On Mon, Apr 6, 2026 at 11:55 AM Jacob Champion <[email protected]> wrote: > From a mechanical perspective, that has clear advantages to me > (especially with the de facto GitHub interpretation), but I think it'd > collide with our practice of rewriting commits to maintain project > voice. Maybe people could get used to that change, but I generally > expect the Author in the Git metadata to be the *literal* author of > the commit message.
Yes, I think that's right. I would have no problem us allowing pushing of commits under the actual author's name if the commit is pushed unchanged, but I rarely push anything unchanged and I think people would be very quickly become unhappy if I started doing so. In the rare cases where that would be warranted, the person usually just gets made a committer anyway. But really, that's a discussion for another time. The discussion here is whether we're going to interpret the authorship information in the existing commits in the way that the committers who created those commits intended, or whether, as Bruce proposes, we're going to do something else. -- Robert Haas EDB: http://www.enterprisedb.com
