First Xouvert Milestone Released

An anonymous reader writes " The first milestone of xouvert, the X-server replacement has been released. Xouvert includes MAS giving the X server its very own sound server. Nice. :) Also, just noticed that enlightenment quietly released an update to the 0.16 series. " (Here's a link to the Xouvert download page.)

Comments

The things people complain about X...

[ObviousGuy]

People complain about X a lot, but when it's all boiled down there really isn't much to complain about. X is a great windowing system.

Re:The things people complain about X...

[fucksl4shd0t]

I'm afraid that Xouvert shows the worst side of Open Source. And that is that anyone can write OpenSource. Where's all the profiling data showing where XFree86 is slow. Why if you're trying to improve on XFree86 are they using a code fork and not starting from scratch? It seems to me this whole project is based on a gut feeling that removing all that socket code will speed it up rather than doing the proper research.

Another poster already showed you their FAQ where they say they cannot remove network transparency.

I think the Xouvert actually shows one of the best sides of open source. They are being non-critical of the fact that the XFree86 organization is slow, bloated, and more or less unable to keep XFree86 in a constant, modern state. Instead, they are providing a 'branch' of XFree86 that will focus on being bleeding-edge and providing fast turnaround for development and testing, so that they can interface with the slow, bloated XFree86 organization to improve XFree86. I think that says a lot of good things about OpenSource, taking care of our own, getting the job done, etc.

Funny....

[IamTheRealMike]

People were talking about XFree forking for so long and nothing ever happened. Now within the space of a few months, we have two!

It seems at least to me that the freedesktop.org x server (kdrive) is where the interesting stuff is happening, but we'll see how the Xouvert guys get on.

sounds nice

[axxackall]

Xouvert includes MAS giving the X server its very own sound server. Nice. :)

Just nice? It's excelent! This is the biggest X Windowing achievement since first actual implementation of X Windows.

It is in human nature to assotiate visual and audio information in the process of percepting it. Therefore video without audio mean seriously broken usability. That's why I think all these years X Windows has been developed in essentially wrong direction. The made in recent XFree86 versions transparency, which is really just a candy, while so important prime functionality was missed all the time.

I am really happy that MAS in Xouvert now. I am going to switch to Xouvert as soon as possible. Good-bye, XFree86 - thank you for keeping me in the void silence all these years.

Re:sounds nice

[axxackall]

I see, you're simply anti-unix. You think there should be one monolitic application that has everything integrated to it.

RTFA - they integrated X and MAS, not merged the code. Unix *is* integration of protocol layers, listeners and daemons, and applications. The integration of MAS with Xouvert is done in completely Unix way.

When GNOME is everything - that *is* monolitic.

In case you haven't noticed, XFree86 does not, nor has it ever, come with sound-drivers, or sound apps, so I have no idea why you think it's the responsibility of X to handle sound as well.

RTFA, Xfree doesn't, MAS does. it's responsibility of MAS to handle sound and it's great that X and MAS are integrated now to handle both graphics and sound in a same network-transparent way.

If X is handling sound, how do I play sounds when X isn't running?

What *graphics* do you see when X isn't running? That's right. TTY is for system management tasks, not for entertainment. When you want to entertain - you run your desktop. When you are not local - you run it remotely. And now it will have sound.

Re:sounds nice

[GreyWolf3000]

When GNOME is everything - that *is* monolitic.

Not really; you don't type "gnome" and have a full blown web browser, media player, kitchen sink all-in-one app. Gnome is, essentially, a set of foundation libraries and application framework for building consistent applications that play nice together. The vanilla Gnome distribution contains many applications (a web browser, applets, gnome-panel, etc.), but Gnome is about as monolithic as a Linux distribution. Fedora or Debian may have a thousand programs on it, but that doesn't mean that they are monolithic.

What *graphics* do you see when X isn't running? That's right. TTY is for system management tasks, not for entertainment. When you want to entertain - you run your desktop. When you are not local - you run it remotely. And now it will have sound.

