Wonderful writing style. Thanks for sharing your thoughts.

I really enjoyed your post. Thank you for sharing.

Wonderful writing style. Thanks for sharing your thoughts.

Really good information. Thanks for putting it out there.

Just installed xorg 1.7.1.
Am I dreaming or this new xorg is much faster than the 1.6?

Rémi: if you're using the Intel driver, there's a pretty good chance it'll be much faster. 2.9 will even be a little faster when we stabilize it :)

not to forget, that you can't and shouldn't fix ''upstream'' in the distro. as a distro you just serve ''upstream'' as best as possible to your users. .

Um.. wow, scratch that last stupid comment I made. Everyone makes typos right?
The newest x11 drivers DO work...

So... anyone care to shed a light on why none of the x11 drivers work with the new xorg-server? Seems that the dependencies haven't caught up with the newest xorg...

This is how ~arch has to be. The not infrequently demanded "stability" and problem-free operation of ~arch is ridiculous.

What should be expected is one-two different compiles and possibly running through an automated test suite, not developers wasting his precious time trying to get things polished even BEFORE it hits real installations.
That duty must be moved to the larger amount of installations and the overall massively larger time budget of ~arch users. This is the only efficient way to produce a stable branch, after all...

@Jeremy: I think it varies from arch-to-arch. However, x86 & amd64 are arches that should be easy to find ATs for.

Nice post. Yes, we shouldn't break the ~arch tree on purpose..But it seems like people just want a second stable tree which doesn't help us with much. But it does point out that the arch teams are overworked because they can't keep up (or we are not filing stablereqs in a timely manor, etc)

Hi,

Thanks for the effort on keeping up with X.

Also, keeping it cool and civil is a great way to respond to -positive and negative- criticism. After all, people criticise because they give importance and want things to be better.

I just wanted to say that I really appreciate your work, and that I was really excited when I found you that released such a large package on the same day as it was officially released. Thanks, and keep up the great work!

Keep doing what you are doing.

Thanks for the prompt addition of Xorg 7.5!

I feel honoured by such a lengthy response. Let me just point out a few things:

- I very much accept your explanation that version number don't mean much in the xserver. As such, my irritiation was unjustified, at least when pointed at you or the gentoo-devs.

- I actually file quite a number of bugs all the time. I used to be a gentoo-dev but quit because of lack of time (I don't want to get too emotional about it, but I just want to let you know that I quite know what I'm talking about)

That sentence really struck me: "The ABI of X libraries has not changed, but I'm pretty sure there will be compile errors in some packages"

The more I think about it, the more I understand what you tried to say, but when I read it the first time, my feeling was: well, when you upgrade, things will badly break very likely. And that's not the way unstable works. Unstable is a place where things in large are settled, but are not yet finalized and some minor breakage still can happen. May be that's the message you tried to get across?

Rémi: Yes, that's what I meant. The ABI is unchanged (meaning currently built packages will still work) but headers have moved around (meaning that new builds may fail). That's where I'm asking for both help and a little indulgence :)

If every unstable package would break in many cases, people would not use it.

>> you can't reasonably expect us to try every possible combination

No, surely not, sir! I know that's not possible and would never expect that


Cheers
Benjamin

Stable for stable.
Unstable for not stable yet (testing).

It seems to *me* like there is nothing to change.
BUT if people always complain about that, we should AT LEAST study the problem, TRY to find a "solution" to their problem...

Maybe you could create an other branch named "bleedingedge". It would contain hot "almost stable" versions of "important" packages (like X, firefox, kde, etc) using as less unstable libs as possible...
It's purpose would be to let people have a more "up to date" environnement and/or use more recent versions "fixing bugs"...

Is it too complicated? Too "expensive" to make?

Thanks for the announcement. If you assume compile problems, why is that thing unmasked? I've always been irritated by the way the xorg team handled masked/unstable/stable releases, as even rc's were unmasked at times.

There are probably a lot of people who put the xorg-server in package.keywords because they needed/wanted feature X/Y or because it fixed some bug for them (it did for me). So now I get a release that possibly breaks build in unstable?

I just thought I'd throw this in as it has irritated me numerous times. Unstable seems to mean different things with different herds and this can be a bit cumbersome to the user

Having said that, thanks for doing all the work - I'll surely test it out soon :)

Will there be a list you can save to /etc/portage/package.keywords like last time?

Remi, thx for it.

just a note:

xorg-x11 meta package should be bumped to 7.5 too.

For those who still haven't migrated to HAL (and/or cleaned up their xorg.conf's) I've put up a lenghtly HOWTO on my page :]

linkie:
http://matija.suklje.name/?q=node/56