jimrandomh

• jimrandomh 12 Nov 2019 19:14 UTC

  2 points

  in reply to: jkaufman's comment on: At­tach Re­ceipts to Credit Card Transactions

  Right, they would cer­tainly do it if you paid them enough (and low­er­ing the fee is a form of pay­ment); this is a rea­son why the price would be higher.

• jimrandomh 12 Nov 2019 19:10 UTC

  7 points

  on: Ban the Lon­don Mulligan

  It sounds like the prob­lem is that mul­li­gans are nec­es­sary to en­sure there’s a game to play, which de­pends mostly on hav­ing a rea­son­able num­ber of lands, but that they have bad side-effects which are mostly the re­sult of giv­ing too much con­trol over which non­land cards you have. So I pro­pose the fol­low­ing mul­li­gan rule:

  • On your first mul­li­gan, draw five, then choose one: draw a card, or search your library for a ba­sic land card, re­veal it, and put it into your hand.

  • On your sec­ond mul­li­gan, draw three, then choose as be­fore, twice.

  • On your third mul­li­gan, draw one, then choose as be­fore three times.

  • There is no fourth mul­li­gan.

  This makes mul­li­gans much bet­ter at en­sur­ing you can play, and much worse at en­sur­ing you can find a par­tic­u­lar card or combo that you’re look­ing for.

  (For com­plex­ity rea­sons, this rule would work bet­ter if there were a key­word for “ei­ther draw or fetch a land”, and if it were in­tro­duced in ad­vance.)

• jimrandomh 12 Nov 2019 17:19 UTC

  2 points

  on: At­tach Re­ceipts to Credit Card Transactions

  Stores don’t want to do this for the same rea­son they make prices that change fre­quently, are one cent off from round num­bers, and are in the least leg­ible font that is legally per­mis­si­ble. They want pay­ing at­ten­tion to prices to be in­con­ve­nient, be­cause pay­ing at­ten­tion de­creases spend­ing and shifts that spend­ing to­wards lower mar­gin items.

• jimrandomh 2 Nov 2019 6:25 UTC

  22 points

  on: How do you as­sess the qual­ity /​ re­li­a­bil­ity of a sci­en­tific study?

  1. For health-re­lated re­search, one of the main failure modes I’ve ob­served when peo­ple I know try to do this, is tun­nel vi­sion and a lack of pri­ors about what’s com­mon and rele­vant. Read­ing raw re­search pa­pers be­fore you’ve read broad-overview stuff will make this worse, so read UpToDate first and Wikipe­dia sec­ond. If you must read raw re­search pa­pers, find them with PubMed, but do this only rarely and only with a spe­cific ques­tion in mind.

  2. Be­fore look­ing at the study it­self, check how you got there. If you ar­rived via a search en­g­ine query that asked a ques­tion or posed a topic with­out pre­sup­pos­ing an an­swer, that’s good; if there are mul­ti­ple stud­ies that say differ­ent things, you’ve sam­pled one of them at ran­dom. If you ar­rived via a query that asked for con­fir­ma­tion of a hy­poth­e­sis, that’s bad; if there are mul­ti­ple stud­ies that said differ­ent things, you’ve sam­pled in a way that was bi­ased to­wards that hy­poth­e­sis. If you ar­rived via a news ar­ti­cle, that’s the worst; if there are mul­ti­ple stud­ies that said differ­ent things, you sam­pled in a way that was bi­ased op­po­site re­al­ity.

  3. Don’t bother with stud­ies in ro­dents, an­i­mals smaller than ro­dents, cell cul­tures, or un­der­grad­u­ate psy­chol­ogy stu­dents. Th­ese stud­ies are done in great num­bers be­cause they are cheap, but they have low av­er­age qual­ity. The fact that they are so nu­mer­ous makes the search-sam­pling prob­lems in (2) more se­vere.

  4. Think about what a sen­si­ble end­point or met­ric would be be­fore you look at what end­point/​met­ric was re­ported. If the re­ported met­ric is not the met­ric you ex­pected, this will of­ten be be­cause the rele­vant met­ric was ter­rible. Clas­sic ex­am­ples are pa­pers about bat­tery tech­nolo­gies re­port­ing power rather than ca­pac­ity, biomed­i­cal pa­pers re­port­ing effects on bio­mark­ers rather than symp­toms or mor­tal­ity.

