[Spice-devel] Proposal: review branches (was Re: [vdagent-win PATCH v6 2/5] Initial rewrite of image conversion code)

[Christophe Fergeau]

On Tue, Jul 25, 2017 at 02:23:28PM -0400, Frediano Ziglio wrote:
> > 
> > On Tue, Jul 25, 2017 at 07:20:23PM +0200, Christophe Fergeau wrote:
> > > On Tue, Jul 25, 2017 at 07:09:22AM -0400, Frediano Ziglio wrote:
> > > > > 
> > > > > > >>> I see several benefits to doing this:
> > > > > > >>> 
> > > > > > >>> 1. We always know exactly which component and branch is being
> > > > > > >>> patched
> > > > > > >>> 
> > > > > > > 
> > > > > > > As long as contributor keep pinging or resending his series, this
> > > > > > > is
> > > > > > > already the case.
> > > > > > 
> > > > > > As Frediano said at the beginning of the series, “I’m tired of
> > > > > > hearing this
> > > > > > reply”.
> > > > > 
> > > > > And this is not an actionable answer... My perception is that there
> > > > > rarely are 'ping' on old series. Does this mean we are doing a good job
> > > > > at reviews? (I doubt it or we would not have this conversation) Does
> > > > > this mean patch senders do not want to do that? Why? Does this mean
> > > > > it's
> > > > > done a lot, but to no avail? All I'm reading is "I'm not happy with how
> > > > > things work", with nothing specific.
> > > > > 
> > > > 
> > > > Patch series are getting old (even years) repeatedly pinged (5/6 times)
> > > > but they continue to not getting any feedback/ack/comment.
> > > > If you can't remember any... this just confirms the problem.
> > > 
> > > So I went through my mails (searched for mails containing 'ping'), in
> > > the last months
> > 
> > Forgot a number here, it's since April 2016, so in the last 15 months or so.
> > 
> > > I found 24 series which needed a ping, among these, 3
> > > needed several pings, and it stopped at 2 pings. Maybe I missed some.
> > 
> 
> As I said with ping you don't count multiple submissions and even
> if you got a 0 ping count this does not mean that series get not
> reviewed and should be. For me the fact that you have to search is
> another confirmation that a tool would be helpful as memory and
> patchwork are not enough.

I guess we are going to disagree, but to me the fact that I forget is
not far from being a feature, and some kind of self regulation more than
being a big problem. I only have a limited review bandwidth, so
forgetting some patch series is (from that perspective) a good thing,
and relying on patch submitters to ping on still important stuff that is
lagging behind moves the burden of remembering to the person most
involved with the patch series.
You can see it as a failure on our part (and for non-core contributors,
I agree it really is, we should be more responsive), but it can also
have its benefits.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/spice-devel/attachments/20170726/349683ec/attachment.sig>