Re: master 4803fba487 1/2: 'C-x v v' on a diff buffer commits it as a patch (bug#52349)

[Eli Zaretskii]

> From: Sean Whitton <spwhitton@spwhitton.name>
> Date: Tue, 30 Aug 2022 07:49:56 -0700
> 
> >> Anyway, the main point is that adepts of any existing VCS or any future
> >> VCS can easily adapt their VCSes for this feature by just implementing
> >> the backend call 'checkin-patch'.
> >
> > I very much hope this is not going to be the official policy of VC
> > development in Emacs.  VC should IMO remain VCS-agnostic in its most
> > important features, and "C-x v v" is definitely such a feature of VC.
> 
> Surely leaving room for things to be implemented for other VCs means
> that it remains VC-agnostic.  With e.g. Magit there isn't even that room.

Yeah, and then let's "leave room" for adepts of documentation to add
it if they feel like it, and for adepts of Lisp style to fix the style
of the code someone else installed, and for adepts of clean design and
implementation to clean up after others, etc. etc.

We have some conventions and some standards in Emacs development, so
I'm surprised to hear that some of them are left to "adepts" to
uphold.  If we go that way, there are good chances that some things
will never happen, because no adepts were in supply, or because we
forgot, or for any number of other reasons.

> > That's basically the only real raison d'être of VC nowadays, because
> > otherwise let's just obsolete VC and switch to Magit.
> 
> VC is far more flexible and Emacs-y than Magit in how it handles
> buffers.

I'm sure frequent users of Magit disagree.

Of course, that's not why I mentioned Magit: I mentioned it because if
we want to cater to a single VCS, a very good solution for that
already exists.