With GNOME 2.6 not out yet, plans for GNOME 2.8 are already starting. First interesting
thing, at least for me, is Alex’s work on the GNOME-VFS network:// module, which will
be using service discovery APIs to discover servers in the network. This is something
that I’ve wanted for a long time, but had never got the time to work on it, so it’s
really nice to see Alex starting work on it.

Also, on the personal front, I’m waiting for the right time to propose GNOME
‘s inclusion in GNOME 2.8 (screenshots here).
It is a full featured application, much much better than the application it was
initially based on (Mac OS X’s network utils tool).

GNOME Programming Language

After following all comments about the “need” of a high-level language
for GNOME, I’d thought about giving my 2 cents. I really don’t think we need
to apply on a single language, be it Java, C# or whatever, we really should
continue to allow anyone to use any language. What would be really interesting
though would be to have that common runtime that both Java and Mono provide.
This, if it were done to be used from any language (like Mono does), will help us in reducing the
work needed for creating bindings (or even removing that need at all, since we’ll
just have to add support for new languages to access the runtime, giving them
access automatically to all features in the runtime).

I think that’s the interesting bit, and something we should really think about. We really
need to have
easily accessible all GTK/GNOME APIs to all languages, with no delay (as it happens
now with bindings, which are released almost always some time after the C library).
Automatic creation of bindings could be a good solution for the time being, although
that seems, from what I’ve heard, a bit complicated, given that human intervention is
needed in a lot of cases.

About development tools, even though I don’t like at all IDE’s (at least the ones I’ve used,
many years ago), it is another thing we should put efforts on. Anjuta seems to be loved
by new Linux developers (that’s what I’ve heard), and if it continues (as I know it does)
to just integrate into the development tools we already use (automake, autoconf, cvs,
and our beloved libtool), I think it can be the perfect entry point for new developers
right now. For the future, whenever we change the development tools, we should make sure
we always have an easy-to-use IDE that makes use of all those development tools. If
that IDE hides the (probable) complexity of the development tools, then the new developers
don’t really need to learn lots of obscure stuff.

Multimedia frustration on SuSE

Everything works as expected on SuSE, except, at least for me, the Totem and
Rhythmbox packages from the xd-unstable channel. Rhythmbox can’t play more than
one song and Totem sometimes says it doesn’t have plugins for formats it was
playing correctly before :-( Of course, since I was a (very) happy user of those
2 applications on Debian, everything points to the Ximian (unstable) packages
being broken. So, to solve it, I’ve been trying to install the totem/rhythmbox
versions from other Red Carpet channels, with no luck at all. Is there some
Swedish conspiracy against SuSE or what?