| 1 |
On Tue, Feb 03, 2004 at 02:58:38PM +0100 or thereabouts, Paul de Vrieze wrote: |
| 2 |
> I think that the whole update issue needs to be rethought. Will that be |
| 3 |
> done by fixed-tree maintainers (in cooperation with package |
| 4 |
> maintainers), or do package maintainers need to do it themselves. |
| 5 |
> |
| 6 |
> I think that in the end the updates need to go through a fixed-tree |
| 7 |
> maintainer in some way. |
| 8 |
|
| 9 |
The creation of each stable tree release will be done by fixed-tree |
| 10 |
maintainers. However, off-cycle updates (as currently proposed) will be |
| 11 |
done by the package maintainers themselves. |
| 12 |
|
| 13 |
Because this is NOT a QA GLEP, I don't see the need to funnel off-cycle |
| 14 |
updates through a central team. It's going to be different enough to |
| 15 |
commit something to the stable tree to prevent accidental oopses in 99.9% |
| 16 |
of the cases. If someone deliberately commits something off-cycle to the |
| 17 |
stable tree, then that's a separate issue. |
| 18 |
|
| 19 |
Down the road, if the QA team decides they want to put more stringent QA |
| 20 |
checks in place to ensure better QA of the stable tree ebuilds, this is |
| 21 |
certainly an option. That is, however, outside of the scope of this GLEP. |
| 22 |
|
| 23 |
--kurt |
Archives