A lot of folks out there use X-less desktops. mplayer for example can output sound AND video to the command line (with framebuffer support), and I think the parent was afraid that if MAS were widely adopted, and integrated into X, then audio support for the command line would fade. The parent (albeit rather rudely) meant that by bundling a sound server with X, people who choose not to run X will not have sound (assuming MAS becomes popular).

What the parent failed to realize is that you can run MAS without running X (or if you can't, there's bound to be an X-less MAS server around).

Personally, I fail to see the importance of tying a sound server to the X server (even if it is merely association), but if it means the acceptance of a standard, network transparent protocol, I'm all for it. I'm sick of sound being nonstandard in Linux.

How's this going to work with KDE/gnome etc?

[carnivore302]

Xouvert has its own sound engine, MAS. If Xouvert catches on, does this mean that the sound engines of KDE and gnome will become obsolete, or will they collide with MAS?

If they collide, it basically means that KDE and gnome will have to support both X11 and Xouvert. I'm not sure if that is achievable. On the other hand, if they don't collide what's the use of MAS? I'm pretty happy with the way it works now. So I'll then continue working without MAS.

Re:How's this going to work with KDE/gnome etc?

[Amiga Lover]

If Xouvert catches on, does this mean that the sound engines of KDE and gnome will become obsolete, or will they collide with MAS?

There's a few places Linux has failed miserably for me as a desktop, and consistent audio has been one. If I get KDE audio working, six other non KDE apps suddenly go silent, If I get those working, KDE audio apps error on me. Same story sadly. Now, perhaps it's just me not knowing what to futz around with, but to repeat a cliche, "I shouldn't have to do that".

Perhaps kernel level device sharing would work, but I don't know if adding another sound engine would help much

Re:How's this going to work with KDE/gnome etc?

[TheRaven64]

Perhaps you should look at FreeBSD? It contains kernel-level sound mixing, and exposes several virtual /dev/dsp devices (/dev/dsp.0, /dev/dsp.1) which are mixed together to produce the final output. I have the KDE sound daemon pointing at one, the Gnome one pointing at another and leave `legacy' apps and games (which can't tolerate the latency imposed by one of these daemons) to use /dev/dsp (which is a symlink to /dev/dsp.0). In the 5.x series, this is handled automatically, and each request to open /dev/dsp returns a new mixer channel, rather than the device.

Having said that, MAS is not a replacement for /dev/dsp. For one thing, it is network transparent (so I can run a MAS enabled MP3 player, for example, in a remote X session, but still hear the sound.) MAS is cross platform, so I can (in theory) post the sound between any combination of machines that run an X server, as I can with X11. MAS uses a stream/filter graph-based model, and so is very flxible. I regularly use a remote X session, and audio is one of the things I have been missing. MAS should provide that, and this is the first real implementation I have seen. Hopefully it should make it into the main XFree86 trunk soon...

X is not bloated!

[lokedhs]

X isn't and never was bloated. People think it's large just because the framebuffer memory is included in the "ps" listing.

Read the explanation on the freedesktop site. There they mention the fact that people developed X on really old VAX machines. I even ran X myself on an old VAXStation II which had several times less memory than your average palmtop computer, hardware which happens to run X as well.

Re:X is not bloated!

[kinnell]

The X server itself may not be bloated, but the XFree86 source distribution certainly is. Everything from client libs to Xeyes are included in the build system, and working out how to configure it to only build the parts you need is not an easy task. Not to mention the amount of cruft you have to download. This is a shame, because there is no reason why it has to be like this, but clearly the XFree86 people aren't interested in the problem. If Xouvert can modularise it well like they plan, this will be a massive improvement in itself.

Re:Gombine and Gonquer, with XouverG

[t_hunger]

A consistent GUI allows for the things you learned in your word processor to be reused in your browser, e-mail client, etc. Thanks to the thousand of toolkits, desktop environments, support libraries, sound backends, printer support solutions, etc. that's plain impossible in X. So a user has to spend lots of time relearning how to do simple tasks for each application he uses (and mixing them up after learning them). That ruins productivity!

Wether someone runs one or onehundred word processors is absolutly irelevant to the GUI consistency discussion.

Who really wants all that garbage?

