Hi Rodrigo,
the difference between ubuntu and fedora is that they use PA for the whole distro, while we use it only for GNOME, unfortunately. This is the cause of most of the problems (not all of course), that users install alsa/oss/flash-based apps which, by default, are not configured to use PA because they are expected to be used in both KDE and GNOME.
Also, the other problems related to PA, most of them are upstream, and are getting fixes upstream, so those problems have shown up also on Fedora and Ubuntu.
Apart from the fact that PA was an example to give an idea, I think that you pointed out exactly the problem. In my opinion, if PA was not implemented for the whole distribution, maybe (I guess) someone did not consider it ready, or there was not enough time to do it. As a consequence, why not waiting for the next release to introduce it, giving upstream developers the time to fix the major bugs and in the same time planning a clean and user-transparent integration in the whole distribution? From a user point of view, I'm sure, there was no urge to have PulseAudio.
I don't think we work that way. When we release software, it has been tested. Of course, problems show up (like with PA) when lots of users (much more than testers) start using the software. Maybe we should be quicker in fixing the issues, you might be right there, but please, don't pretend to say we just release buggy software and hope users won't find problems. Apart from our testing, upstream GNOME gets a lot of testing before being released.
Upstream GNOME gets a lot of testing: my point does not discuss that, as I referred to the openSUSE implementation of GNOME, which gained the name of not being good over time, not just in 11.0, where we had very important improvements, recognised by almost everyone without a bias. Of course I'm not referring to small annoying bugs, but to the very evident ones, that clearly show no one (I include the community, of course) paid attention. To make two examples that I think sum it up well, let's reconsider the old dear main-menu and its memory leak of more than 200 MB. This issue affected SLED 10, openSUSE 10.2 and openSUSE 10.3. It was such an evident problem that it made that menu unusable on production workstations in order not to force the user to kill it manually or logout. It is hopefully fixed in 11.0, but it was plainly ignored for a long time (proofs in bugzilla). As a second, more recent, example that comes to my mind is the status of the Yast-gtk printer module, which was completely crashing in 11.0 final. We had other cases of applications not working in the final release and never fixed also, but I think I gave the idea. This had to be added, of course, to the slow process of fixing reported problems. Excluding who is in principle against GNOME, I think issues like these played a big role to create the current rumours about openSUSE GNOME implementation. Regards, A. -- To unsubscribe, e-mail: opensuse-gnome+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-gnome+help@opensuse.org