Why are 4.8.80 packages already out in the wild?

[Albert Astals Cid]

El Dilluns, 4 de juny de 2012, a les 06:31:56, Martin Gräßlin va escriure:
> On Monday 04 June 2012 01:26:01 Kevin Kofler wrote:
> > On Monday 04 June 2012, Albert Astals Cid wrote:
> > > It's not about hurting or not (that it does, see how Martin got confused
> > > and  probably lost time wondering how something like that could be
> > > happening).
> > 
> > Oh, and what Martin actually got confused by is a failure of processes on
> > the KDE end: The Bugzilla entry for the new (pre)release should be created
> > no later than the first "try" of tarballs. How else are we, the packagers,
> > supposed to report bugs we find in the tarballs, if they're not of
> > release-
> > blocking importance? When you create the tarballs, the component should
> > already be in place.
> 
> No, sorry, that doesn't make any sense. Why?
> * the number of bugs reported by distro people compared to users is
> extremely low (less than 1 %)
> * I cannot remember a bug reported by distro people between tagging and
> releasing
> * for me it is important to be able to reference a git hash with a release.
> The git tags are only pushed *after* the release. In the state between
> tagging and release I am normally not aware what has been tagged, with the
> exception of 4.8.80 thanks to Albert publishing the shasums on the mailing
> list

I'll try to make this "feature" (publishing temporary hash<->tarball values) 
permanent in the releases i handle.

Cheers,
  Albert

> 
> So no, also in future I will create the version tag in bugzilla once there
> is a release and not when the release process got started.
> 
> Cheers
> Martin