Re: [gentoo-dev] Gentoo-sources - should we stable?

[Rich Freeman]

On Fri, Jan 2, 2015 at 3:11 PM, Ian Stakenvicius <axs@g.o> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> On 02/01/15 02:57 PM, Mike Pagano wrote:
>>
>> I understand your point. Maybe waiting a few days to auto stable
>> makes sense, because less than 7 days later, a new version with
>> bug/security fixes is released.
>>
>> Isn't our current rate of stabilization "selling" a promise of
>> stability we can't stand behind?
>>
>> Mike
>>
>
> Well to be perfectly honest, the current-stable 3.16 and 3.17 kernels
> for me at least have some rather unfortunate regressions over 3.15 and
> previous, so even with the stabilization we're achieving now I don't
> think we're living up to our "promise of stability" :)
>

As a btrfs user I went through quite a bit of pain in the whole
3.15-17 series, but I that probably isn't a typical mainstream user
experience.  This sort of thing was why I did suggest targeting
longterm.  When the next longterm is announced then we could
transition to it at our leisure (ie with plenty of testing while
following point releases quickly on the previous longterm).

So, moving between longterm branches would have a more typical Gentoo
QA process.  However, between point releases within a branch we would
auto-stable releases, since it is unlikely that our own QA process is
going to add any real value beyond what upstream already does.

We also need to keep in mind just what our "promise of stability" even
means.  We're not a release-based distro, and we're NEVER going to
offer an experience like RHEL or Debian Stable where the entirety of
the package base is pinned and tested and we only do security
backports.  The kernel stable branches probably represent a lot more
stability than we have almost anywhere else in the distro anyway.

-- 
Rich