Re: PG 19 release notes and authors

On Mon, Apr 6, 2026 at 08:43:00AM -0700, Jacob Champion wrote:
> On Mon, Apr 6, 2026 at 7:47 AM John Naylor <johncnaylorls(at)gmail(dot)com> wrote:
> > That's not how I interpreted it at all, and after seeing commits with
> > both "Author" and "Co-authored-by" I am equally confused as to how
> > people are interpreting it.
>
> In case it helps, here's what I had always assumed the meanings were
> without consulting the wiki, with links to commits I've made so you
> can roast my usage.
>
> - "Author" overrides the default assumption, which is that the
> committer was the author of the patch: https://postgr.es/c/a6483f5ac
> - "Co-authored-by" lists co-authors, who share attribution in some
> unspecified way. (GitHub adds a weak mechanical effect to this tag.)
> https://postgr.es/c/993368113
> - Some people list multiple Author: lines as an alternative to
> Co-authored-by:, which never particularly bothered me.
> - If attribution is more complex than that, people just say that in
> the body of the message: https://postgr.es/c/c2bca7cc9

What you have said above is the way I think most committers have done it
for PG 19. If some have not, it would be good to tell me now. Also, if
we want to change it going forward, that would be good to know.

> In particular, if I don't want official "credit" in the release notes
> for minor changes I made to a patch during commit, I don't need to add
> any tag at all. I just mention that I changed the patch, following a
> style I've seen from Tom and others: https://postgr.es/c/e020a897e

Yes, I have seen that.

> > My take is that the co-author tag has backfired and made things less
> > clear. If we are using it inconsistently, then it doesn't convey any
> > useful information.
>
> It conveys *attribution*, regardless of whether or not it's used
> consistently for a mechanical purpose. I'm willing to bet that "I
> coauthored this patch" has intuitive meaning to most people, inside
> and outside this project.
>
> I'm glad the momentum appears to be in favor of keeping that
> attribution, because the idea that we'd retroactively discard it
> seems... misguided, to me. This is going to be fuzzy in complex cases,
> but it's okay to just write the complexity longhand when needed,
> right?

Yes.

--
Bruce Momjian <bruce(at)momjian(dot)us> https://momjian.us
EDB https://enterprisedb.com

Do not let urgent matters crowd out time for investment in the future.