en­ment/​fruition proper doesn’t even form mem­o­ries, al­though I’ve never been able to sus­tain that state for longer than a few sec­onds. But when it drops, it usu­ally drops back into equa­nim­ity… so I guess be­tween the two, it’d be a se­ri­ous im­prove­ment on “eter­nal con­scious suffer­ing”?

  Un­for­tu­nately, to get into En­light­en­ment ter­ri­tory, there’s a se­ries of in­ter­me­di­ate steps that tend to set off ex­is­ten­tial crises, of widely-vary­ing sever­ity. Any book or teacher that doesn’t take this and the wire­head­ing po­ten­tial se­ri­ously, is prob­a­bly less good than one who does. That said, I still recom­mend it, es­pe­cially for peo­ple who seem to keep hav­ing ex­is­ten­tial crises any­way. But it’s a per­cep­tion-al­ter­a­tion work­bench; its sub-skills can some­times be used to detri­men­tal ends, if peo­ple aren’t care­ful about what they in­stall.

• Spiracular 15 Sep 2019 21:06 UTC

  2 points

  in reply to: Spiracular's comment on: Look­ing for an­swers about quan­tum im­mor­tal­ity.

  Here’s one plus-side that you don’t need the ad­di­tional con­text to un­der­stand: I kinda sus­pect that at least most peo­ple would even­tu­ally find the right com­bi­na­tion of in­sights and ex­is­ten­tial-crises to bum­ble into en­light­en­ment by them­selves, if they had an eter­nity of con­sec­u­tive ex­pe­riences to work with. Espe­cially given that there seem to be mul­ti­ple sim­ple prac­tices that get around to it even­tu­ally (al­though it might take a cou­ple of life­times for some peo­ple).

• Spiracular 15 Sep 2019 21:06 UTC

  2 points

  in reply to: gilch's comment on: Look­ing for an­swers about quan­tum im­mor­tal­ity.

  As some­one who has had stream-en­try, and the change-in-per­cep­tion called En­light­en­ment… I en­dorse your read of it as be­ing po­ten­tially use­ful in this case?

  I’m go­ing to give more de­tails in a sub-com­ment, to give peo­ple who are already rol­ling their eyes a chance to skip over this.

• Spiracular 12 Sep 2019 18:59 UTC

  6 points

  on: Spirac­u­lar’s Short­form Feed

  While I could rat­tle off the benefits of “del­e­gat­ing” or “ops peo­ple”, I don’t think I’ve seen a highly-dis­crete TAP + flowchart for re­al­iz­ing when you’re at the point where you should ask your­self “Have my easy-to-del­e­gate an­noy­ances added up to enough that I should hire a full-time ops per­son? (or more).”

  Many peo­ple whose time is valuable seem likely to put off mak­ing this call un­til they reach the point where it’s glar­ingly ob­vi­ous. Propos­ing an easy TAP-like de­ci­sion-bound­ary seems like a po­ten­tially high-value post? Not my area of spe­cialty, though.

• Spiracular 11 Sep 2019 2:57 UTC

  15 points

  on: Bioinfohazards

  Now that we’ve gone over some of the con­sid­er­a­tions, here’s some of the con­crete top­ics I see as gen­er­ally high or low haz­ard for open dis­cus­sion.

  Good for Open Discussion

  • Broad-ap­pli­ca­tion an­tiviral de­vel­op­ments and meth­ods

    • Vaccines

    • An­tivirals proper

    • T-cell therapy

    • Virus de­tec­tion and monitoring

  • How to re­port lab haz­ards

    • ...and how to nor­mal­ize and en­courage this

  • Broadly-ap­pli­ca­ble pro­tec­tive mea­sures

    • Sanitation

    • Bunkers?

  • The state of funding

  • The state of tal­ent

    • What broad skills to develop

    • How to ap­peal to talent

    • Who tal­ent should talk to

  Bad for Open Discussion

  Th­ese things may be worth spe­cial­ists dis­cussing among them­selves, but are likely to do more harm than good in an open thread.

  • Disease de­liv­ery methods

  • Spe­cific Threats

  • Spe­cific Ex­ploitable Flaws in Defense Sys­tems

    • Ex: im­mune sys­tems, hos­pi­tal mon­i­tor­ing systems

    • It is es­pe­cially bad to men­tion them if they are ex­ploitable reliably

    • If you are si­mul­ta­neously pro­vid­ing a com­pre­hen­sive solu­tion to the prob­lem, this can be­come more of a gray-area. Par­tial-solu­tions, or challeng­ing-to-im­ple­ment solu­tions, are likely to fall on the bad side of this equa­tion.

