Summary: Syncing from testing a success?

[Martin Pitt]

Hello all,

thank you for all your contributions and comments. I tried to sort and
categorize the answers, and want to give a summary:

Pro:
- general perception is that it increased stability

- syncing from testing was beneficial for stability of the "unowned"
  packages in Ubuntu

- large library transitions are tested and reasonably RC bug free,
  instead of half of them hitting us and DIF getting in between

Neutral:
- not much stability difference for the actively maintained packages

- no difference in difficulty of merging packages, just a slight hindrance
  (like X.org)

Con:
- Debian encountered RC bugs which aren't relevant to Ubuntu and held
  up testing propagation
  → compromise: autosync from unstable until DIF, then autosync from
  testing for a bit longer

- required slightly more manual syncs/merges for the actively
  maintained packages

Tool support:
- Developers need to be able to do syncs themselves to eliminate the
  Archive Admin overhead/bottleneck

- Need possibility to merge with both unstable and testing (mentioned by 5
  people) → bzr branches support that, but not MoM

- tangent: we should review removals from testing and check if we
  should remove them from Ubuntu

General:
- no general agreement that we should continue to do so, or whether it
  will continue to work well in future cycles

- in the next one or two cycles Debian will be frozen, thus we should
  go back to syncing from unstable

Thanks,

Martin
-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
Url : https://lists.ubuntu.com/archives/ubuntu-devel/attachments/20100420/c2a321eb/attachment.pgp