This site is new, very imperfect, and a work in progress; your feedback is greatly welcome!

Aaron Bergman's tweets, December 2022

Every public tweet, reply, quote tweet and retweet that Aaron Bergman (@AaronBergman18 on X, formerly Twitter) posted in December 2022 and has not deleted: 120 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: November 2022. Next month: January 2023.

Thursday, December 1, 2022

Reply to @peterwildeford, 19:54 UTC · 2 likes

Peter Wildeford🇺🇸🚀 (@peterwildeford): Wow, the new ChatGPT just outright automates @AdamJShriver and months of @RethinkPriors work

image attached

Seems like a pretty shoddy argument though(?)

A couple sentences begging the question, a few with meaningless terms (“moral behavior” referring to patienthood), appealing to smth not being “necessarily” the case, not addressing the meaning of “proxy”

Reply to @austinc3301, 20:03 UTC · 2 likes

Agus 🔸 (@austinc3301): I actually read moral behaviour as referring to moral agency?

I think it would make more sense that way, but the broader question about moral weight (at least in EA spheres afaik) almost always refers to patienthood. Like “should we care about shrimp” not “do shrimp themselves have a responsibility to act ethically”

Reply to @lexfridman, 20:07 UTC · 1 like

(The tweet being replied to by @lexfridman is no longer available.)

What a strange tweet

Tweet, 23:52 UTC · 2 likes

PSA: if you like Spotify wrapped but wanna know more granular/precise info than what that provides, you can download your last year of listening data: https://www.spotify.com/us/account/privacy/

Friday, December 2, 2022

Reply, 07:49 UTC · 24 likes

Doesn’t seem accurate since the context (GPT is taking an IQ test) isn’t being provided. Bet it hits 100 (low confidence) if “for question ___ of an IQ test that you’re really trying to get right” or smth was put in front of each prompt

Reply to @daniel_filan, 13:07 UTC · 2 likes

(The tweet being replied to by @daniel_filan is no longer available.)

2b. But is it arbitrary that up is considered the better direction 🤔

Reply to @BootsPlatypus, 15:17 UTC · 2 likes

Platypussy in Bio (@BootsPlatypus): i don't understand why SSRIs are considered to have low abuse potential.

every time i take my daily pill, i get such an amazing high for an hour and then have an internal struggle for the rest of the day bc i want to take another but know that i can't.

Huh wow, fwiw I think this is pretty uncommon

Saturday, December 3, 2022

Tweet, 00:25 UTC · 2 likes

CMV: “understanding” as normally understood is a basically fake or illusory concept, perhaps aa wrapper label for a cluster of qualia but nothing more

(best analogy I can think of: “blessed” as in “feeling blessed” if we live in a world where blessings don’t exist)

Reply to @frances__lorenz, 20:13 UTC · 11 likes

(The tweet being replied to by @frances__lorenz is no longer available.)

Bigger than FTX story imo

Reply to @frances__lorenz, 20:28 UTC · 27 likes

(The tweet being replied to by @frances__lorenz is no longer available.)

ntifying the moral value of shrimp atm?

Sunday, December 4, 2022

Tweet, 00:55 UTC

No fake intermediate conciliatory positions…

Consciousness aka sentience aka qualia aka genuine legit private internal experience of which there’s (allegedly) a Hard Problem and which is why there’s something it’s like to be you (iff that’s indeed the case) is _______

Tweet, 06:52 UTC · 4 likes

.@Spotify
Situation:

- smaller artists don’t sell merch but *would* if it was worth their time/$ (+some ppl say ur screwing them over i don’t really have a take here)

- ur a big company with a lots market power and money and presumably experience dealing with suppliers

1/?🧵

Image

Thread, continued, 06:52 UTC · 1 like

So (obviously):

1. use economies of scale+capital+talent+market power to contract at a sub-market rate with whoever makes customized stickers/water bottles/hoodies/whatever

2. ask artists if they wanna upload pics within certain dimensions to go on their merch

(?=5)

Thread, continued, 06:52 UTC · 1 like

3. push notifications to users asking if we wanna buy stuff (e.g. a sticker with some artist’s logo/design)
- ^maybe once a year, maybe rotating by blocs of user, maybe year round for all idk

