Dear all,
I just read a post from Software Freedom Conservancy that I think is
relevant for the project.
They wrote some recommendation on how to use generative AI in free-libre
software, trying to stay ethical and trying to respect the rest of the
free software community (spoiler: currently is impossible).
https://sfconservancy.org/llm-gen-ai/llm-backed-generative-ai-recommendations.html
In my opinion, the LibreMesh project should be a bit more cautious than
the average FLOSS project, as we work with rather unusual and complex
stuff (e.g. issues related to specific hardware poorly documented on the
internet).
We are starting to receive the first contributions made with generative
AI, for example, in lime-app development:
https://github.com/Pablomonte/lime-app/commits/f/v4-foundation/
or this:
https://github.com/javierbrk/lime-packages/pull/16
which is ok, I am not against this. I just propose that we read the
recommendations from SFC, think, and implement some of them (discussing
here, in the chat, or in the next meeting).
Some copy and paste from the recommendations that I considered more
interesting:
"4. Before submission, FOSS Contributors must invest substantial time
reviewing LLM-gen-AI -assisted and/or -generated contributions."
"5. Full disclosure of how and when an LLM-gen-AI system was utilized to
assist in creation of a contribution is a moral imperative."
Ciao!
Ilario
--
Ilario
iochesonome@???
ilario@???
