Releases in 3 months

[laurent Montel]

Le lundi 8 juillet 2013 16:19:02 Àlex Fiestas a écrit :
> I did pasted the link in #kontact and #akonadi a couple of times before
> sending this email :p

I never saw it.
And paste on irc is not speak with guy which works several project.

> 
> The development time is NOT divided and it is NOT limited to 2 months,
> instead now master will be always open to include features and once every 3
> months we'll make a release.

So at a specific date you takes code without freeze before ?!


> For example, if you take a look at this pic:
> http://community.kde.org/images.community/b/b8/Drawing.png
> 
> For example between 4.12 branching (15/10) and 4.13 branching (15/01/2014)
> there will be 3 months. If for whatever reason your feature is not stable to
> go into 4.13, you can delay it for 4.14 which will happen 3 months after,
> so not much is changing of how we are doing things now.

So you will take code from what ?
I will continue to work on master so you will take a broken code ?

It’s more logical to freeze master until release otherwise you will release 
broken code it’s sure.

> 
> Said it in another way, if you want to keep a 6 months release schedule you
> can ! this just allow others to ship features more often.
> 
> Finally, I'm NOT proposing this because kde-workspace and kdelibs are
> frozen, that's just an opportunity to change the release cycle since it
> will affect less people. You can read the motivation here:
> 
> http://community.kde.org/KDE_Core/ReleasesProposal#Motivation

"The main motivation for this change is to reduce the amount of time between 
releases and make them simpler, making us able to deliver new features faster 
to our users while keeping if not improving the current quality. “

"deliver new features faster” ? :) new feature broken yes (less time more 
bugs)

Regards


> Cheers !
> 
> On Monday 08 July 2013 15:45:13 you wrote:
> > Hi,
> > As you know I work a lot on kdepim (but you forgot to ask me if it was a
> > good idea or I didn’t receive mail... :) )
> > 
> > Ok kdelibs and kde-workspace is frozen until KDE SC5 so less work for you
> > for example.
> > 
> > But for kdepim we are not a big team, and we have a lot of feature to
> > implement and it’s not possible in 2 months.
> > 
> > For example I finished my last feature 2 days ago (sorry I was a lot of
> > feature to do for 4.11)
> > 
> > So I disagree to reduce development time to 2 months just because kde-
> > workspace is frozen.
> > 
> > And as it’s frozen why reduce development time ?
> > Why increase number of release ? Just to increase number ?
> > 
> > So for my point of view it’s not a good idea to reduce time, we will speed
> > to create feature so more bugs because not big period of test, or we will
> > stop to develop feature because we don’t have time to finish.
> > 
> > Why is the reason to divide by 2 the release time ?
> > 
> > Regards

-- 
Laurent Montel | laurent.montel at kdab.com | KDE/Qt Senior Software Engineer
KDAB (France) S.A.S., a KDAB Group company
Tel. France +33 (0)4 90 84 08 53, Sweden (HQ) +46-563-540090