…

Thread, continued, 06:52 UTC · 1 like

4. aggregate orders of the same design/exact product (maybe daily for top artists but less frequent for smaller ones idk) -> order a bulk shipment of these -> have printing contractor ship or ship in-house to customers

(can prob ship stickers for cost of an envelope+stamp)

…

Thread, continued, 06:52 UTC · 2 likes

5. take a cut

6. send the rest to artists (zero risk and almost zero cost for them - their time/effort making the designs but could literal album covers they already have)

7. everyone wins (except wherever we'd otherwise have spent that money but thats capitalism man)

(5/5)

Reply to @andy_matuschak, 20:58 UTC · 5 likes

Andy Matuschak (@andy_matuschak): A provocation I'm liking: most piano students have a 1:1 tutor; few math students (say, undergrads majoring in math) have a 1:1 tutor.

Lots of work is predicated on the latter being intractable ("must find some way to make group instruction as good as 1:1!"). Why the difference?

Avg 20 y/o pianist cares about being good at piano more than the avg 20 y/o math major does for math (most of instrumental benefit of latter comes from crossing the binary “yes u graduated” threshold)

Monday, December 5, 2022

Tweet, 12:43 UTC · 4 likes

👀

Thread, continued, 12:43 UTC · 2 likes

No take yet btw - have only briefly skimmed

Tweet, 20:40 UTC · 2 likes

Of all the inconsistencies to have? On literally the second thing I tried?

(Left pic first, right one immediately after

Image

Image

Tuesday, December 6, 2022

Reply to @absurdlymax, 05:37 UTC · 5 likes

Max Alexander (@absurdlymax): What is the best way to buy one individual stock? I would like to purchase a share of Asana as a gimmick

AssEt PrIces REflect aLl pUbLiC iNfoRMaTiON ANd ThUs ConSiStENtlY eSTimATE tHE net pREsENT VAlUE of All FuTUre cAsH FLOwS

Reply to @mattyglesias, 05:38 UTC · 2 likes

Matthew Yglesias (@mattyglesias): ChatGPT is so averse to stereotypes and generalization that it's reluctant to say men are taller than women.

image attached

Welcome to the IDW

Tweet, 06:06 UTC · 24 likes

Ik chatGPT was trained with the goal of predicting the next token from some text, but that’s not literally what it’s doing now, right?

Do we know what it *is* trying to do now, in deployment?

Reply to @absurdlymax, 06:55 UTC · 3 likes

Max Alexander (@absurdlymax): What is the best way to buy one individual stock? I would like to purchase a share of Asana as a gimmick

Also idk if it’s the best but I have an etrade account with individual stock shares. Opened it after guessing some March madness games right when I was 9 and won $400

Reply to @Aella_Girl, 15:08 UTC · 10 likes

Aella (@Aella_Girl): Escort monthly income by BMI, from my escort survey I've had open for a while. Sample size is still kinda low, like n=~30 per bin, and this is sampling escorts online enough to take my survey, but there still looks like a trend I'd expect. Interesting to see the degree of impact!

image attached

In order to not to have my econ major rescinded, I’m legally obligated to call out the “impact” phrasing as an implicitly causal claim that’s unjustified by the methodology

Reply to @HSGrowthGuy, 15:18 UTC · 2 likes

Connor | The HubSpot Guy 🤘 (@HSGrowthGuy): The best example in the world:

http://Berkshirehathaway.com

This has gotta be a meme though, not like “ol warren forgot to update the css file”

Reply to @CineraVerinia, 16:55 UTC · 3 likes

(The tweet being replied to by @CineraVerinia is no longer available.)

It’s for those who saved on bulk glazed rings

Image

Wednesday, December 7, 2022

Tweet, 13:44 UTC · 13 likes

Wait blinded placebo-controlled trials (like for Rx drugs) can’t be the “gold standard”— they only have 2/5 relevant groups! (all still under the informed consent constraint).

The 5 should be:
- A & B know which group they’re in (placebo vs treatment); C & D do not

Thread, continued, 13:44 UTC · 3 likes

- A & C get treatment; B & D get placebo

- E has the same inclusion criteria but never hears about any study. They’re just observed as a sort of zero information control group (tho ofc this only works when there’s individual-level public info on the relevant metric)

Reply to @jimmy_wales, 20:08 UTC · 282 likes

Jimmy Wales (@jimmy_wales): Not for sale.

