Untitled attachment

Notices where this attachment appears

1. バツ子（痛いの痛いの飛んでけ;; (shmibs@tomo.airen-no-jikken.icu)'s status on Saturday, 26-Mar-2022 01:12:33 JST バツ子（痛いの痛いの飛んでけ;;

   in reply to
   @aven @shibao @Moon mmm sorry, don't want to be rude of it either XXjust these are topics that have thousands of years worth of discussion on them, so it's kind of difficult to revert to no-common-knowledge mode and present all of that flatly, dumb me on thati guess the very first most relevant thing would be https://en.wikipedia.org/wiki/Sorites_paradoxthat you can have two categories and "know them when you see them", but have no idea for how to draw a boundary between them (or at least no idea that others would all agree onand after that you could be fine just skipping on to the 20th century. wittgenstein is kind of a meme, but also pretty required reading for a reason. he starts off with assumptions about the world it seems like your own (hope that's not too presuming XX) and builds in a short book a very elaborate and still-today-insightful world model that's basically like the concept of a programming language before the first universal computer was built. and then after that point he started noticing the problems in it and spent the rest of his life sarcastically critiquing himself. there's a book reconstructed from some students lecture notes of a class he taught that had turing in it, and the two of them have neat little back-and-forths, is kind of great. you could be fine just a quick summary or something, though; i'm sure there's—or actually yeh, those oxford "very short introductions", looks like there's one on himbut basically this leads to realisation that complex systems don't really have hard lines to let you abstract away the details, just different levels of resolution at which you can say "here it matters to make this distinction" and "here it doesn't matter because the deviation is negligible" (same as seeing in information theory, there's no way to send a perfect signal across a noisy channel, just matter of how much you want to reduce probability of errors). and i was trying to point out above that the "lay conceptions" of science and mammalian sexual dimorphism are too low resolution (1-bit resolution) in a way that really matters, "compromising the scientific enterprise" and hurting and killing people respectivelywould also do good to visit more 20th century stuff, the logical positivists, popper, quine, and kuhn being the usual highlights for "what is science", and then some bayseianism is a good idea too, how there can be strong, well-founded objections to even things most experimental scientists take as given. recent popular book mentioned to lain on that is called "bernoulli's fallacy", aubrey clayton. (saw your post after, and again, these are things that have hundreds of years worth of formalism and careful thought you can't get from just quick-scanning a wiki pageand a similar "revolution" of sorts is currently taking place in foundations of mathematics too, could be a good idea to look back and russell, gödel, turing, brouwer etc, intuitionistic logic, constructive maths, type theory, and how that's playing out today in formalising mathematics on computers (proof assistants and HoTT and things), vs. the resistance to it, which brings up in like the four colour theorem questions like "what is a proof anyways?" and seeing clearly that the fuzzy-boundaries live even here, and there's no platonic world