On "gardening" code reviews

[Albert Astals Cid]

El dissabte, 18 de febrer de 2023, a les 23:15:07 (CET), AnnoyingRains va 
escriure:
> To clarify, this is the MR process:
> If it has been a month since the last activity, a reminder is sent and the
> stale label is added
> If another two months passes, then the MR is finally closed.
> 
> While I agree that MRs awaiting review shouldn’t be closed, if it has been
> more than three months, including two months after a ping message that
> sends everyone involved an email and temporarily increased visibility in
> the MR list due to activity, I don’t think it will ever be reviewed; and if
> it is going to be reviewed, that can still happen.

Stop closing Merge Requests where the only issue of the MR is that we are not 
reviewing it, that is non negotiable.

Albert

> 
> > Why do you need a label, the stale MRs are all nicely collected at
> 
> https://invent.kde.org/teams/gardening/gitlab/-/issues
> Sure, there is a list of stale MRs there, but most people don’t know it
> exists and it is difficult to sort it per project or team. For example, I
> cannot search for all stale Dolphin bugs as a dolphin developer. It also
> makes it much more hidden from the native gitlab list of MRs. I am also
> hoping that seeing a big red “Gardening: Stale” label on an MR in the MR
> list should turn a few heads and cause it to gain some interest from
> developers.
> 
> > It's not that they don't have visibility
> 
> Correct, but I am trying to not lower the visibility too much when doing
> this, as MRs can always be reopened!
> 
> - Kye Potter, KDE Gardening
> 
> On Sat, 18 Feb 2023 at 6:45 am, Albert Astals Cid <aacid at kde.org> wrote:
> > El dijous, 16 de febrer de 2023, a les 21:55:41 (CET), AnnoyingRains va
> > 
> > escriure:
> > > Hey Luigi,
> > > 
> > > Thanks for the feedback! The stale MR closing has been a thing for a
> > 
> > while,
> > 
> > There's 2 different of stale MR:
> > 
> > 1) Some KDE reviewer has asked for something to be improved in the MR and
> > the
> > author of the MR did not follow up
> > 
> > 2) No one has reviewed the MR
> > 
> > 1 is possibly fine to close (after some prudential heads up time), the
> > author
> > lost interest, that's life.
> > 
> > 2 SHOULD NEVER EVER BE CLOSED. If you are doing that STOP NOW.
> > 
> > > just not particularly active, as I was waiting for my developer account
> > > application to be approved so I could properly label MRs (which does
> > > help
> > > with visibility!)
> > 
> > Why do you need a label, the stale MRs are all nicely collected at
> > https://invent.kde.org/teams/gardening/gitlab/-/issues
> > 
> > It's not that they don't have visibility, it's that we lack reviewer
> > power/
> > time.
> > 
> > Cheers,
> > 
> >   Albert
> >   
> > > While I know this isn’t really a solution, the gardening team also has
> > > “Love Projects”, which, at least historically, helps to renew interest
> > 
> > in a
> > 
> > > project. Possibly a Love Project could be opened for one of these
> > 
> > projects
> > 
> > > with many stale MRs, specifically mentioning them and the label that is
> > > applied with them.
> > > 
> > > I know you said that you would prefer if this is opt in, but this
> > > service
> > > is still opt out, like most KDE Gardening activities.
> > > 
> > > Either way, I do agree with your point, and will see if I can come up
> > 
> > with
> > 
> > > a solution.
> > > 
> > > Thanks
> > > - Kye Potter, KDE Gardening
> > > 
> > > On Fri, 17 Feb 2023 at 2:52 am, Luigi Toscano <luigi.toscano at tiscali.it>
> > > 
> > > wrote:
> > > > Hi,
> > > > 
> > > > I see the the gardening started also on stale reviews.
> > > > I think it should be opt-in, only for projects which request it.
> > > > If that's the case already, please skip the rest (or read it anyway,
> > > > it
> > > > may be
> > > > relevant if the policy changed).
> > > > 
> > > > We have projects which are, let's say, sleepy, were reviews are not
> > 
> > taken
> > 
> > > > into
> > > > account. You may say that, if some project is stale, it should go
> > > > along
> > > > its
> > > > fate and be archive.
> > > > I don't fully agree with that. As the KDE developer status allows
> > 
> > access
> > 
> > > > to
> > > > each and every repository, a project may get some renewed interest at
> > 
> > some
> > 
> > > > point, and we have had several examples of this in the past.
> > > > 
> > > > Now, I believe that automatically closing code contributions (which
> > > > are
> > > > slightly different beasts than bugs) can accelarate the process of
> > 
> > making
> > 
> > > > a
> > > > project look obsolete, rather than helping them to be back. Having the
> > > > reviews
> > > > around may be a way for new contributors as start with a certain
> > 
> > projects
> > 
> > > > (by
> > > > rechecking them and merging). It's true that the reviews are still
> > 
> > there,
> > 
> > > > just
> > > > closed, but it's not the same in terms of visibility.
> > > > 
> > > > Ciao
> > > > --
> > > > Luigi