(quoting @LevineJonathan: I wonder how much Wikipedia would cost? @elonmusk)

Guys you can have your very own copy on a small physical object for <$20

Image

Image

Thursday, December 8, 2022

Tweet, 00:04 UTC · 2 likes

Do you have a backup copy of ~all the information on your main personal computer (like on an SSD or cloud service)?

Thread, continued, 00:04 UTC · 2 likes

FWIW I don’t, but, like, probably should right? iOS makes it easy , apparently, but I don’t (yet) own a big enough storage thing to put the backup on. It’s < $100 for a 1TB SSD tho so thinking I probably will

Reply, 21:46 UTC · 2 likes

Ay congrats!!

Reply to @SpacedOutMatt, 21:51 UTC · 2 likes

Matt 🔸 (@SpacedOutMatt): @TheZvi @NathanpmYoung Why do you say that? In the linked example, it seems like a pretty healthy fraction of the calendar is time blocked off for either family or solo work

I have the same reaction, and its the lack of flexibility and associated lack of slack (cc Zvi lol) + having to keep the “am I in the right time block right now?” module running most waking hours — not the distribution of time per se

Friday, December 9, 2022

Quote tweet, 02:28 UTC · 5 likes

Unless I’m missing something or being very dumb (both possible tbf), taking Martin’s rec here was like getting buyers remorse in reverse: feeling like an idiot for having not signed up before

29 of y’all might wanna consider

Image

Quoting Martin Glusker (@GluskerMartin): @AaronBergman18 Ive heard backblaze is good

Saturday, December 10, 2022

Tweet, 01:36 UTC · 6 likes

Image

Thread, continued, 01:36 UTC · 3 likes

Also lol

Image

Reply to @absurdlymax, 02:41 UTC · 2 likes

Max Alexander (@absurdlymax): @AaronBergman18 So fast it's actually just standing still

It would be like if an f13 simply couldn’t handle the air over St Louis specifically

Reply to @RatOrthodox, 04:36 UTC · 1 like

Brangus🔍⏹️ (@RatOrthodox): Many of the people I most want to work with live in Berkeley.

Lol my reply was bad sarcasm - exact same and I’ve had intentions to try to move ranging in seriousness from “maybe just don’t get on the flight home” to “probably at some point, eventually”

Quote tweet, 12:22 UTC · 7 likes

Ok funny meme haha but I’ll admit it pushes the boundaries of my understanding of evolution…

Like I sure don’t think human-level dog is possible, but what *is* the limiting factor on max(IQ | dog)? Is it dog (species membership whatever that is) constraining IQ or the reverse?

Quoting @goth (@goth600): Alright, who’s got the biggest shrimp tank here? We need to get to work!

image attached

Tweet, 12:38 UTC · 3 likes

We need a general purpose aboutness preposition that sounds less pretentious than “with respect to”

Sometimes its just not worth making ur brain figure out the precise semantic relationship between two things and then translating that into language

Reply to @CineraVerinia, 19:54 UTC · 2 likes

(The tweet being replied to by @CineraVerinia is no longer available.)

If GW bought the house and put out a blog post explaining why I’d be cool with it by default

Sunday, December 11, 2022

Reply to @tummycom, 01:58 UTC · 6 likes

Evelyn (@tummycom): @the_wilderless https://en.wikipedia.org/wiki/Gimli_Glider

Lol (tbc these two pics don’t exactly tell the whole story)

Image

Image

Tweet, 05:59 UTC · 10 likes

Guy who makes bismuth subsalicylate (the active ingredient in pepto bismol) at home from chemistry-grade bismuth and salicylic acid (lifetime supply of both for ~$100 on amazon) bc he’s concerned about the amount of red dye used in commercial products but is bullish on C7H5BiO4

Reply to @HiFromMichaelV, 10:41 UTC · 8 likes

michael vassar (@HiFromMichaelV): Just found out that Aaron Schwartz’s last essay was about HPMOR 😢

Third to last liked tweet, 10 years ago

Image

Monday, December 12, 2022

Reply to @peterwildeford, 03:03 UTC · 2 likes

Peter Wildeford🇺🇸🚀 (@peterwildeford): What’s something you disagree with me on?

Consciousness, at least I think. You’re an eliminativist of some kind right?

