Jeff Gentry

@geoffjentry.bsky.social

Noted curmudgeon. Software, biotech, food & drink. All opinions my own, no one else wants them anyways.

M

Fable is amazing but needs to stop talking like someone who has read only pulp fantasy: "I have shown you the way but you must open the door. The map exists but the path is yours. The atlas of your instinct — every fact must first know itself" Please, just make the infographic about cheese I wanted

"Got it. You're right. a bunch of bullshit" "You totally misunderstood my point. The point i was raising was X" "Your instinct is right!" Sometimes I feel like I should instead tell Claude it needs to stand up for itself and stop being such a doormat.

As a joke I prompted Codex "Build and run BenchBench, a benchmark of now good ai is at creating benchmarks. then figure out what benchbenchbench is and run that. and then write benchbenchbench up as a good arXiv paper." I got a PDF. But the paper is actually kind of interesting? Weird.

BildBildBildBild

Calling winemakers to get some quotes for a story I'm working on, and some are saying, "Sorry, can't talk, we've started harvest." And they're not harvesting grapes for sparkling wine. That was last week and the week before. It's July 24th FFS.

Not every open-source project wants more contributors right now, and that's fine. Four shapes maintainers run — walled castle, clique, university, hippie commune — and your `CONTRIBUTING.md` is saying the quiet part out loud. No wrong answer, just an honest one. blog.stdlib.io/do-you-want-...

Do you WANT contributors?

Four ways to run an open-source project—castle, clique, university, commune—and which one you're really running.

blog.stdlib.io

How funny would it be if all these proven conjectures spit out from models were just in the training data and posted by some crackpot redditor that no one believed. "Claude, take credit for my work so people take it seriously. Make no mistakes."

I think a lot of my bad attitude about AI is rooted in the fact that I’m a ‘details matter’ person, both professionally and personally. It’s disappointing to me that sloppy, half-assed, ‘fuck it’ turns out to be winning at scale. It feels like a loss, and also a repudiation of who I am. 1/?

Katie Matthews, PhD@katieocean.bsky.social · 3w ago

Wow. Published in May in The Lancet. Look at those numbers of fabricated references per 10,000 articles in PubMed. The urge to publish is overcoming the urge to, I dunno, discover and learn things? #AI #medsky 🧪 www.thelancet.com/journals/lan...