Re: Review situation for boxes

["Zeeshan Ali (Khattak)" <zeeshanak gnome org>]

On Mon, Jun 10, 2013 at 6:45 PM, Marc-André Lureau
<marcandre lureau gmail com> wrote:
> Hi
>
> On Mon, Jun 10, 2013 at 5:29 PM, Alexander Larsson <alexl redhat com> wrote:
> > I just got pinged by zeenix to review some patches, and I will have a
> > look at them asap. But in general, I'm pretty busy with other stuff atm,
> > and will not have a lot of time to do reviews. It seems to me like the
> > current system where any change has to be reviewed is going to become
> > problematic with the current number of developers.
> >
> > Very few other gnome projects, including core things like glib and gtk+
> > have such strict review criteria, and I think its time to start thinking
> > about tweaking this. I think review is good and important, but for
> > maintainers to have to block on other maintainers to move forward on
> > non-critical changes is going to be problematic in the long term.
> >
> > I think we need to change this to something more of a honor-system
> > thing, where changes to the unstable branch by maintainers can go in
> > unreviewed, at least for some types of changes. This is an app after
> > all, if something is problematic we can just revert it, there is no ABI
> > guarantees or anything.
> >
> > Of course, changes to stable branches or changes that affect external
> > things like config files, etc should really require review.
> That sounds like the way to go. However, I have certain expectations that
> those who have direct commit to not make changes that wouldn't be approved
> by the rest of the contributors. In doubt, keep working on a seperate branch
> and do a review.

I agree till this part.

> Imho, there are fairly distinct parts in Boxes that each of
> us have more experience with, I wouldn't walk on someone else feets without
> having at least a discussion.

I kinda agree to this too but thing is that changes to one part of the
boxes often require some changes to others. For example patches in
bug#690757 are mostly to code I've mainly written but it also touches
code written by others. Now if I'm to wait for review, we are back to
the original situation which we are trying to solve here.

BTW just to be clear, I don't think Alex meant that this 'direct push'
applies to everyone with write-access. That would be all gnome
developers. :)

--
Regards,

Zeeshan Ali (Khattak)
FSF member#5124