[bt-devel] proposal

[Martin Gruner]

On Sam, 16 Sep 2000, you wrote:
> Hi,
>
> > > Hehe, if we'd do it, who would spend $1500 each year to pay the license
> > > fees to trolltech?
> >
> > Maybe. Why not? Maybe we can beg them for a free version since we are not
> > developing a commercial program, but I don't really think so.
> > Maybe we make a call on the website that people should send money if they
> > want a windows version?
>
> I think very few people would react and sen dmoney. We won't get enough
> IMHO.
> > > Basically it's possible, but we made BibleTime as a KDE and as a Linux
> > > application from the first day because in Linux there was no Bible
> > > program for Linux.
> > > For windows there are so many bible study tools so we have probably no
> > > chance to get some marke tshare.
> >
> > You may be right, but I see no GPL alternative to bibletime... bible
> > study tools are expensive...
> > IMO the poor should profit from bt, hopefully even more with windows...
>
> Don't forget: I do not have windows and I''m not so rich to spend 700DM for
> a buggy OS if I can get a better one for 80DM. I coded for Windows in
> Delphi years ago, and Qt is much better.

This is not the problem, I would send you a 95 license.

>
> > > If we'd switch toQt we have to change these things:
> > > 	-Drop XML stuff
> >
> > I don't mind.
>
> But it's really useful and easy to use.

After the keychooser is finished, we would have to drop the xml for the 
presenters anyway. (Not sure)

>
> > > 	-Install dialog will be very hard because we'd have to write our own
> > > network management code
> >
> > Let us make a separate application of this. Wouldn't be so bad if it were
> > linux-only. Maybe we can port it later...
> >
> > > 	-Integration in KDE is lost
> >
> > Up to this point it does not hurt. We don't use dcop and kparts yet, and
> > there would be other ways to do the things that need to be done...
> >
> > > 	-Lots of small things will go (i18n(), SmallIocn(), LargeIcon(), ... ,
> > > klistview, the KPopupMenu, )
> >
> > i18n() would be a seroius problem. Is it not available on windows? Is it
> > not a GNU thing?
>
> Qt does use tr(), but I'm so familair with KDE.

Ok, accepted.

>
> > The icons could be compiled into the binary.
>
> Nooo! The binary would be megs to huge if we do this! If we'd do it we have
> to port some KDE code.

We could write a simple backend for images.

>
> > Klistview -> Qlistview, IMO easy.
> > same for QPopupMenu
> >
> > > 	-We have to change all the dialogs (almost all use KDialogBase and KDE
> > > widgets)
> >
> > Takes time, but is possible and not too hard. The optionsdialog might be
> > difficult...
>
> Maybe port some KDE sources?

Maybe. Maybe just use QT (designer?) for a static layout...

>
> > > 	-Port KDE partly, at least the widgets we need to Qt and Windows
> >
> > Which? KTextEdit?
> > Why not port KConfig?
>
> Sure, we'd have to port KConfig, don't know about KTextEdit.

I guess we couldn't port KTextEdit even if we wanted to...

>
> > > Somebody would have to buy Qt and a compiler ...
> >
> > Exactly. I think Inprise offers a free compiler, and we don't need an
> > IDE.
>
> We could use KDevelop and compile in windows or vmware.

VMware? Isn't it expensive? Win is ok.

>
> > > Is it really worth it?
> >
> > I am not sure, and I don't want to question all that you have
> > accomplished so far. but I see that a windows version of bt would be an
> > amazing thing...
> >
> > IMO we should aim towards it in the long run. That means we could do this
> > slowly by not introducing new kde stuff, trying to clean things up and /
> > or simplify them...
>
> I don't know. I like Linux, dislike Windows, like Qt and KDE, dislike the
> Windows UI and I don't want to do again all the work!
>
> BTW, KAction things will be lost if we are Qt only!

What about QAction? Btw, IMO QT will use more and more things from KDE in 
future.

Martin
>
> > But those are just suggestions.
> >
> > Martin
>
> --Joachim