Aaron Bergman's tweets, January 2023
Every public tweet, reply, quote tweet and retweet that Aaron Bergman (@AaronBergman18 on X, formerly Twitter) posted in January 2023 and has not deleted: 121 posts, oldest first, times in UTC. Each heading links to the original post on X. When Aaron replies to or quotes someone, their tweet appears as a blockquote with their name; those words are theirs, and everything outside the blockquotes is Aaron's.
This page is one month of the complete archive of Aaron Bergman's tweets, which updates automatically every few hours. Previous month: December 2022. Next month: February 2023.
Monday, January 2, 2023
Reply, 01:44 UTC · 14 likes
The next winning Powerball numbers
Tweet, 03:45 UTC · 9 likes
🎁🎁🎁
Wednesday, January 4, 2023
Reply to @ns_whit, 19:36 UTC · 4 likes
Nick Whitaker 🇺🇸 (@ns_whit): Holden Karnofsky is the most interesting intellectual of our time. And @dwarkesh_sp might be our best interviewer https://www.dwarkeshpatel.com/p/holden-karnofsky
Podcast Host Showdown: @dwarkesh_sp and @robertwiblin simultaneously each other to prove who is the single 🐐
($10 to a charity of each’s choice if this happens and results in a >30min episode)
Thursday, January 5, 2023
Reply to @MichaelDPlant, 14:09 UTC · 2 likes
Michael Plant (@MichaelDPlant): Puzzling over the Q4 website metrics for @HappierLivesIns 🧐. Turns out 'work with us' got 3x more page views than 'give now' - we weren't hiring but we *were* urging people to give. So, I guess we ... pivot into a recruitment agency? Add a 'donate' button to our job page? 🤷♂️
No experience but I’d guess that’s basically true for most orgs - the careers page is like a mystery treasure chest which might hold “your dream job, perfect fit, apply w only resume, $250k comp+infinite stipend for anything”
Reply to @astridwilde, 23:49 UTC · 11 likes
Astrid Wilde 🌞 (@astridwilde): suffering is not inherently Bad. pleasure is not inherently Good.
This is probably the most wrong Tweet in existence (no hard feelings tho👍):
two simple propositions, both of which when negated are in the top 5 “most likely to be true zero ambiguity” claims I’ve ever entertained. And simple enough to be translated into first order logic 🥲🥲
Friday, January 6, 2023
Reply, 21:05 UTC · 2 likes
Not if it’s being sold as a research chemical with a “not for human consumption” label😇
Reply to @peterwildeford, 21:12 UTC · 66 likes
Peter Wildeford🇺🇸🚀 (@peterwildeford): ChatGPT: Explain HTML to me like I'm a dog
“Not something you need to worry about” 😡 😡 😡
Saturday, January 7, 2023
Reply to @Memetic_Theory, 06:36 UTC · 2 likes
mass (@Memetic_Theory): @AaronBergman18 @peterwildeford Yep, it may be controversial, but I do believe dogs have inferior intelligence.
yeah, thanks to comments like that ^
Reply to @SpacedOutMatt, 23:00 UTC · 2 likes
Matt 🔸 (@SpacedOutMatt): @Laura_k_Duffy @peterwildeford My experience (as a 2018 grad) was that there was ~0 academic risk to respectful disagreement and discussion in the classroom. On the other hand, my friends and I definitely would roll our eyes at some of the more annoying conservatives who always needed to play devil's advocate
IME (2021.5) there was a medium amount of asymmetric norm enforcement/isolated demands for rigor (along the basic culture war axis and in the directions you expect) Easiest example: took what was basically the imaginary CRT thing but irl (“intro to sexuality studies”) …
Thread, continued, 23:11 UTC · 2 likes
Oh ya maybe I’m dumb but totally thought “Intro to Sexuality Studies” was gonna be a very normal course introducing the academic study of sexuality through sociology/psych/bio/whatever. Complained on the feedback form that it was a misleading title lmao
Reply to @christineist, 23:33 UTC · 1 like
christine (@christineist): every now and then my body decides to stay up until the sunken hours just to see the fabric of reality shimmer a lil bit
Shit hits different at 4:09
Sunday, January 8, 2023
Reply to @PradyuPrasad, 20:06 UTC · 9 likes
Pradyumna 🇸🇬 (@PradyuPrasad): What's a good book for a smart 14 year old interested in physics?