  5. Cor­rectly con­trol­ling for con­founders is much, much harder than peo­ple typ­i­cally give it credit for. Ad­ding ex­tra things to the list of things con­trol­led for can cre­ate spu­ri­ous cor­re­la­tions, and study au­thors are not in­cen­tivized to han­dle this cor­rectly. The prac­ti­cal up­shot is that ob­ser­va­tional stud­ies only count if the effect size is very large.

• jimrandomh 30 Oct 2019 18:59 UTC

  3 points

  in reply to: habryka's comment on: [Site Up­date] Sub­scrip­tions, Book­marks, & Pingbacks

  (We for­got to run one of the mi­gra­tion scripts)

• jimrandomh 30 Oct 2019 2:17 UTC

  10 points

  in reply to: G Gordon Worley III's comment on: [Site Up­date] Sub­scrip­tions, Book­marks, & Pingbacks

  Same ca­dence, but sep­a­rat­ing them does make sense and I might add that op­tion in the fu­ture.

• jimrandomh 29 Oct 2019 3:07 UTC

  9 points

  on: Cli­mate tech­nol­ogy primer (1/​3): basics

  Nu­clear power is typ­i­cally lo­cated close to power de­mands, ie cities, be­cause of the costs and losses in trans­port­ing power over long dis­tances. This also limits the size/​scale of nu­clear power plants, since if you build larger than the de­mands of a city, you have to trans­port the power over a long dis­tance to find more sinks.

  On the other hand, sup­pose a city were built speci­fi­cally for the pur­pose of host­ing nu­clear power, car­bon-cap­ture, and CO2-to-fuel plants. Such a city might be able to have sig­nifi­cantly cheaper nu­clear power, since be­ing far away from ex­ist­ing pop­u­la­tion cen­ters would lower safety and reg­u­la­tory costs, and con­cen­trat­ing pro­duc­tion in one place might en­able new economies of scale.

  It seems like there are two wor­lds we could be in, here. In one world, nu­clear power right now is like the space launch in­dus­try of a decade ago: very ex­pen­sive, but ex­pen­sive be­cause of in­sti­tu­tional failure and a need for R&D, rather than fun­da­men­tal physics. In the other world, some com­po­nent of power plants (steam tur­bines, for ex­am­ple) is already op­ti­mized close to rea­son­able limits, so an or­der of mag­ni­tude is not pos­si­ble. Does any­one with en­g­ineer­ing knowl­edge of this space have a sense of which is likely?

• jimrandomh 29 Oct 2019 2:11 UTC

  7 points

  on: Cli­mate tech­nol­ogy primer (1/​3): basics

  Some­thing that con­fuses me. This (and also other) dis­cus­sions of af­foresta­tion fo­cus on plant­ing trees (con­vert­ing non-for­est biomes into for­est). But it seems like trees do a rea­son­ably good job of spread­ing them­selves; why not in­stead go to ex­ist­ing forests, cut down and bury any trees that are past their peak growth phase, and let the re­main­ing trees plant the re­place­ments? How much does the cut­ting and burial it­self cost? (I see an es­ti­mate of $50/​t for bury­ing crop resi­dues, which seems like it would be similar.)

• jimrandomh 29 Oct 2019 1:37 UTC

  13 points

  on: Cli­mate tech­nol­ogy primer (1/​3): basics

  I ex­pect to have some ques­tions and ideas, but I’m still work­ing my way through this, as I sus­pect are oth­ers. I re­ally ap­pre­ci­ate how in depth this is!

• jimrandomh 27 Oct 2019 3:56 UTC

  2 points

  in reply to: MakoYass's comment on: Jacy Reese (born Jacy An­this)?

  Un­der Wikipe­dia’s rules, yes.

• jimrandomh 20 Oct 2019 19:23 UTC

  13 points

  on: En­doge­nous Epinephrine for Ana­phy­laxis?

  I ex­pect that know­ing you’re hav­ing ana­phy­laxis with­out a solu­tion is already rea­son­ably close to the up­per end of psy­cholog­i­cal stress, and you can’t add that much more. The rea­son the epinephrine con­cen­tra­tions are so much higher in car­diac ar­rest pa­tients is not be­cause car­diac ar­rest is psy­cholog­i­cally stress­ful, it’s be­cause epinephrine re­lease is trig­gered by hy­poxia.