Also the relationship between that and ethics. Setting aside built-in-uncertainty, IMO illusionism -> nihilism

Friday, December 16, 2022

Reply to @hillelogram, 17:02 UTC · 2 likes

Hillel (@hillelogram): I have discovered the best possible use of ChatGPT

image attached
image attached
image attached
image attached

“Some deeper than the tallest mountains on earth”

Tweet, 21:22 UTC · 3 likes

Image

Saturday, December 17, 2022

Reply to @JacquesThibs, 11:44 UTC · 2 likes

Jacques (@JacquesThibs): If he’s looking for a life partner and also wants to do AI governance, DC does seem ideal

Disappointingly few of them are on Twitter but Georgetown EA is a thing now too FWIW, in addition ofc to DC EA

Reply to @uneventual, 11:49 UTC · 2 likes

lucy 🐧 (@uneventual): done with college👩‍🎓

🎉🎓

Reply to @xriskology, 17:55 UTC · 6 likes

Dr. Émile P. Torres (they/them) (@xriskology): You're not gonna believe this, but an EA-aligned org called "European Summer Program on Rationality" received ~$4.5 million from Bankman-Fried's FTX to buy what *they themselves* describe as a CASTLE in the Czech Republic. A 🧵of actual pics of the castle, bought in April 2022:

image attached

in other units, ~.88 Pennsylvania Rite Aids

Image

Sunday, December 18, 2022

Reply to @EzraJNewman, 00:52 UTC · 11 likes

(The tweet being replied to by @EzraJNewman is no longer available.)

Tbh animal welfare seems to pretty clearly dominate in terms of neglectedness, tractability, and importance - the latter by orders of magnitude

I've dont think I've seen a plausible argument against this from within a pretty default EA/utilitarian worldview/set of assumptions

Image

Reply to @SpacedOutMatt, 00:55 UTC · 2 likes

Matt 🔸 (@SpacedOutMatt): Are you convincible on animal welfare? I've ended up giving about 50/50 between GiveWell top charities and ACE top charities

There's a really solid chance I'm misremembering but bc this itself could be a quite persuasive: did you say our convo a few months back pushed you towards actually donating some amount more to animal welfare?

Reply to his own earlier tweet, 01:12 UTC · 3 likes

Aaron Bergman, December 18, 2022: Tbh animal welfare seems to pretty clearly dominate in terms of neglectedness, tractability, and importance - the latter by orders of magnitude I've dont think…

Although, taking a step back, the thing I actually endorse is donating only a token amount now and using the money you have now to maximize smth like (wlog(w))^.5 or in english try to become rich*

*no fraud tho

Tweet, 01:38 UTC · 1 like

Hypothesis: a person/situation's "vibes" are, computationally, the interaction term between all 5+ sensory input streams, which initially is too high-dimensional to deal with so🧠compresses and then treats (functionally+phenomenologically) similar to, say, "pure" smell or taste

Reply to @SpacedOutMatt, 01:52 UTC · 2 likes

Matt 🔸 (@SpacedOutMatt): @AaronBergman18 @EzraJNewman Yes. I gave 25% to animal welfare last year and (double-checking my spreadsheet) 45% to animal welfare this year. Hard to say exactly how much of that was your influence (I gave a greater % last year to animal welfare than the year before) but it was definitely some

Cool time to update my 'virtue points' spreadsheet (kidding lol)...

acc though that's awesome and "greater % last year...than the year before" is the actual thing I'd highlight!

Tweet, 03:38 UTC · 14 likes

Better than a Nobel🥲

I'd like to thank to Microsoft Word for this commemorative PNG

Image

Thread, continued, 03:38 UTC · 5 likes

If you insist...https://www.lesswrong.com/posts/x6Kv7nxKHfLGtPJej/the-topic-is-not-the-content

Reply to @RomanHauksson, 07:14 UTC · 1 like

Roman Hauksson-Neill (@RomanHauksson): @AaronBergman18 @EzraJNewman What's the best way to do that?

easy just be 99.99th percentile prescient, ambitious, smart, hardworking, lucky, charismatic and create a company that delivers lots value to lots of people while capturing some as profit incl enough to invest duly in regulatory capture and R&D

fr though...