Reply, 23:59 UTC · 2 likes
Not exactly the same but for all the paper/textbook notes I took in college: split screen: reader app (browser or pdf reader or whatever) on the left, Word on the right, and just make a long bulleted, nested list. Can add screenshots, latex-style code, links etc
Monday, January 9, 2023
Reply to @ftlsid, 01:06 UTC
ftlsid (@ftlsid): college admissions as a system that creates angry smart kids with something to prove is probably net positive for society
Lmao https://www.aaronbergman.net/p/most-ivy-smart-students-arent-at
Tuesday, January 10, 2023
Quote tweet, 01:10 UTC · 11 likes
The rarity with which most Westerners today undergo terrible acute suffering today is, I think, literally the best thing that has ever happened.
But a cruel irony is that, largely for this reason, descriptions like Ngo’s are shockingly, misleadingly rare - rendering…
(1/2)
Quoting Richard Ngo (@RichardMCNgo): Just had a routine medical procedure that was nevertheless the most painful experience of my life, leaving me curled up and nauseous. I'm now feeling very visceral empathy and horror for the thousands (millions?) of people who underwent surgeries before anaesthesia.
Thread, continued, 01:10 UTC · 8 likes
suffering’s overwhelming moral importance silent to many and a whisper to most
Why “largely”?
Bc just as important is the topic’s (completely intuitive and natural) lack of memetic fitness. It’s not a fun or sexy topic, even in EA/adjacent spaces!
Quote tweet, 01:12 UTC · 5 likes
Wait this whole thread is incredibly based
Quoting Richard Ngo (@RichardMCNgo): There's also a lot of work to be done in getting people to internalise the horror of extreme suffering. It's insane that we can't even agree that hell is bad! Increasing our civilisational empathy is one of the most important projects possible.
Tweet, 04:00 UTC · 6 likes
Both housemates have COVID. Afaict I do not (yet)
Is there a nerdy cutting edge take on transmission between two non-adjacent rooms that each fork from a single vent?
(Also I know the basics and am not super worried but any other advice is welcome too)
Reply to @astridwilde, 08:57 UTC · 7 likes
Astrid Wilde 🌞 (@astridwilde): only city i've ever been to that looks anything close to this is Barcelona
Wonderful memory of getting lost there after hitting the streets alone at 16, no cell service, while on a family vacation because “the big diagonal road” I was counting on as a landmark doesn’t seem quite as conspicuous irl as it does from the sky/on a map
Reply to @FrostedSapling, 20:05 UTC · 7 likes
Ivan 🔸 (@FrostedSapling): Jeez Louise, @inesferhumi did you have to mute this?! Nearly 100,000 likes!
(quoting @inesferhumi: ???? Why did i think i would remember what this meant)
And did u ever figure out what it was referring to??
Wednesday, January 11, 2023
Reply to @utopiannotions, 09:29 UTC · 2 likes
Conor M (@utopiannotions): It's crazy that jhana exists yet it's extremely rare for people to even TRY to achieve it. What's the explanation? If I had to guess:
99% of people haven't heard of it.
Of that 1%, half don't believe it and many of the remainder think it's undesirable or unlikely... (1/2)(quoting @nickcammarata: We live in a world where jhana exists and less than five people are working on a jhana helmet)
FWIW I’m in the <.25% left over, and idk I think it’s just a combo of laziness/short time preference (meditation is super hard for me) and probably some baseline skepticism my brain won’t let go of
Would like to think I’ll try in earnest in the future tho
Reply to @CineraVerinia, 20:04 UTC · 1 like
(The tweet being replied to by @CineraVerinia is no longer available.)
Lmao
Reply to @The_Equationist, 20:06 UTC · 2 likes
𑘀𑘦𑘡 𑘎𑘨𑘢𑘹 (@The_Equationist): @MarkovMagnifico They want to minimize the cost function.
Happy to use the agent/desire framing, but isn’t this only true during training for DL?
Thursday, January 12, 2023
Reply to @cafreiman, 05:31 UTC · 1 like
Chris Freiman (@cafreiman): Is there any topic that’s totally resistant to politicization?
To be pedantic, politics
Reply to @pleasure_engine, 07:01 UTC · 3 likes
(The tweet being replied to by @pleasure_engine is no longer available.)
I’m not overweight but it seems pretty intuitive: it’s hard not to eat when you’re hungry, and hunger is the basic mechanism driving body weight setpoint
Reply, 07:04 UTC · 2 likes
It’s an effective antidepressant
Reply to @chiefofstuffs, 07:12 UTC · 4 likes
karl yang (@chiefofstuffs): you can tell how old someone is by how they name google docs
How old am I
Reply to @peterwildeford, 19:24 UTC
Peter Wildeford🇺🇸🚀 (@peterwildeford): @ostensiblynil I forgot to mention all the variables have to be bounded [0, 1]. This summation trick results in that parameter violating the bounds.
