Re: [gentoo-dev] rfc: revisiting our stabilization policy

[Jeff Horelick <jdhore@gentoo.org>]

[-- Attachment #1: Type: text/plain, Size: 2845 bytes --]

On 14 January 2014 18:11, William Hubbs <williamh@gentoo.org> wrote:

> On Tue, Jan 14, 2014 at 05:43:57PM -0500, Michael Orlitzky wrote:
> > On 01/14/2014 05:33 PM, William Hubbs wrote:
> > > On Tue, Jan 14, 2014 at 04:57:30PM -0500, Michael Orlitzky wrote:
> > >> On 01/14/2014 04:37 PM, William Hubbs wrote:
> > >>>
> > >>> 2. I would like to see the policy below applied to all arch's [2].
> > >>
> > >> [ ] Yup
> > >> [X] Nope
> > >
> > > The reverse of this would be to let maintainers stabilize on all arch's
> > > after 90 days, then they are allowed to remove all but the latest
> stable
> > > version. This isn't good though because maintainers would be
> stabilizing
> > > packages on arch's where they can't test.
> > >
> > > The stable tree is significantly behind because the arch teams are so
> > > short staffed, and this prooposal is an attempt to fix that.
> >
> > It's attempting to fix a headache with a bullet. The arch teams are
> > lagging behind, you're annoyed, I get it. Give 'em hell. But don't break
> > stable to make a point.
> >
> > For users, both options are worse than the status quo.
>
> The first option would start reverting things back to ~ and users would
> have to unmask them.
>
> The second option would introduce new things to stable which may not be
> stable due to not being tested on the arch.
>
> The second option is worse than the first imo, that's why I didn't
> propose it first.
>
> The status quo is not good, because we are forced to keep old, and
> potentially buggy, versions of software around longer than necessary.
>
> William
>
>


I think the simplest short-term solution might be to add teams that are
looking for ArchTesters to the Staffing Needs page on the wiki and promote
that page like crazy. I'd be more likely to do a lot more stabilizations if
it wasn't just me going on my experience and running through the AT
procedures myself (they're also a bit lengthy if you follow them properly,
which I prefer to).

I do feel some arches should be a bit deprecated. Not quite as severely
other arches the council deprecated a few months back, but something.

Also, to ease the burden on Arches, it'd be nice if the maintainer would do
some of the archtesting work on all their available arches rather than
making the AT's/arch teams do it...For example, almost everyone who has a
amd64 system, can easily make a x86 chroot (or VM) to test in.

Another option (and I don't mean to step on any toes or call anyone out
here, these are just examples) may be to just deprecate stabilizing certain
software. Packages such as the stuff in app-vim/ or app-emacs/ or some
games or some scientific software. For the editor plugins, most people do
not get them from the package manager I feel and a Vim plugin requires
almost as much arch testing work as a new version of grep, for example...

[-- Attachment #2: Type: text/html, Size: 3719 bytes --]