On Mon, Apr 6, 2026 at 10:56:49AM -0400, Andrew Dunstan wrote:
>
> On 2026-04-06 Mo 10:29 AM, Bruce Momjian wrote:
>
> I think having "Co-authored-by:" mean one thing when "Author" appears
> and a different thing when "Author" is missing is too confusing.
>
>
>
> Possibly. I think we're tying ourselves up in knots needlessly here, though. To
> me, without having to interpret the exact meaning by consulting a wiki,
> Co-authored-by signifies that the person made a significant contribution, but
> not as much as the Author(s). These things shouldn't be technical terms of art.
>
> Personally, I'm in favor of being fairly liberal about giving release note
> credits.
So "Co-authored-by:" shows a level of involvement, but doesn't have any
effect on the major release notes. That works too.
"Liberal" here means give it even for the lesser contributions. They should appear in the release notes.
If everyone explicitly lists every author using the author tag for non-committer-only commits the rule that all authors are equal applies and we can move one with that preferred wording. Co-authors becomes unnecessary. But the usage as it stands historically is that co-authors are authors and if a commit doesn't have an explicit author the committer is one. We can leave that stand as historical and when people fall back on old habits.
Maybe add Assisted-by if we want to introduce a intermediate level between author and reviewer. It does seem we failed to make that be co-author and redefining should be avoided.
David J.
