Submitted by “After several weeks of discussion the leaders of Compiz and Beryl have agreed
that the two communities shall reunite. This decision is supported by both David and Quinn and represents the majority decision of the administrators and developers in each community.”
I guess only good things can come out of this, they can always re-fork if beryl and compiz developers can’t get along.
so does this mean the united thing will be dependent on gnome crap or not?
thats kindof a huge plus for beryl.
For a news that show collaboration at his best, you could have resisted more than 3 seconds before taking a shoot at it.
Oh come on, it’s a fair question (thouth he needn’t have dissed Gnome like that). At first, Beryl had far fewer Gnome dependencies (and yes, depending on Gnome for configuration counts as a Gnome dep to me
). For people on other DEs it was a fair reason to use Beryl (note was ).
Rather than mod him down you could answer his question. Before the merge Compiz was working on a config framework that didn’t need Gnome. The combined entity can continue to use Aquamarine, the KDE windeco. Once it’s all joined up it should work just fine without Gnome if that’s what one wants.
Anyway it’s really good news. As much as the plugins differed at times, the core was pretty much the same in both projects, with effort expended to keep it that way. Much better not to have that redundancy. Congrats to both sides for working through stuff!
Does it still depend on GNOME libs?
I just checked my KDE-only Gentoo laptop, and to install compiz, compiz-extra and compiz-settings, the only dependency I have to resolve is dbus. I admit, I don’t know what the situation is on other distributions, but from what I see, compiz doesn’t have to depend on GNOME.
Correct me if I’m wrong.
Correcting becouse You are wrong (unless You have a overlay with compiz that does not need gnome-control-center):
acer downloads # emerge -pv compiz compiz-extra compiz-settings
These are the packages that would be merged, in order:
Calculating dependencies… done!
[ebuild N ] gnome-base/orbit-2.14.7 USE=”-debug -doc” 713 kB
[ebuild N ] gnome-base/libgnomecanvas-2.14.0 USE=”X -debug -doc” 598 kB
[ebuild N ] dev-perl/XML-NamespaceSupport-1.09 8 kB
[ebuild N ] app-text/build-docbook-catalog-1.2 4 kB
[ebuild N ] media-libs/gstreamer-0.10.12 1,799 kB
[ebuild N ] gnome-base/gnome-keyring-0.6.0 USE=”-debug” 466 kB
[ebuild N ] perl-core/Test-Simple-0.67 75 kB
[ebuild N ] app-text/sgml-common-0.6.3-r5 75 kB
[ebuild N ] dev-util/desktop-file-utils-0.12 341 kB
[ebuild N ] gnome-base/gnome-mime-data-2.4.3 USE=”-debug” 551 kB
[ebuild N ] x11-themes/hicolor-icon-theme-0.10 33 kB
[ebuild N ] virtual/perl-Storable-2.15 0 kB
[ebuild N ] dev-perl/XML-LibXML-Common-0.13 13 kB
[ebuild N ] x11-libs/libcm-0.1.1 269 kB [2]
[ebuild N ] x11-libs/libxklavier-3.1 USE=”-doc” 462 kB
[ebuild N ] app-admin/eselect-esd-20060719 2 kB
[ebuild N ] dev-libs/liboil-0.3.10-r1 USE=”-doc” 841 kB
[ebuild N ] gnome-base/gnome-common-2.12.0 USE=”-debug” 59 kB
[ebuild N ] gnome-base/gconf-2.14.0 USE=”-debug -doc” 1,852 kB
[ebuild N ] media-libs/gst-plugins-base-0.10.12 USE=”X alsa oss xv -debug -esd” 1,427 kB
[ebuild N ] gnome-base/libbonobo-2.16.0 USE=”-debug -doc” 1,374 kB
[ebuild N ] app-text/docbook-xsl-stylesheets-1.72.0 1,557 kB
[ebuild N ] dev-perl/XML-SAX-0.15-r1 58 kB
[ebuild N ] media-sound/esound-0.2.37-r1 USE=”alsa tcpd -debug -ipv6″ 375 kB
[ebuild N ] virtual/perl-Test-Simple-0.67 0 kB
[ebuild N ] gnome-base/gail-1.9.3 USE=”-debug -doc” 609 kB
[ebuild N ] gnome-base/gnome-vfs-2.16.3-r1 USE=”gnutls hal samba ssl -avahi -debug -doc -ipv6″ 1,856 kB
[ebuild N ] dev-perl/XML-LibXML-1.62.001-r1 226 kB
[ebuild N ] x11-wm/metacity-2.16.3 USE=”xcomposite -debug -xinerama” 1,791 kB [2]
[ebuild N ] app-text/docbook-xml-dtd-4.4-r1 94 kB
[ebuild N ] app-text/docbook-xml-dtd-4.1.2-r6 74 kB
[ebuild N ] media-plugins/gst-plugins-alsa-0.10.12 0 kB
[ebuild N ] media-plugins/gst-plugins-oss-0.10.5 1,325 kB