In that case, program gradient ascent by hand? Not exactly a full set of code I’m providing
Friday, January 13, 2023
Tweet, 03:38 UTC · 19 likes
Something I wrote months ago and still stand by
(y’all are ~smart~ enough to find the source on your own, should you want it)
Reply to @nkreu113r, 07:55 UTC · 2 likes
noah (@nkreu113r): @AaronBergman18 i think this sort of thinking decays rapidly into "no cognitive traits are moral virtues".
intelligence isn't any more or less "undeserved" or "unchosen" than work ethic, charity, or not being a serial killer. brains are brains.
I see where you’re coming from, basically agree with the physicalist lens, and indeed do not think cognitive traits per se should should be regarded as virtues. But ethical behavioral patterns (eg telling the truth) which sometimes get reified as intrinsic properties (“honesty”)
Thread, continued, 07:57 UTC · 3 likes
at least often should be, because behavior responds to incentives in a way that cognitive traits per se do not
IOW Virtues should have elastic supply
Reply to @the_culturist_, 21:08 UTC · 2 likes
The Culturist (@the_culturist_): A 23 year old sculpted this.
What's your excuse?
Im 22 for 2 more months
Reply to @PradyuPrasad, 21:40 UTC · 2 likes
Pradyumna 🇸🇬 (@PradyuPrasad): @CineraVerinia If bostrom did believe that such differences were genetic (and I do think he does), several actions of his and worldviews would change. Would he statistically discriminate against black people as a result of this? I think that's a direct consequence of this object level belief
It may be a consequence for some, but it’s not a direct, intrinsic, or necessary consequence
Reply to @PradyuPrasad, 23:00 UTC · 2 likes
Pradyumna 🇸🇬 (@PradyuPrasad): @AaronBergman18 @CineraVerinia If I believe that X characteristic causes lower ability, wouldn't I need some evidence to prove that there is normal ability despite X characteristic?
Yeah but that evidence is so universal and routine to collect that, if youre interacting with a person in virtually any capacity, you already have it
Sunday, January 15, 2023
Reply to @RichardHanania, 07:45 UTC · 2 likes
Richard Hanania (@RichardHanania): People wear glasses to correct subpar vision, but has anyone ever advocated using a similar technology to create superhuman vision? How technically feasible is this? It’s strange people are satisfied with things just being good enough.
Breaking: bad person becoming based occasionally
Monday, January 16, 2023
Tweet, 00:57 UTC · 30 likes
This is The Scissor Statement isnt it?
Thread, continued, 01:00 UTC · 11 likes
Reply, 01:03 UTC · 4 likes
Would RT🔒
Reply to @AndrewCurran_, 02:03 UTC · 2 likes
Andrew Curran (@AndrewCurran_): @MatthewAdelste1 @AaronBergman18
(quoting @ESYudkowsky: so I hear the GPT4 prototype got used to generate a scissor-statement apology where half of EA would condemn the apology, and the other half would condemn anyone who condemns the apology)
Ok I swear I didn’t see yuds post before lol
Tweet, 05:41 UTC · 4 likes
Had to do a double-take when my brain parsed the Chalmers/Goff/Frankish convo as one of those fake Fellow Kids history class social media feeds
Like nbd just hashing out the nature of consciousness on the TL
Tuesday, January 17, 2023
Tweet, 07:07 UTC · 6 likes
A world in which moral atrocities are held at bay by empirical belief in universal omnidimensional equality is one teetering on the brink of such acts - and this is so regardless of whether those beliefs are true or false
Thread, continued, 07:07 UTC · 3 likes
Parts of science have, are, and will be wrong. Pseudoscience can be mistaken for its legitimate namesake
But bigotry and the creation of suffering cannot wax and wain with the phases of each scientific revolution; no empirical claim—correct or mistaken—will ever justify either
Wednesday, January 18, 2023
Reply to @ozziegooen, 04:44 UTC · 13 likes
Ozzie Gooen (@ozziegooen): Is there any term for "People who think that EA should expand very quickly, and this is about the most important cause area"?
"EA Expansionism", maybe?
Connotations aren't great, but there aren't many options.
EA/acc
Reply to @algekalipso, 06:35 UTC · 4 likes
Captain Pleasure, Andrés Gómez Emilsson (@algekalipso): https://www.opioids.com/
Who snatched that domain in like 1986?
Reply to @BenHayum, 19:50 UTC · 3 likes
(The tweet being replied to by @BenHayum is no longer available.)
How does one walk like a fir tree?
