GNOME dependencies

jmak jozmak at
Thu Aug 9 22:16:39 UTC 2007

On 8/9/07, Jani Monoses <jani.monoses at> wrote:
> One other argument in favour of GNOME deps. They seem to come whether we want them or not,
> and resisting or working around them is taking up a lot of time. During dapper and edgy
> Gauvain and I spent a lot of time making separate -gtk only packages. There were 2 or 3 only
> and that shows how little you get for much effort if it goes against whatever everyone else is
> doing, in our case upstream GNOME and Ubuntu.
> Now is anyone using Gutsy already - it's a devel list after all :) if you look,
> evince-gtk is depending on libgnomeui (probably fixable) but the new gaim (pidgin) for some reason
> uses libgnome now and gstreamer. Making a pidgin-gtk package seems to me like a lot of waste of time.
> It would be good to have but it's a lot of work upfront and then continual maintenance and bother not
> only for us but for Ubuntu desktop devels. The other choice is to drop gaim because of these dependencies :(
> Here's Tribe4 with bonobo, libgnome and gstreamer on the CD. At least there it does not seem to make a noticable
> difference :)
> So as long as we're using the Ubuntu repos there's only so much we can do, and we do not have the liberty of
> Zenwalk or any other independent distro to optimize the packages to the bone. For this price we get the
> long term maintenance, easy upgrade, lots of packages etc advantages of Ubuntu.

I think, we have difficulty figuring out the direction xubuntu should
go because we have never really devised a clear identity for xubuntu.
We don't have a clear picture in our heads about xubuntu's status.
Unlike ubuntu, kubuntu or edubuntu. Ubuntu is a user oriented
operating system with the gnome desktop, Kubuntu is the same thing but
with kde, edubuntu is a gnome based os targeting educational
institutions. With these flavors everything is crystal clear.
Xubuntu's situation is ambiguous. We don't know exactly who we are
targeting. Slow computers? Maybe, but as xfce gets sexier, an
increasing number of people use it on high end machines as well. In
this year, for instance, I've installed xubuntu on three machines with
dual core processors with 2 gig ram because customers preferred
xubuntu to ubuntu. All in all, we can only come to an agreement if we
first, chart a path for xubuntu, the way ubuntu or kubuntu developers
did. After that everything will fall into its place much more easily,
otherwise these kind of debates will persist and development suffers;
ambiguity doesn't help the development of anything.



More information about the xubuntu-devel mailing list