[ebuild N ] media-plugins/gst-plugins-xvideo-0.10.12 0 kB
[ebuild N ] media-plugins/gst-plugins-x-0.10.12 0 kB
[ebuild N ] gnome-base/libgnome-2.16.0 USE=”-debug -doc -esd” 988 kB
[ebuild N ] app-text/gnome-doc-utils-0.8.0 USE=”-debug” 403 kB
[ebuild N ] app-text/scrollkeeper-0.3.14-r2 USE=”nls” 664 kB
[ebuild N ] gnome-base/gnome-menus-2.16.1 USE=”-debug” 407 kB
[ebuild N ] dev-perl/XML-Simple-2.16 68 kB
[ebuild N ] gnome-base/libbonoboui-2.16.0 USE=”X -debug -doc” 916 kB
[ebuild N ] x11-misc/icon-naming-utils-0.8.2 65 kB
[ebuild N ] gnome-base/libgnomeui-2.16.1 USE=”jpeg -debug -doc” 1,448 kB
[ebuild N ] gnome-base/gnome-desktop-2.16.3 USE=”-debug -doc” 1,174 kB
[ebuild N ] gnome-base/gnome-mount-0.4-r5 USE=”-debug” 357 kB
[ebuild N ] x11-themes/gnome-icon-theme-2.16.1 USE=”-debug” 2,433 kB
[ebuild N ] gnome-base/eel-2.16.3 USE=”X -debug” 658 kB
[ebuild N ] gnome-base/nautilus-2.16.3 USE=”X -beagle -debug -gnome” 4,169 kB
[ebuild N ] gnome-base/control-center-2.16.3 USE=”alsa hal -debug -eds” 1,979 kB
[ebuild N ] x11-wm/compiz-0.3.6 USE=”dbus gnome kde svg -debug” 706 kB
[ebuild N ] x11-plugins/compiz-extra-0.3.6.0 USE=”-debug” 621 kB
[ebuild N ] x11-apps/compiz-settings-0.07 1,125 kB
Total: 52 packages (52 new), Size of downloads: 37,087 kB
You might wanna take a look at your use flags in /etc/make.conf – not to mention those set as default by your profile.
You’re rigth, I have -gnome in my make.conf, but I did put
x11-wm/compiz gnome
in my package.use, I don’t really remeber why I did so but I guess that compiz does not have some features without it. When I was posting I didn’t noticed that so sorry for too fast posting.
You guys need to get over the dependency thing. It is just NOT a big deal! No, your system will not be “bloated” from having extra libraries installed. Unless you are running your system on a 386 with 8 Mb RAM, this is simply not an issue. You only get the false illusion of ‘optimizing’ your system, when what you really get is a millisecond here and there, making absolutely no difference at all.
No, you cannot feel the difference. If you think you can, it’s an illusion. Human-Computer Interaction is generally not a very quick process, and when computers seem sluggish, it is always because of poorly written programs that brings the hardware to it’s knees (such as using the wrong algorithm on the wrong type of data, leaking memory leading to thrashing, etc.).
Linux, however, is not poorly written, and neither is Gnome or any of it’s dependencies. And neither are Compiz/Beryl.
– Simon
And where in my post You see _anything_ that says “I don’t want gnome deps becouse it slows my system” ?
I just noted that it needs gnome deps, I was wrong becouse I had gnome use flag in package.use (in make.conf I got -gnome) and I didn’t noticed that this USE flag is on with compiz but still, there is no word about speed or optimizing anything.
I don’t think I ever will. I was born with AntiGnome emblem tatooed on my forhead.
It IS a very big deal to me. And yes,the system will be very much bloated with extra tens of megabytes of libs and useless Gnome junk.
Exactly, the goal here is to avoid the poorly written, bloated, and badly performing Gnome libraries.
BTW, Compiz/Beryl are still highly experimental and not really suitable for production use machines.
So your last sentence is the evidence of how clueless you really are on this issue.
Correcting becouse You are wrong (unless You have a overlay with compiz that does not need gnome-control-center)
He’s right, you are wrong. Remove that “gnome” use-flag, if you tell portage to install compiz with gnome support, it’s obvious it’s going to install gnome.
Edited 2007-04-07 12:04
Considering that Compiz never had any hard dependencies on Gnome, the answer is no. The only dependency on Gnome that the original Compiz had was an optional GConf plugin. Granted, if you didn’t compile the GConf plugin then there was no way to configure Compiz, but it was optional nontheless.
