On Tue, 30 Jul 2013, Cyrille Berger wrote:
On 2013-07-30 13:07, C. Boemann wrote:
Now as to which branch to use etc. I clearly recall the qt4/kde4 porting -
it
dragged on because things were not ready, but certainly also because we
were
still working on the qt3/kde3 version at the same time.
That is quiet wrong. We only had six months while we had both qt3/kde3 and
qt4/kde4 in different branches. Then we had two years before the first KDE4.0
release, where before being able to code on koffice code, you would have to
update your version of kdelibs4, kdebase4 and some other stuff for
kdesupport, if you were lucky, you would have time to fix what broke in
koffice in conscequences of changes in dependencies. And then it was time to
go to bed before you could work on actual improvement. It is only after
kdelibs 4.0 release, that things started to improve drastically.
Yes... In fact, we were porting for nearly three years, at least that was
my gut feeling. I do want to avoid that very much, and while there are
some very good developments in frameworks (Sebas' plugin work!), I think
we need a more controlled approach here. But also one where we avoid:
* a fork, which could very well happen if a subset of calligra got
ported in a branch. I'm not using the word fork here in a hostile sense,
but what I mean is that two branches grow apart too much to merge.
* that the porting needs to be done twice (once now, and once when KF5 gets
released next year).
* where all the porting burden falls on the shoulders of the funded people
-- because that will cause serious instability in the rest of Calligra,
simply because there is so much code...
Btw, the krita-mvc-rempt branch is progressing again. KoPart no longer
needs to be an xmlgui client. I'm taking my time here, but there is some
progress towards proper components.
Boudewijn
_______________________________________________
calligra-devel mailing list
calligra-devel@kde.org
https://mail.kde.org/mailman/listinfo/calligra-devel
