Le 06/11/2013 07:46, Tim a écrit :
We are still looking at updating the patches to have a build of the new version
to test (that works is going to be useful, even if that's not
this cycle). Then we expect to have some theming issues to resolve (as usual
with recent GTK update). Once those are resolved we can give some
testing to the new version and see what are the costs/benefits of the update
I believe its mainly just the custom menu hack that needs updating?
I guess that's the main one yes (out of themes issues)
Big +1 for this, however the changes also affect gnome-desktop3, which now
depends on mutter for display config and idle manager. Maybe that
needs to be forked as well?
Having a forked library is trickier, what happens to rdepends? (e.g
nautilus, with what lib do we build it? do we need to have another
nautilus build to use the other library?). I need to look into that one
and see if we can fold the different APIs in the same library (or update
and keep a compat API for non gnome-shell users)
I thought the GtkHeader was introduced in gtk 3.10, does that mean some apps
have just gone and mimicked the look, without using the proper gtk API?
We looked a bit and what GTK 3.10 update would imply. It's likely that
if we get the new GTK some people are wanting to update some of the
apps, which makes us hit that issue...
from Ubuntu GNOME perspective we would like to see
- update cogl/clutter to atleast 1.16, or preferably 1.18
I've no real opinion on cogl/clutter, I guess it's up to the Ubuntu
GNOME to decide on that one (assuming that there is no new depends
involved that would impact apps under Unity)
- update to BlueZ 5
That's not going to happen this cycle. The new bluez is incompatible
with the old one and not co-installable. The only desktop ported is
GNOME 3.10, and there has known regressions there (e.g DUN support not
working)
Cheers,
Sebastien Bacher
--
ubuntu-desktop mailing list
ubuntu-desktop@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-desktop
