Quoting Jason Ekstrand (2018-11-28 11:30:32) > Yes, but the point is that we (the reviewers) know that we're conflicting. > That's very different from what I could easily see happening *a lot* were ML > reviewer A is perfectly happy with some bit of code but MR reviewer B asks for > it to be completely reworked. In v2 of the series, the submitter reworks it > but now reviewer A is unhappy. "Why did you change it?" he says, "It was just > fine before!". "Reviewer B requested the rework," says the submitter. "When > did he say that? I didn't see that comment." says B. "On the GitLab MR," > says > the submitter. "Well, I don't read MRs; this kind of feedback should happen > on > the list where we can all read it," says A. > > If you can't immagine that exchange happening, then you haven't been on this > list long enough. :-) (Says a guy who's been on the list for about half as > long as Jordan.) > > We have enough stubborn people on the list that MRs are going to constantly > get > pulled back to the list just because someone doesn't want to use the web > interface. That's really mean to submitters who actually want to use the MR > process and is strictly worse than what we have today. If we're going to > actually try out MRs, we need those people trying it too at least from the > reviewer side.
This is exactly my concern, and the reason I think we need to be "all in" on one or the other. I have a preferences for MRs, but I'm fine with continuing to use the mailing list. This is (IMHO) a case where both is strictly worse than either choice. Dylan
signature.asc
Description: signature
_______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
