Proposal: Better force-tag

[Mark McLoughlin]

On Sun, 2008-09-14 at 12:23 -0400, Brian Pepple wrote:
> On Sun, 2008-09-14 at 17:52 +0200, Denis Leroy wrote:
> > Ignacio Vazquez-Abrams wrote:
> > > Ultimately nothing can stop the user from using cvs tag -F, so any
> > > solution (including this one, which I like a lot) would have to be
> > > server-side.
> > 
> > That's what they're planning to do, despite many objections. When will 
> > the FeSCo minutes be available on the wiki ? I would like to know who 
> > voted for this.
> 
> I'm planning to write up the summary later today, but the IRC logs are
> available as soon as the meeting ends.
> 
> http://bpepple.fedorapeople.org/fesco/

jwb>      so we've had vocal feedback on this i think a simple vote is in order
dgilmore> your saying you want FESCo to control Makefile.common
jwb>      i'm saying FESCo was elected to represent our developer and
          user communities, and changes that impact them should at least
          be run past FESCo

Actually, may I respectfully suggest that FESCo should try and stay out
of matters like this insofar as possible?

IMHO, FESCo should merely be trying to reflect the "rough consensus" of
the rest of the community.

e.g. in this example, a rough consensus is forming on the list along the
lines of:

  1) "force-tag" is used by a number of packagers for reasonably valid 
     reasons that don't conflict with the tracking of what sources were 
     used for a given build

  2) We could disallow (on the server-side) re-tagging if a build has 
     completed, or is in progress, using that tag

  3) Or we could track the CVS/Entries for a given build

FESCo need not play any role here beyond keeping an eye on the progress
of this rough consensus - i.e. that the original change is being
reverted and the relevant maintainers (and others) are hashing out the
implementation of either suggestion. FESCo taking a position via a vote
is not helpful, IMHO.

Cheers,
Mark.