Hello, Am Mon, Jun 29, 2026 at 04:46:56PM +0900 schrieb Maxim Cournoyer: > I think the Motivations section probably needs to be toned down, as > Andreas also suggested; there are diverging views on it in the > discussions, and while it's not part of the proposed commitments/policy > per see, it's still part of the final document that will be > shared/linked on the Guix blog for example and appear as if everybody > agreed on it as a whole.
does this refer to the latest version? It has been considerably toned down, even starting with a positive view of LLMs. There are "some" and "others", so everybody should be able to see themselves somewhere there (or in both; the views are not totally contradictory). > It seems the most disagreeable points are also in the commitments, such > as point 1, which appears to discourage any use of LLM for any Guix > contributors As pointed out by Ludovic, the commitment to not use LLMs only mentions "authoring" and "interacting", so it is not "any use". Personally I regret that the "will not encourage" half-sentence has been dropped, and am opposed to any further watering down of the proposal. For me this point is the central commitment of the proposal; the one that prevents the floodgates of AI slop being opened. > While I will be away myself too during that period, I doubt the reduced > time to discuss this difficult topic will help converging toward a > final, agreeable document. Perhaps, the GCD should be withdrawn and > resubmitted at a later date, e.g. in September when people are back from > vacation and re-energized? My reply is a more general comment on withdrawing or not GCDs. I think that unless there is obviously no consensus, one should keep a GCD and let the team members deliberate; only then do we know what people really think beyond the most vocal ones. I do not see much use in withdrawing a proposal; the "worst" that can happen is that the proposal ends up being withdrawn anyway, but at least people had a voice. And we must also think of the authors: Writing a GCD, reacting to comments, rewriting to try to capture consensus is draining. The timeline ensures there will be an outcome and also protects the authors, who (from my own experience) are unlikely to feel energised for a second round. Andreas
