[Kst] branches/work/kst/kst1kde4/kst/src/libkstmath (silent)

[Brisset, Nicolas]

> Hi Nicolas,
> 
> My primary responsibility is to the LFI Planck team rather than to Kst
per
> se.
OK, fair enough. Over the years I have grown to appreciate your
efficiency at fixing bugs and improving the tool, I understand that
having a fast solution to the problems is great and that you are doing
this well. But I suspect they still have interest in the long run in
having as many people as possible working on improving kst. Creating a
separate branch would mean they have to maintain it themselves... unless
it turns out that everybody only wants to use that branch and that the
portto4 one dies. But frankly, I don't believe in this scenario. 
If you've followed what happened in KDE land during the (long)
transition to Qt4/KDE4, everybody said that porting (a real port, making
use of all the great new features and not relying on qt3support) was
much more work than they had expected. Qt4 is a major overhaul, and that
has a cost. I think the people who worked on the portto4 branch are
excellent coders, and they knew what they were doing all along the way.
Maybe everything is not perfect (memory handling?), but I think it is
better to point out the problems and try to discuss them than to work in
a separate branch. 
 
> The discussion over moving to a KDE-free or Qt4 based version of Kst
> started
> many years ago,
> and the work at Toronto has been going on for a long time.
That's right. I also can't wait to see 2.0.0 out, and hopefully a flurry
of new users and developers. But at the same time I understand that this
takes time, especially as in the last months there was only one person
full-time working on that port, and more recently only Barth in his
limited commuting time. That's exactly why I'd rather see you lend a
hand here than start a new branch. Even if you don't want to participate
in the overall design discussions, there are a lot of well-isolated
areas (datasources for one) where some improvements could be done. But
then, that's only my opinion, I'm not the one who's paying you to work
on kst so I don't have a say in that decision :-)

> It has recently become clear to me that the LFI team, at some time in
the
> near future,
> will want to move away from KDE3 - particularly given the ever
decreasing
> level of support for KDE3 in recent linux distributions. In order to
> support
> that there needs to be a stable and well-tested version of Kst that
does
> not
> rely on KDE3.
Kst 1.x only depends on kdelibs, and I believe these will stay around
for another few releases (i.e. years) even though kde3 applications will
fade away. And nothing prevents you from using kde3 apps in a kde4
environment for as long as you need. I don't see such a high pressure.

> That might well prove to be the Toronto version, but I believe that
there
> has to be a plan B.
You may be right. But I have started working with kst2 lately and even
though there are still quite a few quirks to iron out, it is definitely
usable. And I think under-the-hood improvements have yet to show their
full potential. For me it is definitely too early to make the drastic
decision of forking kst2, which is basically what you are starting to
do... 

Sorry for the long post, but I think this is important. I don't want to
criticize you, and I really appreciate your work. To be constructive,
could we maybe start to discuss concretely the problems you see with the
current portto4 branch and try to solve them before you work on your own
version? 
Everybody would benefit.

Best regards,

Nicolas