• jimrandomh 5 Oct 2019 2:30 UTC

  4 points

  on: Notic­ing Frame Differences

  Cu­rated. I think the meta-frame (frame of look­ing at frames) is key to figur­ing out a lot of im­por­tant out­stand­ing ques­tions. In par­tic­u­lar, some ideas are hard to parse or gen­er­ate in some frames and easy to parse or gen­er­ate in oth­ers. Most peo­ple have frame-blind-spots, and lose ac­cess to some ideas that way. There also seem to be groups within the ra­tio­nal­ity com­mu­nity that are feel­ing alienated, be­cause too many of their con­ver­sa­tions suffer from frame-mis­match prob­lems.

• jimrandomh 28 Sep 2019 18:58 UTC

  27 points

  in reply to: shminux's comment on: Fol­low-Up to Petrov Day, 2019

  Lizard­man’s Con­stant is an ob­ser­va­tion seen in polls of un­filtered groups of peo­ple, but the peo­ple who were given the launch codes were se­lected for trust­wor­thi­ness.

• jimrandomh 27 Sep 2019 2:56 UTC

  42 points

  in reply to: jimrandomh's comment on: Honor­ing Petrov Day on LessWrong, in 2019

  (This thread is our col­lec­tive reen­act­ment of the con­ver­sa­tions about nu­clear safety that hap­pened dur­ing the cold war.)

• jimrandomh 27 Sep 2019 1:27 UTC

  7 points

  in reply to: Ruby's comment on: Honor­ing Petrov Day on LessWrong, in 2019

  
  
  
  
  
  

  ?

  
  

  !

• jimrandomh 19 Sep 2019 1:23 UTC

  4 points

  in reply to: sayan's comment on: Sayan’s Braindump

  • Mul­ti­ple large mon­i­tors, for pro­gram­ming.

  • Water­proof pa­per in the shower, for col­lect­ing thoughts and mak­ing a morn­ing todo list

  • Email filters and Pri­or­ity In­box, to pre­vent spu­ri­ous in­ter­rup­tions while keep­ing enough trust that ur­gent things will gen­er­ate no­tifi­ca­tions, that I don’t feel com­pel­led to check too often

  • USB bat­ter­ies for recharg­ing phones—one to carry around, one at each charg­ing spot for quick-swapping

• jimrandomh 16 Sep 2019 20:07 UTC

  7 points

  in reply to: jkaufman's comment on: Focus

  Yep, one of us ed­ited it to fix the link. Added a GitHub is­sue for deal­ing with rel­a­tive links in RSS in gen­eral: https://​​github.com/​​LessWrong2/​​Less­wrong2/​​is­sues/​​2434 .

• jimrandomh 14 Sep 2019 5:26 UTC

  4 points

  in reply to: Raemon's comment on: Rae­mon’s Short­form Feed

  Note that this would be a very non-idio­matic way to use jQuery. More typ­i­cal ar­chi­tec­tures don’t do client-side tem­plat­ing; they do server-side ren­der­ing and client-side in­cre­men­tal mu­ta­tion.

• jimrandomh 13 Sep 2019 22:24 UTC

  22 points

  in reply to: An1lam's comment on: Jim­ran­domh’s Short­form Posts

  I’m kinda con­fused about the re­la­tion be­tween cryp­tog­ra­phy peo­ple and se­cu­rity mind­set. Look­ing at the ma­jor cryp­to­graphic al­gorithm classes (hash­ing, sym­met­ric-key, asym­met­ric-key), it seems pretty ob­vi­ous that the cor­rect stan­dard al­gorithm in each class is prob­a­bly a com­pound al­gorithm—hash by xor’ing the re­sults of sev­eral highly-dis­similar hash func­tions, etc, so that a math­e­mat­i­cal ad­vance which breaks one al­gorithm doesn’t break the over­all se­cu­rity of the sys­tem. But I don’t see any­one do­ing this in prac­tice, and also don’t see signs of a de­bate on the topic. That makes me think that, to the ex­tent they have se­cu­rity mind­set, it’s ei­ther be­ing defeated by poli­ti­cal pro­cesses in the trans­la­tion to prac­tice, or it’s weirdly com­part­men­tal­ized and not en­gaged with any prac­ti­cal re­al­ity or out­side views.

• jimrandomh 12 Sep 2019 18:13 UTC

  4 points

  in reply to: TekhneMakre's comment on: TekhneMakre’s Shortform

  In my ex­pe­rience, the mo­tion that seems to pre­vent men­tal crowd­ing-out is in­ter­ven­ing on the timing of my think­ing: if I force my­self to spend longer on a nar­row ques­tion/​topic/​idea than is com­fortable, eg with a timer, I’ll even­tu­ally run out of cached thoughts and spot things I would have oth­er­wise missed.