Re: [PATCH] t4013: add expected failure for "log --patch --no-patch"

[Felipe Contreras <felipe.contreras@gmail.com>]

Sergey Organov wrote:
> Felipe Contreras <felipe.contreras@gmail.com> writes:
> > Sergey Organov wrote:
> >> Junio C Hamano <gitster@pobox.com> writes:
> >
> >> > When a command does not behave the way one thinks it should, being
> >> > curious is good.  Reporting it as a potential bug is also good.  But
> >> > it would help the project more if it was triaged before reporting it
> >> > as a potential bug, if the reporter is capable of doing so.  Those
> >> > who encounter behaviour unexpected to them are more numerous than
> >> > those who can report it as a potential bug (many people are not
> >> > equipped to write a good bug report), and those who can triage and
> >> > diagnose a bug report are fewer.  Those who can come up with a
> >> > solution is even more scarse.
> >>
> >> I'm afraid the solution I'd come up with won't be welcomed.
> >
> > My solutions are often not welcomed, and yet I still implement them.
> >
> > It might be a waste of time, but often I've found out that very quickly
> > after attempting to come up with a solution I realize there's a lot of
> > detail I was missing initially, so even if the solution is not welcomed,
> > it helps me to understand the problem space and be more helpful in the
> > discussion of potential solutions.
> >
> > So if I were you, I would still attempt to do it, just to gather some
> > understanding.
> 
> I sympathize, and I did recently. However, I figure I'd rather spend my
> time elsewhere, say, in the Linux kernel, where my experience is
> somewhat different, and allows me to enjoy my work.

Completely agree.

My experience in the Linux project is that of a true meritocracy: Linus
Torvalds doesn't have to like me, if the patch is good, it gets merged. Period.

> > I for one welcome any and all attempts to fix git's awful user
> > interface, regardless of the reception of the maintainer, and the "core
> > club".
> 
> For UI, the problem is that there is no core model defined, nor any
> guidelines are given, so every discussion ends-up being what "makes
> sense" and what doesn't for a user, everyone involved having his own
> preference, that often even changes over time.
> 
> In this situation attempting to fix the UI sounds like waste of efforts,
> as nobody can actually point at the state of the UI to which we are
> willing to converge, so there are no objective criteria for accepting of
> fixup patches.

It's even worse than that. There used to be objective criteria like the old Git
User's Surveys [1], but it turned out Git developers did not care about the
feedback from users, which is why there wasn't any point in continuing them.

And worse: even when all Git developers agree on a UI change, except one, it
doesn't matter, because that one has absolute veto power.

Not very hopeful prospects for Git's UI.

Cheers.

[1] https://archive.kernel.org/oldwiki/git.wiki.kernel.org/index.php/GitSurvey2016.html

-- 
Felipe Contreras