Thread, continued, 07:17 UTC · 1 like

it seems like quant trading is the highest paying kinda normalish 'job.' Not sure how, say, 95th percentile swe compares. But 2 second BOTEC says that maxes out at something like $100M donated, so some sort of entrepreneurship is probably needed to get above that, which means

Thread, continued, 07:20 UTC · 2 likes

a mildly toned-down version of my first sarcastic tweet to exploit some latent market or inefficiency that hasn't been picked up by the market yet. Probably in tech, advertising, healthcare, AI, energy, or finance - doesn't exactly narrow things down a ton tho haha

Tweet, 10:29 UTC · 7 likes

1) Claim: “Total utilitarianism does not imply that any amount of suffering can be offset or morally justified by even an arbitrarily large amount of wellbeing”

Epistemic status: Ive thought about this one a lot (prob more than any other q) and am pretty damn sure at this point

Thread, continued, 10:29 UTC · 5 likes

I made a more formal and complete argument in the section of this post titled “II A case for suffering-leaning ethics” but have been meaning to make it it’s own thing somewhere - because it at least might be novel (???), true, and important!
https://forum.effectivealtruism.org/posts/eeDsHDoM9De4iGGLw/questioning-the-value-of-extinction-risk-reduction-1

Thread, continued, 10:29 UTC · 3 likes

2) Still quite AI-pilled, but down from my peak by like 7.5% 🥱

<more to come but the 1 mg of melatonin is hitting hard rn😴>

Reply, 14:02 UTC · 10 likes

Paper: http://eprints.lse.ac.uk/107415/1/manning_edit_revision.pdf

Blog (mine fair warning): https://www.aaronbergman.net/p/a-progressive-ish-case-against-the

Image

Reply to @metakuna, 14:03 UTC · 3 likes

metakuna (14/100 blog posts) (@metakuna): just got the worst haircut of my life, maybe occupational licensing is good actually

Did u get it from a licensed barber tho?

Reply to his own earlier tweet, 18:42 UTC · 3 likes

Aaron Bergman, December 18, 2022: 2) Still quite AI-pilled, but down from my peak by like 7.5% 🥱 <more to come but the 1 mg of melatonin is hitting hard rn😴>

3) ik this is vague + generic but:

Increasingly believe that success and performance at anything is mostly dumb luck (often genetic) with some room for intentional high-level tool selection. But effort or what subjectively feels like hard work is a tweak on the margin at best

Tweet, 23:54 UTC · 3 likes

Darn I kinda like(d?) Twitter :/

Monday, December 19, 2022

Tweet, 10:17 UTC · 3 likes

what is a server? a universe? a to[o]ot[er]?? Is '.social' a normal web domain? Is there at least an analogy?

Is M_____n is like a blockchain/file extension/operating system/programming language/internet/Twitter/Github/Drive/Unicode/goddamn tennis ball? please?!

Reply to @GjMcGowan, 11:52 UTC · 9 likes

George McGowan (@GjMcGowan): By volume, your fridge is mostly

Probably gas

Reply to @allthefrensy, 15:35 UTC · 17 likes

(The tweet being replied to by @allthefrensy is no longer available.)

Emma Darwin: “I wish Chuck would cool it a bit with the birdwatching”

Reply to @StefanFSchubert, 15:37 UTC · 2 likes

Stefan Schubert (@StefanFSchubert): The good old days when work only had a one percent fatality rate

image attached

Pretty responsible including both numerator and denominator, still hasn’t fully caught on

Reply to @EzraJNewman, 20:28 UTC · 3 likes

(The tweet being replied to by @EzraJNewman is no longer available.)

Ha, unambiguously the middle option

Reply to @tszzl, 21:18 UTC · 11 likes

roon (@tszzl): move fast break things era is over. it’s time to think deeply and make generationally good decisions

What has this man seen? DiscussGPT?

Reply to @astridwilde, 21:25 UTC · 1 like

Astrid Wilde 🌞 (@astridwilde): it's the FTX version of Facebook's Little Red Book

Ok what’s FB’s LRB?

