Life is Tetris @tetrislife@qoto.org

The @nlnet GenAI policy is the only one I've come across so far actually making a clear statement on the incompatibility of LLM outputs with open source: https://nlnet.nl/foundation/policies/generativeAI/

Have any of the big license organisations (thinking OSI, FSF, Creative Commons) said anything like this (either pro or anti)? Can LLM-derived code be released under the GPL if you don't know where it really came from?

Or would org funders (I'm thinking of Google, MS, etc here) disapprove of such statements?

@corbet Very tough.

But such content mills can be reported in some places - e.g., https://blog.kagi.com/slopstop

How realistic it is to keep up with the slop? Unclear.

But validated "small web" sites (which is another layer Kagi supports) might be a way. Back to curated Internet site lists as if it's the 90s again!

@mjg59 Reading this and thinking about LuaJIT-FFI's approach, which is that instead of parsing C header files it defines a easily-parseable subset of C and parses that. You wind up editing your header files into long strings and passing them into Lua.

Maybe this "parseable header C" should be a cross-language standard.

@mjg59 We could trigger even more people by stating that using whatever the fuck the rust people cook up will probably be worse than what we have now.

@royaards if only it were a joke: https://rentahuman.ai/

if you happen to have #GooglePlay on some device, please give #ArcaneChat a review so it gets more visibility in the main stream, your sacrifice getting your hands dirty with Google will not be in vain 🎖️

https://play.google.com/store/apps/details?id=com.github.arcanechat

#boostswelcome #pleaseBoost #boost
#foss #openSource #privacy #encryption #messenger

"I am injust"

I still remember trying to sell people on Erlang back at an old job and it went something like “think of it as concurrent functional lisp” and they were like “Lisp’s syntax is terrible” and I was like “then good news because this is Prolog syntax” and that wasn’t better.

#erlang #programming

@cliffle @b0rk A bit of advice I was given long ago when someone proposes doing something apparently dumb and which has been tried before is never to say "that doesn't work" or "we've tried" but to always say "This is what we couldn't figure out last time"

They now know where you came unstuck and can see if they have answers, they know what to think about if they've not got that far, and you've not stopped them if they actually do have the answer.

When I discover that AI can produce this type of OCaml code https://github.com/mtelvers/ocaml-smtpd, which probably comes from what I have been writing for several years, and that investigators ultimately do not take the time to look at what the community can produce... This heralds a bright future for open source where people no longer form a community. I hate AI.

@garius That reason why? One Rich Asshole Called Larry Ellison

@Distante #DeltaChat, it is decentralized, doesn't require phone numbers or any private data, onboarding is smooth much easier than in signal, similar UI (it is a signal fork) it has much better multi-device and multi-profile support, secure in-chat mini-apps like notepads, to-do lists, shopping lists, etc

there are alternatives, but they will not gain momentum if you don't promote them and stop misinforming people selling a centralized silo running on amazon as #sovereign
@leanderlindahl