Re: notification process [was Re: gnome-terminal awfully slow]
- From: Luis Villa <louie ximian com>
- To: desktop-devel-list gnome org
- Subject: Re: notification process [was Re: gnome-terminal awfully slow]
- Date: 13 Feb 2003 15:30:54 -0500
On Thu, 2003-02-13 at 13:37, Bill Haneman wrote:
> >
> > Bill, if Havoc had send an eMail with "Maintainer decision: switch zvt->vte" as
> > subject to the d-d-l, would have been enought for you?
That really shouldn't have to happen. Everyone in GNOME knew this was
coming; it had been announced over and over again. It seems unreasonable
that the QA team, the doc team, and the UI team were all aware of
platform changes and the a11y team needs special notification.
This goes for _all_ teams:
We don't have a centralized model where we have infinite amounts of
resources to test things and let people know what happens. If change
impacts you, responsibility /must/ fall on your shoulders to test,
notice, and scream. The best that the release team can do is to announce
things repeatedly (which we did), schedule in time for people to find
problems (which we did), and react when people raise the problems during
that period (which we did.) We can't special-case groups of people or
features, because otherwise the entire release process will block on
features no one cares enough about to test or use.
If there was some flaw in the notification process more useful than 'we
didn't list at the code level every single change between 2.0 and 2.2'
then I'd think we'd be happy to consider it. But 'we're replacing vte
with zvt' should be detailed enough- if you care for zvt, at that point
it falls on your shoulders to test things and scream about them. The
same goes for everything else in the platform.
Luis
> Yeah, if it had noted that the change was a compile-time change and not
> a runtime option.
> I think all dependency changes and similar should have such a "we've
> just taken a fork in the road" notification (in addition to the run-up
> discussion).
>
> > In this way we all would see a non buried mail that just states the decision.
> > This could be something very good. But only if people respect such a
> > decision ;-)
>
> Yeah, the current thread was intended to be about notification and
> process, not rehashing the decision itself.
>
> > We can't expect the maintainers to repeat themselves over and over
> > again.
> >
> > Martin H.
> ...
>
> But sri says:
>
> > Ack, too much paperwork. The proper way is to create a bug in bugzilla
> > that says "fix vte accessability" and put that as a showstopper or
> > something. Lets not create more paperwork than we have to. We have a way
> > to track things and we should use that.
>
> This doesn't work since we don't have the explicit concept of
> 'showstoppers' in GNOME; and besides, probably most people would not
> accept accessibility issues as stoppers at this time. I think that's a
> shame, and I hope it changes once we've established an a11y user base.
>
> There were bugs filed (and closed) against vte, the problem was that we
> couldn't fully confirm the fixes and the closures were therefore
> premature. That said, we appreciate the fact that Nalin was listening
> to our concerns, thanks Nalin.
>
> -Bill
>
> > sri
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]