Re: GR: Ban LLM contributions from Debian

Lucas Nussbaum writes ("Re: GR: Ban LLM contributions from Debian"):
> So, combined with (4) and (5), I understand that "We request" here means
> something similar to "We recommend" or "We strongly advise", not "It is
> required". Maybe it would be clearer to title that part
> "RECOMMENDATIONS", and use "strongly recommend" instead of "request" in
> 1, 2, 3?

"Recommend" (and "advice") have very different meaning to "request".
I think "request" is clear that whether to do as requested is
optional.

> Shouldn't you add commit messages here? Especially since they are a
> starting point for discussions, for example in code reviews.

I would love to.  However, unfortuantely the authorship and intended
audience for commit messages is quite different and much broader than
that of Debian-specific communications.  Commit messages are,
essentially, code.  We share them with our upstreams in both
directions.

I deliberately left code out of my prohibition because (1) it leads to
a double standard for Debian-authored versus upstream code (which
would be fine by me but leads obviously to accusations of lack of
principle) and (2) it would lead to logic-chopping and
fake-hat-switching, where maintainers would claim their commits were
not written as part of Debian but then merely imported.

Ian.