One of the first “features” developed for Beryl (and one of the reasons for the fork in the first place) was the Beryl Settings Manager. I wouldn’t call that monster of a UI a “huge plus” for anyone.
The Beryl-Settings-Manager (especially its later incarnations) is still much easier/better to use than the GConf plugin.
This is good news. While you often hear about projects forking, you rarely hear about projects merging into one… and this one seems like a particularly good idea since the developers are still trying to do the exact same thing with the same toolkit in the same way.
Yet, I have good memories on merging projects.
Specially the remerge of egcs and gcc.
hopefully, upgrading to the new version will not be too much of a pain in the ass. i’ve emerged compiz and compiz-settings into my gentoo installation, it looks great, but it was not very easy to get it working.
anyway, i welcome this merger, and i hope that Compiz-Extra / “Composite Community” / … will just be named “beryl” to make the beryl community happy.
cheers!
From what I have read, the Compiz folks will be working on the core and the beryl folks will work on plugins to the core. IE one group for stable core development and the other for extras. Sounds like a good plan to me since, from what I know, the Beryl folks were using the Compiz core mostly anyway and just building on top of it.
…divided, we fall!
I’m looking forward to the best of the Compiz and Beryl worlds in the aftermath of this merge. Will we finally get the (relative) stability of Compiz combined with the cutting edge (and sometimes bleeding edge) awesomeness that is Beryl?
Not to be a naysayer but which license are they going to use because I know that Compiz and Beryl both used different software licenses. Does that mean that some work will be lost one way or another?
Also I thought that they were going to choose a new name for the whole project which I personally thought was a better idea but if this what they want more power to them. I hope that this merger works out well for all parties concerned as well as the OSS desktop. Who knows one day I may even be convinced to switch from Win XP to Linux one day, but I do find Knoppix very excellent for fixing and sometimes saving Windows partitions that have been nuked in some way. (KDE4 looks very exciting to me and maybe once 4.1 is released then FOSS will have an extremely compelling reason to switch; sorry but I just don’t care for GNOME that much.)
Further is the community ever going to have an answer to Direct X? I know that FOSS has OpenGL but that is only a competitor for Direct 3D and not the other components that make up the Direct X spec. I can’t drop Windows until I get games support, because I play things like WoW and Civ4 which as far as I know aren’t available for the consoles. I have considered running Linux primarily but my Windows desktop works fine for now and its all legal so why change right now.
I could give a rat’s ass about eye candy, but if this can increase the accessibility of the FOSS desktop then the project will be a winner. Governments will require accessibility frameworks and features before FOSS can ever be fully implemented. Good luck and I hope that one day in the future I won’t need Windows to run any program because every program (or most programs) that I want to run will operate on Linux (or maybe even OpenSolaris) and I will have choice.
Once again congrats to the Compiz/Beryl teams.
The license isn’t really much of an issue. The projects are very young, so anybody who has contributed code and hence holds copyright can’t be too hard to find to get consent for changing over to a new license.
As for an alternative to DirectX, try SDL.
SDL is very good but not an alternative to DX as such.
People forget that the game studios want to spend time developing their game, not in forums looking for support.
And Microsoft, NVida, ATI do support DirectX a lot. Just go to their booths in the Game Developer Conferences.
And most publisher will only support Windows as an option when releasing for the PC. Just see what happened to NWN 2.
The core will be under the BSD license, the plugins will be the choice of the developer.
Adam
small correction , I believe the core will remain MIT (to be compatible with X etc). Unless there’s a license change I don’t know of
The core will (and always has been)licensed under the MIT license so code from it can go back into x.org which is MIT.
http://lists.freedesktop.org/archives/compiz/2007-February/001413.h…
Edited 2007-04-07 17:00
D’oh… Of course. I’m always mixing up BSD and MIT.
Adam
Gaming support for Unicies is a hard thing to get gaming companies to do. ID and some others are the only ones I know that actually pump out Linux versions. However, I suggest You look at transgaming. Its a projected based off wine for running games in Linux. Most of your big games are supported. I will note that it does cost to download and thats to help support the project. As a gamer I still use Win primarily but I donate every month to help them with their cause.
I just want to know why there is no alternative to DirectX. It looks to me like there is incentive to develop such a thing so that development costs would be cheaper for Nintendo, Sony, Apple, and the FOSS community. OSS often works in cases where sharing development resources makes since and these different companies have the incentive to fight back against MS. Having a common and flexible API programming model for multiple systems that is like DirectX, but open and developed by MS competitors would make sense to me, but maybe I’m just missing something.