  • Much of the syn­thetic biol­ogy sur­round­ing this topic

  • Ar­gu­ments for and against var­i­ous agents us­ing dis­ease as an M.O.

• Spiracular 7 Sep 2019 20:33 UTC

  6 points

  on: What are the biggest “moon­shots” cur­rently in progress?

  Warn­ing: Big pile of text, De­com­pos­ing definitions

  I no­tice that there’s a lot of dis­putably-rele­vant axes for as­sess­ing if some­thing is a moon­shot, and that was mak­ing me hes­i­tant to an­swer. So… I’m go­ing to be “that guy” who deep-dives defin­ing terms. Hope­fully this will be con­struc­tive?

  Here are some dis­putable axes for as­sess­ing “moon­shots” that popped to mind:

  • How pos­si­ble is it to profit off of in­cre­men­tal progress on this pro­ject’s sub-goals?

    • At the limit: Does it have to be com­pletely un­prof­itable un­til you win, af­ter which there is a steep, step-like func­tion af­ter which there are mas­sive re­turns?

      • But even space wasn’t this ex­treme! There were rock­ets and mis­siles be­fore there were rocket ships. And my loose un­der­stand­ing of the way Elon Musk is do­ing things is about as in­cre­men­tal as you can make space rock­ets and still have it be rocket sci­ence (ad­mit­tedly: not very). Those many fancy land­ing-con­trol tests don’t make the fi­nal launch not a moon­shot, though.

    • At lower lev­els: At what point is this just the kind of “prof­itable-in­cre­men­tal-progress via se­lec­tion” al­gorithm that cap­i­tal­ism en­courages and sup­ports just fine?

  • How se­vere was the pro­jects “start-up cost”? How much hard-to-con­soli­date in­fras­truc­ture, in­tel­li­gence, data, and re­sources were re­quired be­fore this pro­ject was even con­ceiv­able?

  • For one rea­son or an­other, is the most plau­si­ble coun­ter­fac­tual that if this one group wasn’t do­ing this, no group would be do­ing this? Or would be do­ing this much more poorly?

  • Clas­si­cally, megapro­jects re­fer to large-scale works of ar­chi­tec­tural in­fras­truc­ture, but we live in an in­for­ma­tion age. Do mas­sive broadly-benefi­cial pro­jects of in­for­ma­tional in­fras­truc­ture count, or not?

  • How much of hu­man­ity needs to be im­pacted by the out­come of this pro­ject? Does the out­come need to be benefi­cial? Or is this re­ally about af­fect­ing the nar­ra­tive hu­man­ity has about it­self, and not the effect it has on the day-to-day way that peo­ple ex­pe­rience their lives? (ex: space pro­jects)

  (And… in­ter­est­ing! ryan_b se­lected a al­most com­pletely differ­ent set of axes. Maybe it’s not a par­tic­u­larly con­sis­tently-defined con­cept?)

  I feel con­flicted about in­clud­ing any of these as re­quire­ments (sev­eral don’t ac­tu­ally seem that de­sir­able), but I think the nor­mal way moon­shots are thought of and defined tends to cen­ter around anal­ogy to the Space Race. And there­fore, tends to in­volve al­most all of these fea­tures be­ing pre­sent.

  “No in­cre­men­tal progress mea­sures” seems like a par­tic­u­larly key part of the defi­ni­tion, and yet a po­ten­tially nega­tive thing to filter by. When­ever you can, good in­cre­men­tal progress as­sess­ment is usu­ally a pos­i­tive thing to add to a pro­ject.

  Thought ex­per­i­ment: If you had to com­pare 2 iden­ti­cal cold-fu­sion pro­jects, one of which came up with a bunch of in­ter­me­di­ate steps and tested them, and one which didn’t and just had one big “did you get ev­ery­thing right?” as­sess­ment right at the end… which one is the moon­shot? But which one is prob­a­bly the bet­ter pro­ject?

  Un­der this lens, there’s a pretty im­por­tant dis­tinc­tion to be made be­tween things that are be­ing treated as moon­shots, and prob­lems that have to be ap­proached as moon­shots.

