i hadn't "taken poison on purpose and just full on vibe coded as a bit" in awhile, so i gave it a whirl trying to mimic the manic intensity of some of the PRs i see nowadays. i guess i didn't really appreciate how anyone who talks about using LLMs for code is eliding the fact that they are stupendously wealthy or otherwise have access to resources, because doing some stuff i do for free for like an hour on cheap mode cost me $15, which is the maximum i will spent on a bit of this caliber. if i was actually "working" with that there's no way it would cost me less than $200 a day to simply do things i do on my own now.
Here's another #FEP for representing torrents on activitypub :) short, sweet, and with a reference implementation and tests! towards a federated bittorrent tracker with #sciop ! PR: Discussion: (or this thread) #FEP_d8c8 #BitTorrentOverActivityPub #FederatedP2P #BitTorrent
it's never too late to drop out and dedicate your last academic breath spitting sand in the gears of the machine so that new fertile ground might bloom View quoted note β†’
I can't hang on bsky because it brings back the twitter in me. I saw someone saying that Wikipedia is bloated and inefficient because "why spend $100m on labor when the hosting is only $3m." Like a) thats an absolute bargain for a site that serves free high quality information in 27b views a month, and b) how do you think hosting costs are that cheap? Ive seen bloated nonprofits, but look at wikimedia's 990 - the top execs are taking like $400k. Thats half what society for neuroscience's CEO makes for a way more complicated and impactful organization
question: "is the title field in a work registered in the DOI system a cost-effective means of storage" answer: curl https://api.datacite.org/dois/10.57874/531s-1f11 | jq -r '.data.attributes.titles[0].title' | base64 --decode | play -t mp3 - https://www.octopus.ac/publications/n6k2-p549/versions/1