Status of xserver/debrix/modular tree?

[Bernardo Innocenti]

Daniel Stone wrote:
> On Thu, Feb 10, 2005 at 01:28:55AM +0100, Bernardo Innocenti wrote:
> 
>>The Xserver wiki page says the project is dead, pointing
>>users to Debrix.  The link to Debrix, however, has
>>disappeared from FreeDesktop's Software directory and
>>the Arch repository doesn't work any more.
>>
>>Actually, the page was still there:
>> http://www.freedesktop.org/wiki/Software_2fdebrix
>>
>>...so I found out that Debrix migrated from arch to bazaar.
>>I tried checking it out, but it's mostly old stuff which
>>doesn't even build.
> 
> 
> Interesting.  If you check out daniel at fooishbar.org/debrix--devel--0.1
> (as opposed to daniel at fooishbar.org--2004/...), it should build and work
> fine.  I put in some work a few weeks ago into ensuring that you could
> bootstrap cleanly from the bottom of xlibs to the top of Debrix, so it
> should all work.

Thank you!  Checking it out right now...  I'll let you know as soon
as I find some time to test it.  Could you please update the wiki and
restore the link to it from the Software page?


>>The modular xlibs and xapps trees seem to receive very
>>little development.  It also looks like they're not
>>being kept in sync with the monolitic tree.  The XCB
>>enabled libX11 would be a nice new feature which already
>>works fine today.
> 
> Er, you do realise that the modular xlibs is fully in sync with the
> modular tree?

I didn't notice, and I vaguely remember reading a post in the xorg
list saying that xlibs and xapps were abandoned/unmaintained.
Later on, I was surprised to see more commits going into it.


>>I'm curious about the future deriction of these projects?
>>Is there a plan of some kind?  If so, where is it being
>>discussed?
> 
> Some would argue that the future is modularisation.  Others have
> argued that it's doomed to fail and we shouldn't even try.  That
> debate is happening right now (on xorg@ -- it's very hard to miss),
> and that will determine the future direction of the project.

As silly as I may seem, I did miss it.  Now I've browsed
through February's xorg archives and I still can't find it.
What's the exact subject?

By the way, my $0.02 on this is that modularizing is the
only way to keep a huge project maintainable.  Distributors
split up xorg's distribution in several smaller packages anyway.


> xserver@ and xlibs@ are considered dead lists, as we moved them to be
> within xorg at .

That's a good move IMHO.


> As for the communication thing, well -- sometimes the communication
> happens on IRC, or sometimes there just isn't any.  People know what
> they have to do, and often do them.

I tried logging on #xorg once, but there were mostly users,
not Xorg developers, so I gave up.


> There just isn't the critical mass of developers in xorg that I think
> you think there is.0

I'm curious why Linux distributors don't contribute as
much to Xorg as they do to the kernel and GCC.  It's
quite clear that at this time Linux as an OS needs a
better desktop infrastruture more than it needs a better
kernel.

Some hardware vendors chose the way of maintaining proprietary
drivers instead of contributing to XFree86.  After one year
of open development with Xorg, they still haven't changed their
minds.

Developers of desktop systems (KDE and Gnome) are mostly
waiting for composite to build MacOS-like effects in their
window managers.

Toolkits (QT and GTK) are not helping out too.  GTK appears
to be moving to Cairo, while Trolltech said they would have
used XCB if it was ready in time for QT 4.0.

This complete lack of cooperation puzzles me a lot: why
isn't developing for X as exciting as developing for the
Linux kernel or GCC?


>>I know Xorg is based on volunteer work.  All OSS
>>projects are.  I might have overseen something, but
>>in order to be successful and attract more developers,
>>Xorg appears to need more coordination/PR work.
> 
> X.Org's only position as relates to development right now is release
> manager, which changes each release.  For the 6.7.x series, it was
> Keith Packard, Egbert Eich and Kevin E. Martin; for 6.8.2 it has been
> Roland Mainz.  I believe 6.8 was managed by Kevin again, and 6.8.1 was
> managed by a bastard trio of myself, Kevin, and Egbert Eich.

I think all of Xorg's releases have been very good, both in quality
and schedule.


> Again, X.Org really does not have the critical mass you think it does.
> Everyone is too busy working on code, really, and if you want to join
> and contribute, you are invited to do so, but at the moment, the
> project is desperately undermanned.

I see, but I still can't understand what's keeping people
from joining the Xorg project.  Maybe we need more PR work?

No, I think the f.d.o. site is quite visible and informative.
Getting a CVS account here is much easiter than in GCC, and
you also seem to have an open policy for accepting patches
from outside.  Xorg also gets some coverage on Slashdot and
LWN...

The only explanation I can give is that people come here and
see little activity on the mailing-lists, they get no idea
what the other developers are doing and what the future
direction is.  So they just loose faith and go away.

-- 
  // Bernardo Innocenti - Develer S.r.l., R&D dept.
\X/  http://www.develer.com/