Reply to @austinc3301, 21:27 UTC · 2 likes

Agus 🔸 (@austinc3301): @CineraVerinia I would be extremely wary of anyone that proclaims their community just happens to have the best ideas, even if you've gone beyond

Disagree when the community in question is voluntary and intellectual - if so it wouldn’t just “happen” to have the best ideas; it would be selected for that very reason

Tuesday, December 20, 2022

Tweet, 01:11 UTC · 7 likes

Boring take but Twitter is a good instance of a basically good social media functional framework and, after adding a dash of laziness and convenience, I hope it remains the dominant Twitter-like-thing

Reply to @peterwildeford, 08:10 UTC · 2 likes

Peter Wildeford🇺🇸🚀 (@peterwildeford): @AaronBergman18 @frances__lorenz @Mihonarium the answer is (as always) @RethinkPriors

Started trying to think of a clever reply, a few tabs later... are Comstock and his papers like well known in the shrimp moral patienthood quantitative evaluation community? https://www.researchgate.net/publication/353852027_Pain_in_decapod_crustaceans_A_novel_solution

https://www.researchgate.net/publication/361451600_The_Shrimp_Hypothesis

Surprised that these dont seem to be referenced...

Thread, continued, 08:10 UTC

on the Forum at all (or Comstock at all)

Reply to @sam_atis, 12:20 UTC · 48 likes

(The tweet being replied to by @sam_atis is no longer available.)

Twist: she works with EAs, the speech disorder is when you can’t stop caveating every expression with “in expectation” or some version of “plausible”, and the reply was both intended and received as high praise

Quote tweet, 15:46 UTC · 2 likes

Update: I think I can now take a list of urls and output a max quality single merged pdf of how each page looks when fully loaded (there's a reason I wanted this haha). Think "every SSC post in a 1GB, 100k page pdf"

Afaik there's no [user-friendly] app or site that can do this?

Quoting Aaron Bergman 🔍 (@AaronBergman18): I seriously think someone (in fact, someone reading this if they wanted to) could make big tech-level money by getting really good at making custom shortcuts and Automator scripts for clients on iOS.

Plus you could live wherever, donate 75% of your income, be own boss, etc.

Tweet, 19:59 UTC · 134 likes

It’s a Thing that kids ask how babies are made, but tbh I never understood this because, like, there was never a question in need of an answer!

Wasn’t it the obvious default hypothesis that women just spontaneously start growing babies around a certain age?

Tweet, 19:59 UTC · 60 likes

This would explain all the observations an 8yo (who knows pregnancy is a thing, but not sex) might have: why out of wedlock births are Bad (she couldn’t find a husband *in time*), why kids look like their dad (doesn’t run into *that* many dads who look super different from mom)

Thread, continued, 19:59 UTC · 59 likes

Adult teeth grow in spontaneously at a certain age, people grow, plants have seeds, etc.

For a wrong hypothesis it sure fits the data!

Tweet, 22:41 UTC · 7 likes

Impulse-bought a 6 sq foot poster version of this (slightly cropped), I am very excited. From http://www.pedworld.ch/ENTREE/BIOCHIMIE/metabolic_pathways_poster.pdf

Image

Wednesday, December 21, 2022

Quote tweet, 01:02 UTC · 2 likes

Something this drives home:

All this complexity, bajillions of distinct processes, enzymes, molecules etc. are to even a *2nd* approximation shuffling around exactly 5 things: carbon, hydrogen, oxygen, phosphorus, and nitrogen atoms.

The odd sulfur and iron gets you to 7

Image

Image

Image

Image

Quoting Aaron Bergman 🔍 (@AaronBergman18): Impulse-bought a 6 sq foot poster version of this (slightly cropped), I am very excited. From http://www.pedworld.ch/ENTREE/BIOCHIMIE/metabolic_pathways_poster.pdf

image attached

Quote tweet, 10:35 UTC · 2 likes

Maybe we should retvrn

Quoting $700k TC Staff Engineer (@KotlinChad): @skooookum I do a lot of work as a senior engineer on a single 13” MacBook Pro, no external monitors or anything. TC over $400k btw

Thursday, December 22, 2022

Reply to his own earlier tweet, 01:45 UTC · 2 likes

Aaron Bergman, December 18, 2022: 3) ik this is vague + generic but: Increasingly believe that success and performance at anything is mostly dumb luck (often genetic) with some room for…

ooh forgot about this thread...

4) Another vague and generic one but:

this year I sorta kinda started peaking behind the curtain of various computery things, and feel like I have a better idea in some cases how they work. It's also meant an appreciation of how damn complex...