Thursday, January 19, 2023
Reply to @daniel_271828, 04:25 UTC · 11 likes
zzDANIEL_ETH (@daniel_271828): My very hot animal welfare take is that someone who eats both animal meat and plant-based meat is doing more good than someone who eats neither. Subsidizing the nascent (hopeful) eventual replacement industry is more impactful than your relatively tiny personal consumption
Plausible but the subsidization is tiny as well. And just going on intuition Id be surprised if $100 spent on a bundle of alt meat is better than a $1 donation to GFI, so under most peoples preferences it’d be better to buy only animal meat and donate the net $ difference
Quote tweet, 21:00 UTC · 96 likes
*Unless it’s an EA org, in which case the median candidate wrote a PhD thesis on whatever topic applicants should understand, the lower quartile candidate looks generally great and checks every box, and the top candidate is God
Quoting billy (@billyhumblebrag): Nothing white pills you quicker on your prospects than hiring and seeing the quality of the median job application.
Thread, continued, 21:03 UTC · 21 likes
^figuratively and directionally informed by real experience
Reply to @daniel_271828, 21:12 UTC · 6 likes
zzDANIEL_ETH (@daniel_271828): @AaronBergman18 I think this is an exaggeration
Yeah it is (even without the last part) and my sample size is 1, but fr I was very positively surprised and that was starting with high expectations
Reply to @patrickaturley, 22:16 UTC · 27 likes
Patrick Turley (@patrickaturley): I move that we change the name of the "United States" to "AAAAAA United States" so it shows up at the top of dropdown lists where you have to select your country.
Or make the alphabet start with UNITEDS
Reply to @NathanpmYoung, 22:26 UTC · 7 likes
Nathan 🔎 (@NathanpmYoung): this is why we need moral circle expansion
Reply to @QualyThe, 22:36 UTC · 3 likes
Qualy the lightbulb (@QualyThe): AI safety is not a marathon or a sprint but a secret third thing
Unironically it’s like a race of unknown length
(This would actually be neat irl, plz add @Olympics)'
Reply to @ilex_ulmus, 23:03 UTC · 3 likes
Holly ⏸️ Elmore (@ilex_ulmus): Like, I think race/IQ stuff is so hurtful bc we know good and well that many of the people "just talking about the facts" (themselves morally neutral) value their intelligence most about themselves or at least think that intelligence is determinative of moral worth...
FWIW I think this is a strong argument for decoupling. Sure, “many” people group intelligence and moral worth, but I’d guess in % terms the belief is anticorrelated with decoupling (which IMO makes sense because it’s default in our culture but wrong)
Reply to @nullpointered, 23:05 UTC · 3 likes
(The tweet being replied to by @nullpointered is no longer available.)
Yeah imma blame it on this
Friday, January 20, 2023
Reply to @adambinksmith, 01:38 UTC · 9 likes
Adam Binksmith (@adambinksmith): I hear god's airtable skills are a bit rusty tho
Yeah He didn’t get the internship
Quote tweet, 01:41 UTC · 9 likes
Every time
Quoting Aaron Bergman 🔍 (@AaronBergman18): POV: you've figured out why you're getting a double digit number of notifications
To Mr. Buterin's 3.9M followers, hello!
Thread, continued, 01:43 UTC · 9 likes
Thanks @lastpositivist for applying to the leftist-philosophy-and-Twitter internship. We regret to inform you that you have not been selected for round 2. We had many very strong applications and…
Reply to @QualyThe, 01:49 UTC · 2 likes
Qualy the lightbulb (@QualyThe): if you were sceptical of web3 but think AI is the next big thing then congratulations you have good judgement
Stay contrarian
Reply to @frances__lorenz, 03:53 UTC · 1 like
(The tweet being replied to by @frances__lorenz is no longer available.)
Ayy lfg
Reply to @LexerLux, 03:56 UTC · 15 likes
(The tweet being replied to by @LexerLux is no longer available.)
Most people earning 40k aren’t surfing instructors tho
Reply to @lastpositivist, 14:31 UTC · 10 likes
Liam Bright (@lastpositivist): Mix of fair and unfair critiques in this. True - there is often socio-political naivety among EA rank and file, and upper echelons seem to avoid social theory in a way one cannot help but suspect is related to maintaining relationships with pay masters...
https://jacobin.com/2023/01/effective-altruism-longtermism-nick-bostrom-racism
Holy shit this is infowars-level bad
Reply, 22:06 UTC · 1 like
FWIW some new info makes FLI’s actions seem not nearly as bad IMO
Saturday, January 21, 2023
Reply to @absurdlymax, 02:14 UTC · 3 likes
Max Alexander (@absurdlymax): So it turns out tweeting about how dates should be written might be the path to engagement farming
There is a _relevant_ double entendre here
Sunday, January 22, 2023
Tweet, 01:05 UTC · 5 likes
Deleting bc they got a partial fill at the 5th or 6th pharmacy they tried—a CVS—after being told repeatedly that all local CSVs were out of stock
Posting this in case it's useful info
Reply to @focusfronting, 03:41 UTC · 3 likes
(The tweet being replied to by @focusfronting is no longer available.)
me fr
Reply to @robbensinger, 03:47 UTC
Rob Bensinger ⏹️ (@robbensinger): ............... What????
