Release Strategy Proposal

[Martin Graesslin]

On Sunday 28 April 2013 18:46:57 Àlex Fiestas wrote:
> > what i'm not OK with is delaying the entire PW2 effort for powerdevil
> > (e.g.)
> what I'm not Ok with is waiting until PW2 is ready to do big changes in
> powerdevil. I also don't want to have pressue to put stuff in 4.11 just
> beacuse there won't be a 4.12
we can turn that around: I'm not OK with not being able to port KWin to Qt 5 
because powerdevil wants to have big changes in 4.12.

It sucks we just have two different aims which are conflicting.

But I must say that I don't understand your arguments. I already didn't 
understand them last week when we discussed the ideas in Nuremberg and I still 
don't understand them. I see many possible solutions to the problem of 
powerdevil releases which do not conflict the freeze:
* mark as bug fix (I'm against that due to the regression potential)
* split out powerdevil from kde-workspace repository and make independent 
releases
* branch a KDE/4.12 branch from KDE/4.11 and release that as 4.12 with the 
rest of KDE SC (would obviously not part of the supposed long term 
maintenance)
* wait for PW2

Overall it's also the question whether the need to have a new powerdevil is 
high enough to question the plan of the other projects in workspace. Please be 
reasonable here and accept that it's the best for the project to do the freeze 
even if it would mean that powerdevil has to wait till it gets to the users. 
And please don't rush anything into 4.11. This would be exactly the opposite 
of what we would want to provide by a long term maintenance release (I'm a 
little bit scared since my system stopped suspending due to a recent change 
;-). The code should be pretty stable and without regressions. Something the 
enterprise distributions can safely use as a base for their next five+ years 
distribution.

Cheers
Martin