ktouch/next ready for merge

[Sebastian Gottfried]

Hi Laszlo,

> > Not to mention, KDE 4.10 would be the carry of the previous version
> > for Windows. People would come to report issues and so forth since
> > that is the official Windows version then for KDE 4.10. If all the
> > responsibilities are put towards the distributors, they can achieve
> > the same in the end, but people will still come to report issues about
> > KDE 4.10 KDE Edu, KTouch because that would be KTouch 4.10 for
> > Windows.
It's  KTouch Version v1.7.1 the reports will be reported against. And most of 
them I'm going to be able to close as fixed in v2.0. But I don't expect a 
flood of the new bugs against the old version in the light of three new bugs 
in the year 2012 at all for KTouch.

> 
> Another thing to consider is that, at this point we do not know which
> KTouch version will be better in the future for cross-platform. You
> may disappear at some point for a valid reason, and someone picks up
> the 4.9 version, and extend that, then that will be the official
> version, I guess.
Very true. Still I now which version is better in general. By far. Someone may 
use that criteria for determining what should be the official version, I hope.

That anyone would continue on working on the old version I think is rather 
unlikely. I didn't happen in two years before I did show up and not in the six 
months I've developed on the new version in public. Why should it happen now, 
with the new version there?  And of now I'm still around and I'm at least not 
planning to disappear.  Porting it to Windows isn't impossible, just time 
doesn't permit to do so for KDE 4.10.

I've think the discussion should factor more arguments than availability on 
all platforms. For example all the wishes and bugs for KTouch in Bugzilla this 
version would fix when merged. I haven't closed any issues yet, because of the 
uncertainty when my version will be released, but there a lot them.

> 
> Making the 4.9.X-{windows,linux} branching may be a bit duplicated as
> well and would be different than for the rest. Fwiw, khangman linux
> and harmattan are also different, and do not share much. They could be
> different branches, but it was simpler to distribute, and it did not
> become a harm either. As for Harmattan, I was not even lucky to detect
> the platform as it is possible for Windows with cmake.

There is one big difference: all your distributions are maintained and up to 
date. Basically they are different implementations of the same application for 
various platforms. And they have common grounds: the shared engine/ folder
and the themes data. Those shared artifacts make it really advisable too put 
everything under in one tree from a maintenance point of view.

On the contrary for KTouch this would mean that I would create new top-level 
folder "old" or "non-X11" with the complete KDE 4.9 KTouch tree under it and 
and some CMake magic to use that on non X11 platforms. I'm not totally against 
it but I still not totally convinced of the benefits. It would make the life 
of the KDE Windows installer team a bit easier, but that's all I can see for 
now.  The contents of that folder would be as still unmaintained as before the 
branch. At least as long no one steps up and maintains it what I don't see 
happening. I'm really more than busy with making the new version as good as 
possible. And the end user still would get the old version on Windows, 
obviously.

Best regards,

Sebastian