This literally ever happens?? Huh? What do you mean by "conscious" here??
(quoting @ElodesNL: @Malcolm_Ocean I remember it as being a fairly clear single moment, around when I was three or four years old; suddenly one day I was conscious and I haven't stopped since I also had a few memories already, which was interesting; they'd been saved for me while I still lacked…)
Reply to @jordanwpieters, 04:33 UTC · 8 likes
(The tweet being replied to by @jordanwpieters is no longer available.)
🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸
🇺🇸 🏈💵WELCOME TO THE UNITED STATES OF AMERICA 🇺🇸🏈💵
🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸
Reply to @lainewrites, 08:32 UTC · 2 likes
elaine (@lainewrites): used to think it was an introvert thing, but now i’m convinced that when you feel drained after a social event it’s bc you weren’t comfortable being yourself w the ppl you were with.
it’s super energy-consuming to suppress a part of you or be anyone other than yourself.
Wait isn’t this a (the?) definition of introversion tho
Tweet, 11:55 UTC · 8 likes
Google’s gonna train their LLM on the 25 million books aren’t they https://www.theatlantic.com/technology/archive/2017/04/the-tragedy-of-google-books/523320/
Thread, continued, 11:55 UTC · 4 likes
That’s an interesting pull quote
Reply to @Laura_k_Duffy, 20:16 UTC · 2 likes
Laura Muresianu 🌐🔸🇺🇸 (@Laura_k_Duffy): @AaronBergman18 @laine_writes I think the social-draining definition of introversion is kinda bad. I would define it as “do you feel anxiety before/during social encounters” bc this, imo, is a better predictor of how much I’ll seek out such encounters. (I’m poorly calibrated in how much I’ll enjoy them)
I’m a weirdo, ~zero social anxiety but any amount of human interaction must be followed by a recharging period of appropriate length
Reply to @mirandahzhang, 20:20 UTC · 12 likes
Miranda Zhang (@mirandahzhang): There are forms of suffering that persist to the extent of being a permanent debuff (e.g. chronic pain). Are there currently equivalent forms of happiness?
(If not, I guess this is part of the argument for wireheading?)
Unipolar hypo/mania is a rare ‘condition’ I think. The opposite of depression, hedonically
Reply to @captgouda24, 23:56 UTC · 4 likes
Nicholas Decker (@captgouda24): Christians are not freaking out enough. Like really -- how can you accept that hell is real, you go there if you do not believe, and yet not support having the government seize the children of non-believers?? We want laws against child abuse, but not against eternal torment??
I’ve seen tweets like “why aren’t EAs doing terrorism if they really believe what they say?” and am inclined to think an “effective Christian” would also find this kind of thing instrumentally counterproductive (perhaps in addition to being intrinsically bad)
Monday, January 23, 2023
Reply to @NathanpmYoung, 00:00 UTC · 3 likes
Nathan 🔎 (@NathanpmYoung): Who is your favourite critic of EA that you'd like EAs to hear a talk from?
Me (super egotistical ik but I’m allowed to integrate all the critiques I like right?)
Reply to @utopiannotions, 00:43 UTC · 3 likes
Conor M (@utopiannotions): I just finished writing a novel!
It's the first time I've ever finished something this long. It's 95,000 words and very messy, but it's mine. I'm so happy.
One. Long. Thread.
Reply to @justjoshinyou13, 00:55 UTC · 2 likes
Josh You (@justjoshinyou13): They're sea lioning so hard smh
Idk if there’s enough room in this thread for both squads
Reply to @peterwildeford, 02:32 UTC · 4 likes
Peter Wildeford🇺🇸🚀 (@peterwildeford): Failing to detect the human hiding behind a tree is something that ChatGPT also falls for
“a movement of 6 feet would be unusual for a tree”
Quote tweet, 02:34 UTC · 5 likes
I am officially concerned
Quoting Nick (@nickcammarata): It’s as good a time as ever to start an equanimity practice. One of the most potent and broadest hedges you can have against a wild future
Reply to @paulg, 03:06 UTC · 3 likes
Paul Graham (@paulg): One sign that determination matters more than talent: there are lots of talented people who never achieve anything, but not that many determined people who don't.