Thread, continued, 01:47 UTC · 2 likes

and impressive a lot of software is. Like for instance if you take a random substack post and trace back all the HTML/CSS/JS code to any code they cite (like JS libraries I think) and iterate until there are no levels of citations, you're looking at at least tens of thousands and

Reply to @BootsPlatypus, 01:59 UTC · 2 likes

Platypussy in Bio (@BootsPlatypus): @AaronBergman18 How does one even get into this? Did you intentionally set out to develop an interest in the most intimidating field 😛

Lol😆 mostly troubleshooting random personal medical issues. Also I wouldn't say I'm that interested in metabolism or microbiology generally - only insofar as they relate to something happening in the brain (which can def run through things like nutrition and general health)

Reply to @parafactual, 02:06 UTC · 1 like

You (@parafactual): @AaronBergman18 wait how

I think this is the core of the argument

Image

Quote tweet, 02:11 UTC · 2 likes

I think this is the core of my argument for the QT claim

Probably shouldn't stay buried [checks notes] 3,984 words into a not-super-popular Forum post

Image

Quoting Aaron Bergman 🔍 (@AaronBergman18): 1) Claim: “Total utilitarianism does not imply that any amount of suffering can be offset or morally justified by even an arbitrarily large amount of wellbeing”

Epistemic status: Ive thought about this one a lot (prob more than any other q) and am pretty damn sure at this point

Friday, December 23, 2022

Tweet, 01:05 UTC · 3 likes

Twitter, help: I have a very simple interactive chart application as a website (http://aaron-bergman.shinyapps.io/topics_data2/ for EA forum nerds! ) which I can replicate by running one local file even w/ wifi off

When I save the site as "Webpage, complete" I get the html and bunch of js and css files

Thread, continued, 01:05 UTC · 2 likes

(which I didn't...write) which I can *also* use to generate the page...except this time it behaves like a fillable form that doesn't do anything else

So what's missing? How would you get the site to run locally if all you had to start was the url?

Saturday, December 24, 2022

Reply to @johncoogan, 00:55 UTC · 2 likes

John Coogan (@johncoogan): If you think Lex Fridman lacks the proper academic credentials to host 3 hour podcasts… I have some bad news for you about Joe Rogan.

Rogan is a good interviewer

Reply to @SpacedOutMatt, 03:00 UTC · 3 likes

Matt 🔸 (@SpacedOutMatt): Shrimp welfare advocates in tears right now

Maybe she raised these in an extra spacious fish tank with lots of oxygen before euthanizing them with heroin

Reply, 03:02 UTC · 13 likes

(The tweet being replied to is no longer available.)

Wait but I think dietary cholesterol really does support testosterone synthesis

Reply to @AaronBergman18, 03:57 UTC

(The tweet being replied to by @AaronBergman18 is no longer available.)

…most of the empirical correlation is probably due to dietary saturated fat, and no studies really hone in on dietary cholesterol alone per se as the single treatment variable

Tweet, 05:02 UTC · 6 likes

Twitter showing views is a textbook discontinuity analysis natural experiment: there’s gonna be a big csv somewhere with 200M user’s tweet stats before and after the change

Prediction: small but significant negative effect on mean and median likes/view

Reply to @AndyMasley, 05:03 UTC · 10 likes

Andy Masley (@AndyMasley): Strong disagree. Everyone at any professional conference should be treated as a coworker and that's it. There's already way too much blurring of professional/personal in general.

This but only so I have a good excuse for being single

Reply, 05:07 UTC · 7 likes

Contra the “bad publicity” replies: smallish neolib twitter accounts shouldn’t be optimizing to convince the median American president election voter. ~all true and important claims ought to be said *somewhere*, and it’s hard to think of a better place than “small neolib account”

Reply to @PradyuPrasad, 05:11 UTC · 2 likes

Pradyumna 🇸🇬 (@PradyuPrasad): It doesn't matter if its true or not. It is grossly offensive. There are social boundaries you shouldn't step beyond, and they exist for a *very* good reason. Once again, I'm imploring you to delete it

I honestly don’t find it grossly offensive (and fwiw Jewish family, bar mitzvah, etc)

It could be offensive or merely cringe/bad taste if said in a different setting ofc

Reply to @MaxGhenis, 05:14 UTC · 2 likes

Max Ghenis (@MaxGhenis): @realjudeleelind I'm pretty skeptical of this but would read an analysis quantifying the case

