[ANN] What with all these releases ?

[John Arbash Meinel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 9/20/2010 9:32 AM, Max Bowsher wrote:
> On 20/09/10 14:59, John Arbash Meinel wrote:
>> On 9/19/2010 8:58 PM, Martin Pool wrote:
>>> On 19 September 2010 09:31, Max Bowsher <maxb at f2s.com> wrote:
>>>> All I meant is that the SRU process involves extra effort for us, and
>>>> the Ubuntu SRU team, and this effort is mostly separate for each
>>>> separate distroseries - so we need to decide for which ones it is worth
>>>> the effort. However, a lot of that effort disappears if we get a
>>>> MicroReleaseException.
>>
>>> ... because they don't need so much individual verification of the changes?
>>
>>
>> Reading the MRE page, it definitely seems more about packages that are
>> "stable but don't fit the SRU guidelines". I saw that Martin Pitt
>> pointed out the same thing.
>>
>> The big thing is figuring out why there is so much friction for us. From
>> the time we cut a new .x release, to how we can get it into -updates.
> 
> AIUI, The friction exists because the Ubuntu SRU process is designed to
> facilitate extremely cautious critical updates. Bazaar micro releases
> contain more than tightly-targeted fixes for critical updates - i.e.
> they contain more general bugfixes.
> 
> The MicroReleaseException process is designed to alleviate this for
> project with sufficient QA.

I understand that, but both Mark and Martin seemed to think that bzr's
changes were close enough to the SRU categorization, and thus wouldn't
actually need the MRE.

Anyway, whatever it takes to get less friction from releasing an update
to getting it into users hands is a good thing. This doesn't seem like
MRE is the whole story, but if it works, great.

John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkyXdlsACgkQJdeBCYSNAAMBOACfUMfHqoB6I6WGlC8xUSY0QzZB
75cAn3XA9ohcpZ9Qa3HMVSgM9nZctdKc
=zvtH
-----END PGP SIGNATURE-----