A rare mistaken take
Reply, 05:04 UTC · 9 likes
To be annoying, there is no line, it’s all just a spectrum from Will MacAskill to “anyone who’s friends with anyone who’s ever donated to GiveWell”
Or self-identification if you don’t wanna essentialize
Reply to @QualyThe, 11:00 UTC · 2 likes
Qualy the lightbulb (@QualyThe): GPT-4 prediction thread 👇:
FWIW https://m.slashdot.org/story/409852
Tuesday, January 24, 2023
Reply to @Laura_k_Duffy, 08:42 UTC · 5 likes
Laura Muresianu 🌐🔸🇺🇸 (@Laura_k_Duffy): A bit of personal news (which some of you already know): In about a month, I’ll be moving from Montana to (most likely) the DC area.
Why? A myriad of reasons, including desires to heal past wounds, find new friends, and, of course, experience much warmer winters.
Out: Oxford, SF/Berkeley, NYC, Austin, Great Falls, Boston/Cambridge, New Haven, Bahamas, London
In: DC
On the bubble: Tulsa, outer space
Quote tweet, 09:03 UTC · 4 likes
Endorse. Pulling this off would be some serious SFE Robin Hood shit (not the app)
Quoting Captain Pleasure, Andrés Gómez Emilsson (@algekalipso): The fact that we, as a civilization, haven't eradicated cluster headaches even though DMT can get rid of them for peanuts, is perhaps the most damning thing about our species.
EA 2.0 focusing on pragmatic solutions for extreme suffering would be a beautiful plot twist arc.
Reply to @NathanpmYoung, 16:20 UTC · 1 like
Nathan 🔎 (@NathanpmYoung): Hanania seems on an interesting journey right now:
- his recent support of the press
- his criticism of his own followers (IIRC he said some repulse him)
- his continued support of prediction markets, will force him to be more accurate than many others(edited)
(quoting @RichardHanania: Legalize prediction markets now and try to create a world where people this stupid are held accountable for the ways in which they pollute public discourse.)
I think Scott Alexander’s “worst person” remark might have gotten to him
Fine, “might”=35%
Wednesday, January 25, 2023
Reply to @sullyj3, 00:35 UTC · 2 likes
░▒▓▒░▒▓▒░▒▓▒░ (@sullyj3): @AaronBergman18 @NathanpmYoung Which one was that
https://astralcodexten.substack.com/p/links-for-june-1e7
Reply to @ConnorTabarrok, 00:41 UTC · 9 likes
Connor J. Tabarrok 🏗🚀🌏 (@ConnorTabarrok): "Actually, whaling is unassailably good because of the welfare benefits to shrimp."
Yes but have you fully modeled the complete nth order effects on every form of ocean creature for all of eternity?
Thursday, January 26, 2023
Tweet, 17:04 UTC · 19 likes
Not Thing 1 or Thing 2, but a secret, more complex third Thing
Tweet, 21:50 UTC · 10 likes
Shipped my first web app a few hours ago, a Spotify playlist analyzer :)
Had to do some weird API workaround tho, not sure it was 100% kosher lol
Quote tweet, 22:12 UTC · 3 likes
What's the biggest thing YOUR gc has achieved? "Big" can include super niche if it was consequential for *someone*, btw
Bonus points for small size, penalties assessed for being a Slack or >15 people
Quoting Nathan 🔎 (@NathanpmYoung): don't name names, but what is biggest thing a group chat has achieved?
roonchat seems maybe to have moved A16Z so that's pretty big. anyone got better?
Friday, January 27, 2023
Reply to his own earlier tweet, 03:52 UTC · 3 likes
Aaron Bergman, January 7, 2023: IME (2021.5) there was a medium amount of asymmetric norm enforcement/isolated demands for rigor (along the basic culture war axis and in the directions you…
Guy in Maine and guy in Florida debate the weather on the East Coast
Reply to @peterwildeford, 04:02 UTC · 4 likes
Peter Wildeford🇺🇸🚀 (@peterwildeford): Interesting to see what the list of most donated charities are in the US:
Would be curious to see a Gini index for each (and every charity ideally), where the units are individuals’ total annual donations (I guess both counting and ignoring non-donors?)
Like are there 10 figure donations to Feeding America? Mostly made of 2/3/4/5 figure ones?
Reply to @MatthewJBar, 06:06 UTC · 6 likes
Matthew Barnett (@MatthewJBar): New estimate of the consumer surplus of the internet just dropped. If it's accurate, then there was no great stagnation.