[Nichademus]

After checking out the following screenshot: http://freedesktop.org/~keithp/screenshots/sharp_s hadow.png and then reading the contents of the X-Chat window, specifically, "I'm hoping to do things that won't be fast enough with 2D/3D hardware as it exists today.", I have to ask: Who really wants all this shadowing, and translucent windows, and animated desktop graphics? I mean seriously, what's the point? Does it help you get you work done? Does it increase your productivity? I see it being more of a nuisance and distraction.

It certainly shows that Mr. Packard works for HP, what with him writing software that would require users to purchase new hardware just to have the next generation desktop. Hell, the desktop might as well be free, if we have to shell out the dough to purchase a new video card.

Re:Who really wants all that garbage?

[Just Some Guy]

Who really wants all this shadowing, and translucent windows, and animated desktop graphics?

I do. I really don't know how they'll benefit me, but I guarantee that someone will make a newly-possible feature that, once you see it, you'll wonder how you lived without.

OK, here's a tiny example. What if your window manager used translucency to indicate window selection: the window with focus is opaque. The one you just left is slightly less so. The one before that is starting to become transparent. I think that'd be a much stronger (and faster) visual indicator than "window with focus is dark blue, windows without focus are lighter blue".

Is that a trivial example? Sure. But the point is that we don't know what will turn out to be the productivity enhancing killer feature that we've been waiting for until we try it. These new features may very well be useless and unused, but they could also change the way we use our systems. New functionality is good.

Enlightenment release

[sirReal.83.]

"Also, just noticed that enlightenment quietly released an update to the 0.16 series."

uh... that was over a month ago, on November 5th. It was a good little bugfix release, though.

Enlightenment is a good example of....

[bnavarro]

Why keeping a project in an alpha/beta state is a bad idea. I used to use E a long time ago, but they never went 1.0, and all the distros just started ignoring it, so now I use Sawfish.

This is a real pet peeve of mine. There are many OSS projects that do this. OpenSSL, anyone? The question is, why?? There must be a stable enough "beta" version of E that could be considered production quality, and should have been bumped up to 1.0 release status. I know that this is the case for OpenSSL, and a lot of other OSS projects out there. The fact is companies and non-hackers don't like adopting software that's advertised as "beta" quality. If you wan't your project recognized in the Real World, step up to the plate.

I know this sounds like a whining rant, but I belive that the plethora of OSS projects forever stuck in a "beta testing" phase is one reason for hesitation for mainstream adoption of Linux.

Re:Enlightenment is a good example of....

[who what why]

There are plenty of 1.0, stable window managers out there. E16 has never gone gold because the developers (i.e. Rasterman) want to turn it into an entire desktop environment, not just a window manager. They launched on a massive rewrite (and then another, and another).

E has become a place for experimental ideas that just wouldn't be accepted into a more stable project. Check out Rasterman's research into software vs OpenGL hardware rendering for Evas.

We already have enough stable window managers (especially for use with GNOME and KDE)... The Enlightenment team are working on something new and different, the lack of which is something often lamented around here always complain about in these X discussions. Let them work it out at their own pace, and maybe you'll be blown away by the next release.

Re:Gombine and Gonquer, with XouverG

[adrianbaugh]

Asking for consistency between desktop environments is unreasonable. For one thing, it imposes a burden on developers who are ultimately trying to scratch their own itch. For another thing, nobody asks for consistency between MacOS and Windows environments, yet KDE and gnome have no reason to be any more similar than those two. The fact that they both use the same server application (X) is irrelevant - the projects themselves are as different as chalk and cheese (one written in C, one in C++; one using bonobo for IPC, one using something else, one focussed towards strict HIG, one using different UI guidelines etc.) and it is quite remarkable that they coexist as well as they do. If you stick to one or the other then you get consistency, just as you want. If you mix and match, that's your lookout.
Besides which, have you ever really considered the "consistency" of Windows apps? Internet Explorer has a different feel to Office apps, which in turn are different to apps made by third parties (nobody will convince me that Windows Explorer's CD-burning capability shares anything in terms of look or feel with Roxio CD Creator, or that Excel is consistent with Quattro Pro).