> You must not share projects that mostly consist of code written by "generative AI"-tools (including services such as Claude, OpenAI Codex). Such projects having an unclear copyright status (see requirements § 2 (1) 1 and § 2 (1) 3) and furthermore have little safeguards to ensure that they do not include harmful code (c.f. § 2 (1) 5).
I have read it. This effectively narrows down the use of AI drastically IMO. Which is what I said.
This effectively limits it to stuff like code completion and using it as a chat assistant and doing some medium level refactors.
They should have just completely banned it. As the ambiguity itself causes problems. Even if you use LLMs maybe say 25% of the time, sooner or later that means that a significant portion of your code base is AI generated.
> They want to get rid of slop projects and vibe coders, not any project that has any developer that used any LLM. But again, requires you to read through the actual text to actually get their intention here.
I am not a vibe coder (I am a software dev with almost 20 years experience) and am using Heavy LLM assistance (I am running Qwen 3.8 locally) for some of my projects. I am making sure the code output is decent, well documented, has tests etc. Sure what I am doing is something betweeen "vibe coding" and "proper development" but there would be no way for me in the few hours an evening to even get to the stage I would be at in some of them without LLM assistance.
I would not be able to use codeberg for hosting my project under their rules.
Federation is a people-problem, not so much the resource problem.
Mastodon servers defederate because the people (or rather, the admins, perhaps prompted by the users) of instance A don't want anything to do with the people (who toot at the discretion and policies of the admins) of instance B.
Codeberg's AI position, as stated and voted upon, does not mention resource usage at all. That's completely invisible from the discussion. It is exclusively argued against on accord of 1) unclear copyright ownership, and 2) users have no idea what's in the code, therefore it could be full of security flaws or harmful code.
So they would not want to federate with another instance that used a lot of AI-generated code, even if it claimed an open license, even if that federation caused little-to-no usage of their CI. They would fundamentally disagree with any other instance that allowed mostly-AI-generated projects. They would therefore block such instances, if federation was a thing.
They did not disallow LLM contributions, and they definitely didn't disallow using LLMs to research security vulnerabilities. They only disallowed projects that are majority LLM-written.
Regardless of that, these don't seen to play for the same audience.
Codeberg Terms of Use[^0] makes it clear that it's only for "projects covered by a licence for free and open source software, free and open source hardware, or free cultural works", and "You must not share projects that mostly consist of code written by "generative AI"-tools". They also only provide private repos where the FOSS system need it for their infra or what not, and not just for any purpose.
These are totally fine for Codeberg to do, and it's wonderful for FOSS!!
However it is not a viable place for many projects that are on GitHub/GitLab now and would want to migrate, say: any commercial project, that would need "permission-free" private repos, or when the teams want to decided themselves how much generative AI their developers should use, not their Git-hosting provider...
You've made the same exaggerated claim in two comments. Not only are you not banned from Codeberg if you have used LLMs to help you write code "in any of your projects", but you're not even prohibited from using LLMs to help you develop projects that you host on Codeberg.
https://codeberg.org/Codeberg/org/src/commit/96fac426a32d1ba...
I have read it. This effectively narrows down the use of AI drastically IMO. Which is what I said.
This effectively limits it to stuff like code completion and using it as a chat assistant and doing some medium level refactors.
They should have just completely banned it. As the ambiguity itself causes problems. Even if you use LLMs maybe say 25% of the time, sooner or later that means that a significant portion of your code base is AI generated.
> They want to get rid of slop projects and vibe coders, not any project that has any developer that used any LLM. But again, requires you to read through the actual text to actually get their intention here.
I am not a vibe coder (I am a software dev with almost 20 years experience) and am using Heavy LLM assistance (I am running Qwen 3.8 locally) for some of my projects. I am making sure the code output is decent, well documented, has tests etc. Sure what I am doing is something betweeen "vibe coding" and "proper development" but there would be no way for me in the few hours an evening to even get to the stage I would be at in some of them without LLM assistance.
I would not be able to use codeberg for hosting my project under their rules.