Let me explain.🧵
(quoting @JgaltTweets: 'By far the worst part of Bankman-Fried’s nine-day Bahamian incarceration was the lack of internet. “I didn't realize how much more important than everything else combined internet access is to me, but that was like 80% of the total cost of being in prison," says Bankman-Fried')
Not sure how seriously this is intended but, assuming for argument SBF is right about his own experience, Id bet most of the delta is caused by *losing* access per se after hedonic/habitual adaptation, rather than merely lacking it
Even still, I agree consumer surplus from
Tweet, 06:20 UTC · 3 likes
Everyone should pin their 10+ most-liked and favorite Tweets as an act of community service. No limit to the number as long as they’re descending in subjective value
Reply to @Duderichy, 20:35 UTC · 1 like
the Rich (@Duderichy): yes landlords will frequently accept $0 a month in order to make more money
I highly doubt it's significant but the monopoly/oligopoly supply restriction logic still checks out in this market
Saturday, January 28, 2023
Reply to @EigenGender, 11:16 UTC · 7 likes
EigenGender 🔸 (@EigenGender): whether or not he's correct, Yudowsky's writings are expertly and perfectly and magnificently crafted to lure smart people into wanting to believe him completely. That's why Yudowsky should argue that no one should read his own work.
Lots of hot takes here but IMO he’s just a very good writer - good stylistically and excellent at actually communicating ideas with high fidelity. And I certainly don’t agree with him on everything, maybe consciousness most of all
Tweet, 16:43 UTC · 6 likes
It’d be cool if everyone had a unique name. The one many acquaintances would know, so as things stand middle names wouldn’t count.
Like kinda dumb that there are 50ish Aaron Bergmans just on FB
Thread, continued, 16:43 UTC · 3 likes
And no scarcity of possibilities: 25B 10 letter combos with no adjacent vowels or consonants=enough for everyone
Hi I’m Opuwasilol
Reply to @NathanpmYoung, 17:00 UTC · 8 likes
Nathan 🔎 (@NathanpmYoung): Do you do the @givingwhatwecan pledge?
If so who do you give to?
No(t yet, pending more $ stability/certainty) but to RP and Fish Welfare Initiative
These are my best guesses for ‘best institutional recipient of $1’ but also guess a 10 hour investigation would yield otherwise just bc the possibility space is so large
…
Thread, continued, 17:05 UTC · 5 likes
Seriously, change my mind!
No plurality bullshit I want to know which single donation-accepting institution it is most ethical to give $1 (or $1k, maybe not $1B)
Ideally I’d put up a real bounty but rn I can mail you an EA Georgetown sticker
Quote tweet, 19:35 UTC · 12 likes
Why didn’t I learn about this in Hebrew school?
Quoting Boze Herrington, Library Owl 😴🧙♀️ (@SketchesbyBoze): One of my favorite things in the Talmud is the intense debate about whether the plague of frogs in Exodus was a million tiny frogs or one *very large* frog. (1 / 5)
Sunday, January 29, 2023
Reply to @absurdlymax, 00:41 UTC · 2 likes
Max Alexander (@absurdlymax): @AaronBergman18 This would have made passover way cooler
The time for CRISPR has come
Reply to @JacquesThibs, 00:47 UTC · 5 likes
Jacques (@JacquesThibs): My tweets when being as factual as possible:
12 likes, 1 retweet, 3 commentsMy tweets when I’m a little hyperbolic:
99999 likes, 9999 retweets, 999 comments
Does this make your tweet true or false?? 😰 😰 😰
Tweet, 01:20 UTC · 13 likes
Would bet that most exercise <> health outcome correlations are due way more to health → exercise causation than (I think?) basically anyone else believes
And tbc I don’t think this looks only like “person with [condition] doesn’t exercise for obvious, legible reason”…
Thread, continued, 01:20 UTC · 7 likes
More like “going from 30th to 50th percentile in systemic inflammation thanks to [exogenous reason] as measured by CRP somehow causes the brain to find all physical activity more difficult/higher threshold/less pleasant”
Not sure I could specify this well tho
Reply to @Aella_Girl, 07:35 UTC · 1 like
Aella (@Aella_Girl): my data was too big for google sheets, so i installed and learned python. but now it's too big for my computer, python won't run it. off to buy a new computer i guess?
How many GB? With some difficulty/crashing, R+RStudio+datatable package+decent but not amazing gov’t laptop with 8GB RAM handled 16M rows over I think 4GB
Reply, 10:48 UTC · 4 likes
(The tweet being replied to is no longer available.)