  Maybe the right ques­tion is… “What are moon­shot prob­lems that some­one is se­ri­ously tack­ling?” Or just drop­ping the moon­shots fram­ing, and get­ting a list of in­ter­est­ing megapro­jects. Or just pick­ing the 1-2 axes you most care about, and sort­ing on them ex­plic­itly.

  (FWIW; any one of those is a valid thing to want, and it’s a good ques­tion! That is part of why I put in the effort to try to break it down.)

  P.S. Is there some way I should have used the Ques­tion-on-a-Ques­tion /​ Re­lated Ques­tion func­tion to do this? If so, could some­one walk me through how that’s sup­posed to work?

• Spiracular 7 Sep 2019 20:21 UTC

  6 points

  on: What are the biggest “moon­shots” cur­rently in progress?

  It’s not quite moon­shots, but here’s wikipe­dia’s list of Me­gapro­jects.

  When it comes to more clas­sic ar­chi­tec­tural in­fras­truc­ture pro­jects, I’d be shocked if China (and maybe Sin­ga­pore) don’t have sev­eral. But I also ex­pect them to lean to­wards the in­cre­men­tal-progress-is-visi­ble model where they can. Be­cause when you can, hav­ing as­sess­able in­cre­men­tal progress re­ally is al­most always the bet­ter choice?

  Se­ri­ous New Physics seems to be full of night­mare-to-fund megapro­jects. What’s that up­com­ing telescope that’s go­ing to be able to test if weird ex­tra­so­lar as­ter­oids like ’Ou­mua­mua pass­ing by Earth are a com­mon or rare event? Some­body’s got to be build­ing a new and more pow­er­ful col­lider, some­where in the world. Who’s work­ing on fu­sion re­ac­tors? Who’s work­ing on craft­ing quan­tum com­put­ers?

  Think­ing of megapro­jects & moon­shots got me think­ing of some in­for­ma­tion-in­fras­truc­ture stuff that might or might not count, de­pend­ing on defi­ni­tion. So I’m go­ing to throw on a list of some ten­ta­tive and dis­putable “In­for­ma­tion Age Me­gapro­jects” too...

  • Wikipe­dia

    • Some­one back-of-en­velope es­ti­mated that Wikipe­dia was the cul­mi­na­tion of 100 mil hours of work in 2008, and as­suredly more by to­day.

    • While it’s suc­cess feels pre­des­tined nowa­days, when it first out­did for­mal and pri­vately-funded at­tempts at on­line en­cy­clo­pe­dias, that was a very sur­pris­ing out­come to many

  • If you loosen the re­stric­tions, biol­ogy prob­a­bly has tons of megapro­jects. (BLAST, GenBank, PubChem… NIH is fund­ing and main­tain­ing vast databases, and all the in­fras­truc­ture that makes them nav­i­gable. How much work that is prob­a­bly de­pends on whether you in­clude or ex­clude the work it took to ex­tract the data they’re ped­dling.)

    • It feels al­most-in­co­her­ent for me to think of biolog­i­cal pro­jects as “moon­shots”? There are in­ter­est­ing and novel things be­ing done con­stantly (Many of them clever, ir­re­pla­ca­ble things. Some of them fun­da­men­tal things! Some of them po­ten­tially high-im­pact!), and yet the term just does not seem to fit. Just about ev­ery in­ter­est­ing syn­thetic biol­ogy pro­ject nowa­days has a “merely” mod­er­ately-large to large start-up cost, a mea­sure of in­cre­men­tal progress, and a huge el­e­ment of chance. Some of those could have huge im­pacts if they work. Should I call those moon­shots? They’re not re­ally big enough for “megapro­jects” to fit, cost-wise.

  • Google (Alpha­bet? GoogleX?) seems to have a taste for this

    • Google’s book-scan­ning pro­ject (be­fore it was shut down)

    • Loon is their in­ter­net-ac­cess bal­loon project

    • The An­droid Oper­at­ing Sys­tem?

    • The Google search-en­g­ine it­self?

  • The In­ter­net Archive?

  • Peo­ple build­ing 3D mod­els of en­tire cities, in games such as Se­cond Life?

  • Does Bit­coin count?

    • I sus­pect not, but I couldn’t re­ally ar­tic­u­late why. If I wanted to ar­gue one way or an­other with this one, I ac­tu­ally wouldn’t know where to start.

• Spiracular 7 Sep 2019 20:14 UTC

  1 point

  in reply to: Liron's comment on: What are the biggest “moon­shots” cur­rently in progress?

