Re: GNOME & KOM/OP

[Stefan Westerfeld <stefan space twc de>]

   Hi!

On Thu, Aug 06, 1998 at 03:18:30PM -0500, Miguel de Icaza wrote:
> > Another way would be - if there are too wide gaps, to restart from scratch,
> > while ensuring that reasonable bindings to Qt & KDE exist from the
> > beginning. Then there should be active cooperation, to ensure, that
> > the KOffice developers will actually switch to the GNU Object Model
> > as soon as possible.
> 
> Our Component model will be based on CORBA, which means you will be
> able to use it from practically any language you want (which is one of
> the goals of GNOME).  The component model does not require Gtk or any
> other graphical toolkit. 
>
> The libraries for the Component model will be released under the terms
> of the GNU LGPL to allow linking with more restrictive licenses (which
> includes the Qt license) so it will be possible to have proprietary
> components talking to free components. 

Being LGPL'd and toolkit independant makes it possible for KDE developers
to use your Component model. But just because it is possible, it doesn't
mean that it will be done.

> I talked to Torben at the Kongress about this and he did not seem very
> interested in redoing OpenParts/KOM with something different, as it is
> part of his master's thesis.

I asked him about it:

  - KOM exists/works since quite some time now
  - KOM does not depend on the toolkit either
  - KOM works on top of CORBA as well
  - there is lots of code modelled around KOM
  - using KOM/OP in KDE applications is very simple (good supported),
    you just need to create an OPControlFrame, and you can embed OP-
	Controls easily (same with OpenParts, Document/View model,...)

KOM has some nice features like a signals & slots implementation as well.

He doesn't want do kill KOM due to these reasons. But since every KOffice
extension will require you to write KOM code, and since a lot of KDE
developers will do such or similar stuff (for instance working with
OpenParts), they will use/learn KOM and not yet-another-object-model,
I suppose.

Would it be _that_ difficult for you to make your glorious GNU object
model (which is done the-right-way(tm)), work as drop-in-replacement
for KOM?? You would help many people, really!

But then, if you go that way, _talk_ to them, and ensure that you don't
develop something that is well designed, but useless, since shaped
without looking at the programmers needs.

   Cu... Stefan
-- 
  -* Stefan Westerfeld, stefan@space.twc.de (PGP!), Freiburg/Germany
     KDE Developer, project infos at http://space.twc.de/~stefan/kde *-