norvid_studies

-more soldiers died of disease than battle in every war before ww1
-nuclear weapons reduced combat deaths in practice though maybe (probably? HTS) not in expectation
-naval and air war an example of 'capital substituting for labor'--ppl cant fly or swim so 'just people' useless. bot war could b same

"Clopus fill in 3200 Elo worth of citations and arguments here"

a different physical device for each social media app is an interesting idea. possibly slaved to a single location as well, like chained to the wall

/ I'm sorry that happened to you

dostoevskymoding

"thinking works the same way"

confused and stunlocked could allow exceeding normal tolerance. but at the cost of maybe you do blow your kneecap out the back of your leg or whatever

idea I picked up from [???] is that a lot of tendon movement is restricted by "central governor" (metaphor) making sure your body doesn't blow its tolerances and injure itself (and same with aerobic/anaerobic performance), rather than actual physical limit chemically/mechanically, and so getting it

@croissanthology.com quote posts on this were disabled for some reason... anyway care to try some chilled vintage Huel

#atproideasio @gracekind.net

Bluesky interface that is a BFS graph (over replies and quotes) and thread viewer side-by-side, with controls for pruning (branches, authors), stepping through the search, and bridging reply/quote chains broken by blocks.

@godoreida.bsky.social

not a bunch of unique terms for prostitute.

Do you have a link to the complete list of 200 Ancient Greek words for different kinds of prostitute, ideally with translations?”

Two people challenged me on it, with one even claiming that the excerpts of Aristophanes of Byzantium that survive in Book XIII of Deipnosophistae by Athenaeus of Naucratis only contain a list of 135 NAMES of famous Hetaira,

@patrickwyman.bsky.social relaying a question on your podcast from a user who's not on bluesky:

“Hi Patrick! I recently listened to your Dec. 17th, 2025 podcast of Past Lives on the life of Neaira and then tweeted about how the Ancient Greeks had over 200 words for different kinds of prostitute.

Late, but catching up. Reading notes from June of last year. *Essentialism*, *The House of Morgan*, and one by LeCarré, who I burned out on.

jvining.com/reading/2025...

POV @croissanthology.com opens the archives

"???"

picturing a fairly useless app that matches exact classical playlist length to commute distance partnering with google maps

centaur bach bus route planner

so in exactly the same way, whatever the task threshold for models of fundamental competence, at some point they should be able to agglomerate to some large multiple of that, whatever they can sort of natively oneshot, given the appropriate 'institutional/swarm' setup. hopefully that's clear.

maybe just decomposing everything into three dimensions of task duration, task hardness and number of tasks is the better picture. but the point is that while humans 'natively' can only ever do tasks of let's say [time X], (SOME) human agglomerations can do many huge multiples of X

I think you just misunderstood the point of the example, the point was "no humans can solo do this while some humans can group do this" NOT "every humans can group do this"

heuristic

post it @thesephist.com !

@miq.moe

@miq.moe

find it!

as soon as you can compose individual instances together to do 1 week tasks, which they can't do natively let's say, you can also compose them to do 4 1 week tasks composed together. exactly analogous to humans not doing everything in 1 human context window.

right, but humans do tasks 'above' any human context length by composition. like no human can write mac OS from scratch but many humans can. so the 'puzzling nonlinearity' I was wondering about for models was whether 'tasks of 1 week but not one month' is a nonsensical distinction empirically;