  I’m used to think­ing of those more as “Me­gapro­jects”. “A List of Me­gapro­jects” is a great goal, but I feel like it might be worth clar­ify­ing that in the ques­tion up-top a lit­tle bit.

• Spiracular 25 Aug 2019 18:53 UTC

  2 points

  in reply to: Kenny's comment on: Paper Trauma

  Sin­gle dat­a­point, but… I love mark­down for notes. The for­mat­ting short­hands are great, and an au­to­mat­i­cally-gen­er­ated Table of Con­tents is a nice plus.

  It hasn’t been easy for me to find any­thing that offered both flow-chart di­a­gram­ming and cloud-sync, though. And both mark­down and HTML are baf­flingly awk­ward to make ta­bles with (Why can’t some­body just set up a painless csv wrap­per? Or if that ex­ists, please tell me?).

  Most de­cent mark­down ed­i­tors are also LaTeX ca­pa­ble, which I prob­a­bly use even more than di­a­grams. There’s a bit of a learn­ing-curve for that, but nowa­days I can type out most equa­tions with­out so much as look­ing at the key­board (let alone look­ing up sym­bols).

• Spiracular 14 Aug 2019 7:02 UTC

  2 points

  in reply to: ChristianKl's comment on: Why is the ni­tro­gen cy­cle so un­der-em­pha­sized com­pared to cli­mate change

  But when the ar­gu­ment for the al­ter­na­tive boils down to “eat shit, not chem­i­cals”...

  I’m kid­ding, but only slightly. Or­ganic fer­til­iza­tion is a bit gross, and I think most food com­pa­nies would pre­fer to not be as­so­ci­ated with any of dead leaf mat­ter, rot­ting lef­tovers, or ma­nure.

  Coun­ter­ar­gu­ment: Com­post­ing to­tally be­came a thing, and that po­ten­tially puts the gross­ness right in your back­yard.

  (Huh. Com­post­ing is cer­tainly some­thing peo­ple do on an in­di­vi­d­ual level to micro-com­bat the us­age of ni­tro­gen fer­til­iz­ers, with prob­a­bly a very neg­ligible effect. And some peo­ple do seem ded­i­cated to it. But I sus­pect that if you asked most peo­ple who do it, they would claim it’s about land­fills or some­thing, not soil ni­tro­gen con­tent.)

• Spiracular 14 Aug 2019 5:39 UTC

  24 points

  on: Why do hu­mans not have built-in neu­ral i/​o chan­nels?

  Even if it had evolved, any de­tailed form of com­mu­ni­ca­tion that had the po­ten­tial to trans­mit hard-to-break im­per­a­tives is some­thing you want to be very, very care­ful with.

  Defec­tion, ma­nipu­la­tion, and novel av­enues for dis­ease-trans­mis­sion or par­a­sitism heav­ily dis­in­cen­tivize this. It’s in­tu­itively “gross” for a rea­son.

  TL;DR: In­fec­tions and defec­tions would prob­a­bly ut­terly wreck this. The blood-brain bar­rier ex­ists for a rea­son. While we did get lan­guage, in other ways we’ve evolved speci­fi­cally to hide in­for­ma­tion from each other; it’s not straight­for­wardly evolu­tion­ar­ily fa­vored. Large clusters of highly-re­lated or­ganisms have more in­cen­tive to do this (bac­te­rial mats, ants, our own cells, etc.), and the in­for­ma­tion-band­width they share with each other through pheromones and chem­i­cal sig­nals is ac­tu­ally pretty stag­ger­ing. But at a glance, I do think they pay a cost in in­creased (and more elab­o­rate) av­enues for ma­nipu­la­tions and in­fec­tions to reap the benefits of this priv­ilege.

  Edit to add: Link­ing some ad­di­tional strongly-re­lated ar­ti­cles! SSC’s Maybe Your Zoloft Stopped Work­ing Be­cause a Liver Fluke Tried To Turn Your Nth-Great-Grand­mother into a Zom­bie and the pa­per it cen­ters on, In­visi­ble De­sign­ers: Brain Evolu­tion Through the Lens of Par­a­site Manipulation

  Iso­lat­ing Brains from Infection

  It bears point­ing out that evolu­tion usu­ally seems to want the op­po­site of this for our cen­tral­ized-de­ci­sion-maker or­gans. No­tice that in­stead of mak­ing the brain more open-ac­cess over time, most species went the di­rec­tion of mak­ing it iso­lated from even our own bod­ies, us­ing things like the heavy-fil­ter­ing brain/​blood bar­rier. The risk of that sen­si­tive or­gan get­ting poi­soned, dis­eased, or biolog­i­cally ma­nipu­lated was just too high to risk it.

