Fedora 18 Beta to slip by two weeks, Beta release date is now Nov 27

[Adam Williamson]

On Thu, 2012-11-08 at 16:39 -0700, Tim Flink wrote:
> On Thu, 8 Nov 2012 23:00:42 +0000
> Matthew Garrett <mjg59 at srcf.ucam.org> wrote:
> 
> > On Thu, Nov 08, 2012 at 03:48:31PM -0700, Tim Flink wrote:
> > > On Thu, 8 Nov 2012 20:14:05 +0000
> > > Matthew Garrett <mjg59 at srcf.ucam.org> wrote:
> > > > "we" are? I see approximately nobody offering assistance in that 
> > > > respect.
> > > 
> > > If it would make you feel better, I can stop building test images,
> > > updating test repos, testing fedup, writing documentation, answering
> > > questions in bugzilla as best I can and attempting to help
> > > coordinate. I'm obviously not helping in the least and it sounds
> > > like my time would be better spent elsewhere. It'd certainly take
> > > things off my plate and then you can do the work.
> > 
> > Oh man I'm sorry - I meant other than you and Will, and I'm hugely
> > sorry that it sounded like I didn't think you were being useful here.
> > I really really don't want to make it seem like I don't appreciate
> > all of that. But we do have many people who are nominally active in
> > writing code who are doing absolutely nothing here.
> 
> That may be true but that isn't true of everyone. I understand your
> frustration, I just think it was directed at the wrong person who
> happened to say "we". Adam isn't doing much with fedup at the moment but
> I'm not doing much anaconda testing at the moment, either and it's not
> a coincidence.

Sorry for introducing the confusion. When I said 'we' I was thinking of
'anaconda team and QA team' - as Tim pointed out, he's been working more
or less exclusively on testing, debugging, documenting, co-ordinating
and generally helping out with fedup for the last week or so. I and the
rest of the QA team have been working on anaconda testing and keeping
the test day cycle going. Division of labour :)

Aside from that - I can understand your frustration that you think
people are chinwagging and not helping, but my point is kind of that you
(anaconda team) have brought that on yourselves. You regret the lack of
contribution, and dcantrell talks about problems with communication: one
of the main points of the feature process is precisely to act as a hub
for communication and contribution. If fedup had gone through the
feature process, people would have known about it, and if the feature
description had been done correctly, people would have known far earlier
that we needed to figure out stuff like packaging for F16 and F17,
generation and shipping of the initramfs, and so on.

We wouldn't have wound up in the situation we're in now where Will is
tied to his desk frantically coding the thing and Tim - who has no
official standing - is trying to prod other groups into action to deal
with things like documentation and testing and providing the appropriate
infrastructure for building and shipping various bits of the framework,
while simultaneously trying to make sure he actually knows what needs
doing and is prodding the right people in the right direction (which
involves bugging Will, who would rather just be coding). If fedup had
been run through the feature process, FESCo - which has a lot more
prominence and weight - would have done that weeks or months ago, and
all the groups ought to have been on board and known what they needed to
do. Everyone would still have had their own idea on what color the
initramfs bikeshed should be, but they would also have known what color
FESCo had decided it was going to be, and they would be working together
to paint it that color, even if they were grumbling the whole time that
it was a stupid color and their idea for the color was way better.
No-one would be able to play the 'well we have no idea what the hell is
going on, no-one told us' card. That's how Fedora works when it's
working right; we all still think our plan is the best one, but we work
on the plan that got agreed on anyway. If we don't use the position and
oomph of a group like FESCo to agree on a plan ahead of time (however
fractious the process of agreement is), we wind up in this mess; the
buck-passing and chinwagging is kind of just an inevitable outcome when
the decisions aren't made sufficiently clearly and sufficiently far
ahead of time.

> I'm of the opinion that more widespread testing of fedup isn't really
> going to help right now until a couple of bugs are fixed. I'm not
> blaming anyone; we are where we are and the blame game is counter
> productive right now. I just don't see the point in creating more noise
> than we need to at the moment.
> 
> When it's ready, I'm certain that plenty of QA (and not-always-qa) folks
> will be testing fedup.
> 
> > > Asserting that we aren't helping in any way is incredibly insulting
> > > and in my opinion, rather naive. I realize that I don't write code
> > > that gets shipped with Fedora and I realize that through some eyes,
> > > that's what happens when you have no talent or useful skills. If
> > > that's what you want to think, go for it; you wouldn't be the first
> > > person to think so and I sincerely doubt that you'd be the last.
> > 
> > Again, my wholehearted apologies. That wasn't my intention.
> 
> No worries. We're all in this boat together, wherever it ends up. I
> think that I'm with most people here in hoping that the boat ends up
> at a good release without more slipping :)

Amen!
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
http://www.happyassassin.net