Sony and Nintendo sharing a common API and development environment would ease costs for both especially since they seem to be going after different markets in the console business now (and possibly in the future). Apple has a stake in order to make it easier to program for OS X. FOSS has obvious needs for such a thing in order to allow for more standardization across Linux, *BSD, and Solaris. Easing porting between the various flavors of Nix should encourage more development for those markets. If you don’t have to rewrite as much code between the various systems (basically more shared code) when porting then porting costs should be cheaper. OpenGL helps with this some, but there need to be answers for audio, networking, and peripherals as well. Maybe this is unrealistic but I think it is something worth working toward and I’m sure that it would be difficult to make compatible with the different systems but I’m sure that it could be done, such as the Wine project attempting to reimplement all the Windows API’s. If that’s possible than this should be relatively simple and there are examples of higher level abstraction such as KDE4’s Phonon (which I know is still Alpha at this point).
I know that SDL exists but that only seems to be a small part of what I’m talking about but maybe heading in the right direction. I mean for all the big names to come together and create a real alternative to DirectX that would standardize more things across the non-dominant or redundant parts of the software ecosystem.
Edit: Added last paragraph
Edited 2007-04-08 23:12
SDL offers graphics, sound, and input support, and you can always use plain OpenGL, OpenAL and just use SDL for the input. The simple fact is that if DirectX were an open standard like these are, it would also be chopped up into multiple bits because different groups want different things out of them and different companies have a stake in them.
This is great. Not only do we get more manpower on a single project (more devs = faster release times and more stability) but it will also help to remove the confusion over Composite Desktops. I for one was very confused between Beryl/Compiz when I first heard about it. Now with only one main project providing the same things, even the less advanced users will be able to get a sexy desktop up and running in no time.
Go TEAM!
NOTE: Probably should work on getting a better name for the compiz-extras part of the project. “Composite
Community” just doesn’t have any ZING to it.
The article says
“temporary name of “Composite
Community”.”
I know that… I was just saying that it should be a priority.
I’d like to take a minute to ask the Linux and other X Server based communities to come together and support our reunification efforts and goals. I personally believe that in a few years we will look back at the composited desktop as being nearly as big of a breakthough as the GUI was to DOS. But MS and Apple aren’t going to sit back and let us eat their lunch. They have composited desktops of their own and they won’t be sitting still. If we want Linux to be a serious contender on the desktop, we all need to reunite behind the technology that can make that happen. I believe that Compiz/Beryl is one of those technologies.
There were many issues that led to the initial fork, and none of them matter today. Over the last few months we came to realize that our visions were still very similar. In fact it recently became apparent that they were actually converging. Each project was independently re-defining themselves to be essentially the same project. At that point most of realized that there was an obvious answer: Reunification.
I don’t want to paint a rosy picture or minimize the problems we will encounter. There will be plenty of conflicts and this won’t be easy. I do want to thank all of you who support our efforts and let you know your support is needed and appreciated. We’re bringing together a large group of talented and enthusiastic developers and users. It’ll take us a little while get organized. But once we get our stuff together, man this is going to be fun…
Personally I think that this is one of the best things about FOSS. If enough people see a need for a fork, then a fork exists, but only as long as there is value for the fork. If times change and either one side sees that the other was right or that both see the wisdom of the other, the fork disappears.
I think they will, at least Microsoft. While Apple is years ahead of everybody with integration of composited desktop Microsoft has just entered the competition and it will be years before next desktop Windows will be released.
I had most of the same thoughts, but…
Microsoft might be slow to compete in certain markets, but once they enter a “market” they work very hard at convincing people that they are the only good option. They don’t always succeed, but they have a lot of resources to allow them to absorb failures.
Windows XP SP2 was mostly about improvements to their outdated security model. They needed to earn back some trust. I imagine that Vista SP2 (SP1 is generally more for bug squashing and security hot fixes than anything else) will mostly be about adding “wow” to Vista (and maybe getting UAC to function more sanely).
I don’t know what Apple is trying to do since they’re so secretive, but I’ve already seen some Linux demos that make Aero look very weak. If the developers can take the next year to make compositing effects easier to use for everyone, Microsoft will have a bit more motivation to improve their composited desktop before the next Windows release (or release a new version sooner).
It’s things like this that give me faith in open source.
I’m sure MS isn’t happy to hear such news – fragmentation and duplication in the open source OS world is the only thing that’s keeping them ahead of the pack at the moment 😉
“I’m sure MS isn’t happy to hear such news – fragmentation and duplication in the open source OS world is the only thing that’s keeping them ahead of the pack at the moment ;-)”
…well yeah, there’s that I guess…but the whole “you can have any flavour of OS on your shiny new machine as long as it’s Windows” thing doesn’t hurt either…
Congratulations to the Beryl and Compiz community!
I had a really good read on this, very detail,
and very useful information.Thanks.
MPEG to Zune
http://www.mpeg-converter.net
I had a really good read on this, very detail,
and very useful information.Thanks.
MPEG to Zune
http://www.mpeg-converter.net
Torrent link would be nice