Ted Chiang: ChatGPT is a blurry JPEG of the web. 👏👏👏, amazing Ted Chiang. Recent AIs (ChatGPT but also DALLE2 & Stable Diffusion) amaze me by the mere compression they achieve. Ted explores this, with a very pertinent reminder of Xerox’s JBIG2 mishap, and asks a million excellent questions. One of the best articles I’ve ever read about ChatGPT.
[pdf] Field Guide to NOTAMs, a tastily-presented snark about the irrelevant crap that goes into official pilot notifications. Said differently, “notification overload”, but for airplane pilots instead of your phone. Via Nelson.
Good reminder by Nick: “Smart” home devices are security & privacy trash. IT’S ALSO RIDICULOUSLY FRAGILE AND MOSTLY-USELESS OBSCENE FIRST-WORLD TECHNO-GADGETO-BABBLE BULLSHIT, GODDAMNIT. MORE TECH ≠ BETTER, ever dared thinking about it?
Scott Alexander: Why I’m less than infinitely hostile to cryptocurrency. Good points that make me reconsider my admitted own infinite hostility to it. One point that Scott doesn’t mention, though, is the ecological disaster that Proof-Of-Work coins imply, but one might hope this will improve (or be fully solved with PoS?!). This is one particular axis of the problem, for which “In 2010, it was probably 100x worse than the regular financial system; now it’s maybe only 10x worse. Might there come a day when it’s no worse at all, or even better?”HN blah.
Pinafore’s (a Mastodon client) interesting “wellness settings”, “designed to reduce the addictive or anxiety-inducing aspects of social media”. Via Nelson.
People Make Games: Valve’s gambling problem. Wow. Had the impression that Valve was less shitty than Epic/Fortnite about gambling (probably due to a positive prior I have about Valve because of the objectively good / ecosystem-healthy work they’re doing with Linux/Proton), and oh boy was I wrong. The whole thing absolutely stinks.