QA 'vote' at the go/no-go meetings

[Adam Williamson]

On Wed, 2011-05-18 at 16:00 -0400, James Laska wrote:
> On Wed, 2011-05-18 at 02:09 +0000, "Jóhann B. Guðmundsson" wrote:
> > On 05/18/2011 02:03 AM, Jon Stanley wrote:
> > > On Tue, May 17, 2011 at 9:16 PM, Adam Williamson<awilliam at redhat.com>  wrote:
> > >
> > >> I hope that's acceptable to all! If not, or anyone has ideas for
> > >> improvement, do say so...
> > > As a formerly very active QA guy, and less so now since I got a new
> > > $DAYJOB, I think that this makes sense, bit I do have minor concerns
> > > over what happens if someone goes and clears the blocker list just
> > > before the meeting for the sake of being able to say that there are
> > > none. I hope that this concern is mostly academic, though :)
> > 
> > Well let's just hope for our own sake that we will notice if someone 
> > suddenly goes and clears out the blocker list out of nowhere..
> 
> Yeah that's lame ... I think the likelihood of someone sabotaging the
> process is low, and we thankfully have plenty of smart folks involved
> monitoring the bugs should anything happen.

Note that this process doesn't preclude adjusting the blocker list
during the go/no-go meeting (in fact, it actively suggests it in a
similar case). What I expect would happen if someone maliciously cleared
the blocker list is that we would simply restore it before casting the
vote. This should be non-controversial, as the conditions for changing
blocker status are clear and specified in the blocker process SOP, so
there should be no problem identifying a malicious change.

I don't think it's worth codifying this, there's always a point at which
it becomes futile to try and write down every little thing and you just
have to go with common sense.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net