  (Ne­ma­todes make what hu­mans call “em­bod­ied think­ing” look like a joke; the sero­tonin from their di­ges­tive tract is felt by their brains di­rectly.)

  Peo­ple used to die in droves (and at young ages!) of measles, tu­ber­cu­lo­sis, and a mil­lion other things. Even be­fore cities, herd-liv­ing put us un­der im­mense pres­sure to de­velop a pretty in­tensely spe­cial­ized im­mune sys­tem. If we had to worry about giv­ing a dis­ease a high­way to our cen­tral ner­vous sys­tem, that would be… very, very bad. It might even guaran­tee that such a species would never get to cen­tral­ize such things at all, such a force is the risk of in­fec­tion.

  Not to even get into the pos­si­bil­ity of phys­i­cal brain-hi­jack­ings by con­cepts (like memes, but oh-so-much-worse!), or even just catch­ing com­mu­ni­ca­tion-trans­mit­ted Kuru… but here’s a pretty vivid spec­u­la­tive de­scrip­tion of just how bad be­ing an evolved “open book” that granted oth­ers write-ac­cess would prob­a­bly get.

  And with re­gards to the “benefits” of open com­mu­ni­ca­tion -of in­for­ma­tion con­veyed in a lan­guage that’s very hard to fake- we do still have some in­for­ma­tion trans­mit­ted in body-lan­guage and words. That cer­tainly cap­tures some of the benefit. But it bears point­ing out that we’re a species that un-evolved any ob­vi­ous pre­sen­ta­tion of whether a fe­male is in es­trus, and has very strong in­hi­bi­tions around try­ing not to gain in­for­ma­tion from each other’s body odor. “Com­plete, to­tal hon­esty” is not some­thing evolu­tion typ­i­cally se­lects for, and it didn’t veer en­tirely that way for us. Even in the less-cut­throat mod­ern era (at least, com­pared to our dis­tant sa­van­nah past), Greg Egan’s Closer feels like a pretty re­al­is­tic de­pic­tion of how we might feel about it if we ever did fully share our men­tal ex­pe­rience with even one an­other per­son. I’ll avoid spoiling it too badly, but we’d prob­a­bly quickly un­cover a lot of things about one other that we wished we didn’t know.

  Some Liv­ing Approximates

  Bac­te­rial mats, gi­ant net­works of fungi, and eu­so­cial in­sects with strong ge­netic kin­ship might have strong enough evolu­tion­ary in­cen­tives for this to line up, al­though higher re­lat­ed­ness ac­tu­ally ex­ac­er­bates the in­fec­tion con­cern. And be­tween the cells of mul­ti­cel­lu­lar creat­ues, cer­tainly quite a lot does get com­mu­ni­cated. Many of these ex­am­ples do seem to “share their mind” in at least some mean­ingful sense. They trans­mit a lot of in­for­ma­tion and or­ders to one an­other, and have a com­mu­nal de­ci­sion-mak­ing pro­cess at vary­ing de­grees of cen­tral­iza­tion/​de­cen­tral­iza­tion. Pheromones for in­sects, var­i­ous sig­nal­ling se­cre­tions from bac­te­ria, the oo­dles of trans­ac­tions and de­liv­er­ies be­tween our cells at ev­ery mo­ment… com­bined these can be very high-band­width. Al­most in­com­pre­hen­si­bly so, if you’ve ever seen at­tempts to mea­sure and chart such things.

  Ants could prac­ti­cally be said to have a pheromone “lan­guage,” com­plete with clan-iden­ti­fi­ca­tion tags. And as a way to se­lec­tively trig­ger a highly-spe­cific neu­ral path­way, or ac­ti­vate a known set of be­hav­iors in a con­spe­cific with the same brain-con­figu­ra­tion, pheromones are not a bad way to go? The be­hav­ior pat­terns pheromones set off can get oddly spe­cific at times.

