Re: Embedded uses diverging Re: GTK+ 2.8.13 released

["Matthias Clasen" <matthias clasen gmail com>]

On 2/28/06, Sean Kelley <sean sweng gmail com> wrote:
> At some point as Gtk+ progresses on to Gtk+ 2.10 and Gtk+ for embedded
> use remains pretty much a heavily patched Gtk+ 2.6 (nokia 770 et al),
> does it make sense to have a real branch for strictly embedded use?
> The reason I say that is the emphasis on things like Cairo, which is
> heavily floating point based really has no place in an fpu-less ARM
> processor.
>
> It also becomes hard to backport various and sundry bug fixes.
>
> Any thoughts?
>

It doesn't make sense to have a non-Cairo branch. Many of the maemo patches
are already in GTK+ head, and more are coming. In the long term,
we need to work on making Cairo more suitable for embedded use.
In the short term, I have no objections if people using 2.6 want to maintain
2.6 bugfixes in the gtk-2-6 branch in cvs, and maybe even do another
bugfix release from that branch. But it should be pretty clear that the
GTK+ team doesn't have the man power to maintain that branch.

Matthias