Lmao
Reply to @lisatomic5, 15:46 UTC · 2 likes
lisatomic (@lisatomic5): everybody postin takes about ending compulsory education, ain't nobody wanna hire an eighth grader
I had a camp counselor job then. It paid, and I promise this is literally true, $100/week≈$2/hour. Not sure why I agreed to that
Reply, 17:53 UTC · 2 likes
how about “df%>%filter(dummy!=1)”
Quote tweet, 20:54 UTC · 15 likes
Hot take: LW-brand rationalism is good
I think because this bundle of norms is:
- non-obvious,
- requiring of a bunch of social cooperation but non-coercive
- in fact conducive to its self-defined objectives; and
- empirically used often for good ends and rarely for bad ones
Quoting Aella (@Aella_Girl): Its hard to describe what rationalist culture is "about"; here's a good summary of the conversational guidelines:
Monday, January 30, 2023
Reply to @mnolangray, 00:14 UTC · 6 likes
M. Nolan Gray 🥑 (@mnolangray): Fascinating, MGM Grand set a trap in my room.
Is this kind of thing legal? Could they just declare the charge to be $5k instead? $500k?
Reply to @CineraVerinia, 01:00 UTC · 1 like
(The tweet being replied to by @CineraVerinia is no longer available.)
Lifting makes me way hungrier fwiw, meaning I’m pretty sure that periods of more regular gym my bf% actually increases without intentionally trying to gain (not gigantically but some, alongside lean weight)
In early HS I did naive linear extrapolation, actively trying to
Reply to @norabelrose, 03:08 UTC · 3 likes
Taylor Belrose (@norabelrose): I don’t find moral realism to be conceptually coherent. I can’t imagine a world in which it’s true. That said, most philosophers do seem to think it’s coherent. I wonder if you could formalize the specific assumptions under which moral realism is coherent or not
Secular moral realist (and I’ve given it a fair bit of thought), and I stopped listening to this episode after 20 minutes because it was apparent that the guest and I already agreed >99.9% on this one niche take https://80000hours.org/podcast/episodes/sharon-hewitt-rawlette-hedonistic-utilitarianism/
Reply to @somewheresy, 05:01 UTC · 4 likes
(The tweet being replied to by @somewheresy is no longer available.)
Attempted explanation on 5 mins of research:
Human body is mostly water, and I’m pretty sure proteins are the main category of secreted molecules
there are lots of (mostly?) water-soluble proteins, so making protein-water solutions seems pretty easy
Reply to @calleymeans, 06:59 UTC · 18 likes
Calley Means (@calleymeans): I assume Japan is mass-prescribing surgeries and injections to combat obesity?
With self-reporting countries removed
Reply to @PradyuPrasad, 17:43 UTC · 8 likes
(The tweet being replied to by @PradyuPrasad is no longer available.)
Not the opposite you intended but my parents put basically zero pressure on me academically and I think this was great
Tweet, 21:52 UTC · 3 likes
ChatGPT updated (for the third time, apparently) https://help.openai.com/en/articles/6825453-chatgpt-release-notes
Tuesday, January 31, 2023
Reply to @NathanB60857242, 09:08 UTC · 6 likes
Nathan Barnard (@NathanB60857242): much shorter timelines, much higher p(doom)
Hard to imagine MacAskill putting 3% on AI x-risk (from WWOTF) if him+team wrote and researched the book from the same office as MIRI. Idk whether this would make the estimate more less accurate tho tbc
Reply to @NathanpmYoung, 11:56 UTC · 2 likes
Nathan 🔎 (@NathanpmYoung): Capitalism and extinction-risk
It seems likely to me that without capitalism, extinction risk would be lower. But capitalism is very effective at bringing people out of poverty.
If you want to replace capitalism on risk grounds, how and what would you replace it with.
Not a complete answer, but the worst case of all would be low extinction risk + 2022=end of political and economic history. That just means factory farming persists indefinitely at large scale. Almost any other trajectory is better besides true s-risk
Reply, 19:27 UTC · 10 likes
“Innocent until P(guilty)=x” - makes sense for x to vary depending on the associated consequences
Reply to @Aella_Girl, 20:40 UTC · 3 likes
Aella (@Aella_Girl): critiques i get from ppl who don't know stats: "Selection bias! Internet surveys aren't 'studies'! Just Aella's Audience!"
from ppl who do know stats:
"You should try [different new statistical method] instead, would account for [obscure math thing I need to go learn now]"
Not disagreeing, but worth noting that most (all?) qualitative methodological critiques can be formally described and sometimes corrected for using some obscure Stata command, and the obscure math things can (almost?) always be described in plain English
So approximately:
Reply to @EzraJNewman, 23:19 UTC · 2 likes
(The tweet being replied to by @EzraJNewman is no longer available.)
what's the best x then for everything?
End of January 2023. Previous month: December 2022. Next month: February 2023. Back to the archive index.