  And… ants also get tricked by pheromones into feed­ing the brood par­a­sites that eat their own young. And what we call “bac­te­rial sex” (high-band­width com­mu­ni­ca­tion of DNA?) is ac­tu­ally virus-es­que plas­mids try­ing to trans­mit them­selves to new bac­te­ria, like an in­fec­tion. Some plas­mids might even come with ad­dic­tion molecules, which is an ex­tra-douchey way for an plas­mid to con­vey “repli­cate me, or die.” And in co­or­di­nated bac­te­ria, you do some­times see defec­tors. So… it’s still pretty ma­nipu­la­ble, and it sure gets ma­nipu­lated.

  The more stereo­typed be­hav­iors you can set off through ex­ter­nal sig­nals, the more you have a “broader at­tack sur­face,” in the cy­ber­se­cu­rity lingo. And biolog­i­cal par­a­sitism is ubiquitous, and frac­tal­ine, and adap­tive, and uses any damn at­tack sur­face it can get.

  Hu­mans? A fluke. Par­a­sites are evolu­tion’s true dar­lings.

• Spiracular 7 Aug 2019 22:20 UTC

  9 points

  on: Why is the ni­tro­gen cy­cle so un­der-em­pha­sized com­pared to cli­mate change

  I thought this was an in­ter­est­ing ques­tion… al­though I definitely get the feel­ing like I’m miss­ing some of the con­text be­hind this “plane­tary bound­aries” write-up.

  (What is the Stock­holm Re­silience Cen­ter? What are its mo­tives and meth­ods? Why was it do­ing this anal­y­sis, and how did you end up run­ning into it?)

  I agree that fer­til­izer runoff gets talked about a lot less than cli­mate change, and I’m not en­tirely sure of why that is. I just looked it up, and “Or­ganic”-la­beled things ap­par­ently do already man­date or­ganic fer­til­iz­ers (which should be N-neu­tral on net?). So there’s at least that.

  Re­gard­ing their as­sess­ment… one of the fac­tors that seemed to be weighted a lot in their as­sess­ment was “level of antro­pogenic change vs nat­u­ral vari­a­tion in Pro­cess X.” I’d ex­pect that to have heav­ily-weighted the Nitro­gen Cy­cle, since our in­terfer­ence in it dwarves the vari­abil­ity due to nat­u­ral pro­cesses (some­thing they them­selves spell out. The Haber pro­cess is a strange, pow­er­ful, mag­i­cal, in­or­ganic thing the hu­mans cooked up).

  This met­ric is… not pre­cisely the same as “level of dam­age this change can cause.” They seem to have set up some sort of thresh­old for what they con­sider “dan­ger­ously high”, and I don’t re­ally un­der­stand the thought-pro­cess or rea­son­ing they used in pick­ing those thresh­olds. But I think they fac­tored “change vs nat­u­ral vari­a­tion” into their think­ing a good deal.

  My un­der­stand­ing from my own Ag back­ground is that fer­til­izer runoff re­ally can be a big prob­lem; ni­tro­gen and phos­pho­rus are a ma­jor limit­ing nu­tri­ent for plant-life, both on land and in fresh­wa­ter ecosys­tems (the ocean sur­face is iron-limited, in­ter­est­ingly). In a sense, this is ex­actly why we use it; we want our food crops to grow at a wild rate, one rarely seen in na­ture.

  When there’s a sud­den in­flux of ni­tro­gen into a fresh­wa­ter sys­tem, one of the pos­si­ble con­se­quences is an al­gal bloom. Th­ese go through a boom-and-bust cy­cle (with the sea­sons, or re­source-availa­bil­ity pat­terns), lead­ing to a mas­sive al­gal die-off. Dur­ing this die-off, de­cay bac­te­ria wipe out the un­der­wa­ter oxy­gen-sup­ply, lead­ing to knock-on effects on fresh­wa­ter ecosys­tems like mas­sive fish die-offs. En­closed spaces like lakes are per­haps es­pe­cially vuln­er­a­ble, since there is no way for the N and P to ever exit the sys­tem, po­ten­tially per­pet­u­at­ing the cy­cle in­definitely.

  That said, sud­denly cut­ting fer­til­izer us­age would prac­ti­cally en­sure that food pro­duc­tion would sud­denly drop way down, and that would ill-serve a lot of other hu­man val­ues. Re­duc­ing runoff looks like a very hard prob­lem, to me. Find­ing ways to re­move ex­cess P and N from en­vi­ron­ment seemed to be an area with at least a lit­tle bit of in­ter­est, but it doesn’t seem very ac­tion­able on an in­di­vi­d­ual rather than city or state level, which might ex­plain the low pub­lic­ity? Un­sure.