RE: Proposed: Evolution (and associated libs)
- From: Jon Kåre Hellan <hellan acm org>
- To: desktop-devel-list gnome org
- Cc: Chipzz ULYSSIS Org
- Subject: RE: Proposed: Evolution (and associated libs)
- Date: Wed, 07 Jan 2004 05:15:20 +0100
On Wed, 2004-01-07 at 04:15, Chipzz wrote:
> On Tue, 6 Jan 2004, JP Rosevear wrote:
>
> > From: JP Rosevear <jpr ximian com>
> > Subject: RE: Proposed: Evolution (and associated libs)
> >
> > On Tue, 2004-01-06 at 06:32, hellan acm org wrote:
> > > If Evolution goes in, gal should be polled into the evolution source
> > > tree. That way, we avoid making gal a part of the platform, and we
> > > make sure that other apps don't start depending on gal.
> >
> > This has always been our intention. Radek has code in his tree to yank
> > the gal dependency from gtkhtml already. The main trickiness in
> > evolution is pulling in the gal a11y stuff, but it shouldn't be too bad.
>
> You can't just delete a library - packages that link to it, while still
> being able to use the old version of the library, can't benefit from the
> improvements and bugfixes to an internal version of gal.
I'm not suggesting that gal be deleted. I'm urging that it not be
included in the platform. gal has never been a platform library.
When you include a library in the platform, you make guarantees about
stability and continued availability. Historically, gal has been very
unstable, so it's not suitable for the platform. Besides, only the
gtk_combo_ widgets seem to be of general interest, and those are not
enough to merit a separate library. And the names clash horribly with
current gtk+.
Non-core apps can kepp using gal as they please. Core apps which want to
use gal widgets should their own copy of gal, just like libegg is used.
Regards
Jon Kåre
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]