Not very rigorous but

Reply to @austinc3301, 06:19 UTC · 3 likes

Agus 🔸 (@austinc3301): @AaronBergman18 @realjudeleelind I think there are plenty of ways to convey the important facts behind this statement without sparking controversy over holocaust denialism, and it doesn't seem like there's really a tradeoff

I honestly don’t see any denialism here; the strength of the claim (semantic and rhetorical) depends on it being common knowledge between writer and reader that the Holocaust is real and horrible

Reply to @austinc3301, 06:19 UTC · 3 likes

Agus 🔸 (@austinc3301): @SpacedOutMatt @AaronBergman18 @maiamindel that's just the leftover hedons for the consumer 😭

A lil extra flavor too

Tweet, 17:38 UTC · 5 likes

Blue tribe vs red tribe: the final battle (over whether dietary cholesterol increases testosterone production)

Image

Reply to @EzraJNewman, 17:40 UTC · 2 likes

(The tweet being replied to by @EzraJNewman is no longer available.)

Lmao

Quote tweet, 17:43 UTC · 2 likes

Can we get a team of intellectually-disagreeable LWers on this one?

Quoting Christopher K. Mellon (@ChrisKMellon): The lack of news coverage compels me to point out that President Biden has just signed into law far-reaching legislation that could solve the UAP mystery: https://www.christophermellon.net/post/unprecedented-uap-legislation

Quote tweet, 18:09 UTC · 3 likes

Everyone dunking on this for the wrong reason: unless you’ve got a super nice heating system, you’re just not gonna make every room in a house 70° if it’s 15 outside

Concentrated heat means *more* heat where you are

Quoting Alaric (@0xAlaric): When leftists post stuff like this, it’s a deep betrayal of their psychology.

The inadvertently say the quiet part out loud.

1/

image attached

Thread, continued, 18:09 UTC · 2 likes

14 outside, 52 in rest of house, chillin in shorts with a window cracked (CO2 ofc) in a space heater-made sauna 😎

Image

Reply to @mealreplacer, 23:57 UTC · 1 like

(The tweet being replied to by @mealreplacer is no longer available.)

https://wtfhappenedin1971.com/

Sunday, December 25, 2022

Reply, 08:31 UTC · 3 likes

This is both annoying and doesn’t make sense lmao

Reply to @IgorBrigadir, 09:00 UTC · 2 likes

Igor Brigadir 🇺🇦 (@IgorBrigadir): It blew my mind when i first learned about approximate counting - it's genuinely extremely difficult to internalize - what could possibly be simpler than counting?!

(quoting @moyix: It’s a dirty secret of the industry that all counts in a distributed system are to some degree fake)

Wait what

Monday, December 26, 2022

Tweet, 10:57 UTC · 12 likes

Sorta in honor of Christmas:

What’s the probability that the basic scientific metaphysics/worldview is mostly wrong e.g.,:
- there’s a full-on quasi-anthropomorphic diety
- we’re in/are a simulation
- universe< 1M years
- “history” didn’t happen
- Truman Show type of thing

Thread, continued, 10:57 UTC · 2 likes

Honestly not sure how to divide up the probabilities - might put another poll below if results are concentrated

Reply to @JesseTayRiver, 19:30 UTC · 2 likes

Jesse Smith (@JesseTayRiver): @AaronBergman18 I'm not on lesswrong so how does the 2nd contradict this worldview?

Doesn’t necessarily contradict, I was trying to include “extremely incomplete” as a subset of “wrong”

Thursday, December 29, 2022

Quote tweet, 02:37 UTC · 14 likes

Lfg, big pharma is good now

(dw current placement is temporary)

Image

Quoting Aaron Bergman 🔍 (@AaronBergman18): Valuable info right here 🧫🧬🔬

Friday, December 30, 2022

Reply to @Laura_k_Duffy, 03:47 UTC · 2 likes

Laura Muresianu 🌐🔸🇺🇸 (@Laura_k_Duffy): AMA to celebrate eleventy-one followers (and bc everyone else’s doing it and it seems fun!)

Why great falls Montana?

Thread, continued, 03:48 UTC · 2 likes

No disrespect to the third most populous city in the great State of Montana ofc

Image

Reply to @mealreplacer, 03:53 UTC · 6 likes

(The tweet being replied to by @mealreplacer is no longer available.)

Guy who doesn’t download his hinge data spreadsheet


End of December 2022. Previous month: November 2022. Next month: January 2023. Back to the archive index.