Archived

This topic is now archived and is closed to further replies.

What is going on with XFree86?

This topic is 5029 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

First off, I admit that most of my knowledge on this subject comes from reading the headlines/summaries on slashdot. Second, this is kind of me just rambling and trying to make sense of some news. Anyway, what is going on with the XFree86 team? It seems like I heard they had forked the project, but then maybe they had merged again. Now, they have altered their license so it may not be GPL-compatible. As a result, a lot of distro''s are not going to use XFree86 4.4. Is that a fairly accurate? Did I miss anything important? I''ve also heard a lot of talk about replacements and alternatives to XFree86. Have any of you guys used any of them? Are there any good ones? It seems like people may be ready to move in large numbers over to an alternative, and there also seems to be interest in a new replacement system for X. Are there non-X based environments that are any good? Anyway, thanks for the help, just wondering what you all know or think.

Share this post


Link to post
Share on other sites
replacement for X? No.

Replacement for Xfree86? yes.

The roadmap looks like this at the moment:

1. see if David Dawes backs down on the license thing.
2. if not, for now just keep up with xfree 4.3
3. when that becomes problemsome (new hardware, etc) move to another X server (brand new one that''s a result of this, or xouvert is already there, i think)
4. eventually fd.o''s xserver will mature, and looks very innovative.

Share this post


Link to post
Share on other sites
/. had some interasting news about a new replacement project for XFree that was started: The Y Window System.

As of the 4.4/4.3 issue: Some of the code in 4.4 has some lisencing mishaps, so they are reverting back to 4.3 to stay out of legal trouble. I am not 100% acurate about this, I havent done my research

Share this post


Link to post
Share on other sites
quote:
Original post by C-Junkie
1. see if David Dawes backs down on the license thing.
2. if not, for now just keep up with xfree 4.3
3. when that becomes problemsome (new hardware, etc) move to another X server (brand new one that''s a result of this, or xouvert is already there, i think)
4. eventually fd.o''s xserver will mature, and looks very innovative.

Even the early releases of XFree86 4.4 were under the nicer old license. I''m sure Xouvert will not be accepting the newer brokenly licensed code (just look at the people running Xouvert), so Xouvert may unintentionally become a "clean" XFree86, as it did seek to mostly share its code base with XFree86 originally. FreeDesktop''s Xserver is pretty nice already, it just needs a lot of drivers (XFree86''s drivers can''t be directly ported because FreeDesktop''s Xserver requires additional features from the drivers). FreeDesktop apparently wants to use OpenGL to accelerate the entire X server when possible, which also requires changes from the driver architecture of XFree86 4.x.

So, in the end, it''s no big matter when it comes to the average user''s X experience. While I don''t know Dawes'' inspiration for dirtying the licence for new code, he isn''t going to get much leverage for his plans through XFree86; there are just too many ways to work around the issue almost painlessly rather than just put up with it.

quote:
Original post by aftermath
/. had some interasting news about a new replacement project for XFree that was started: The Y Window System.

The Y Window System isn''t meant as a replacement for just "XFree", but for the whole X Window System. Like the so many other seeking-to-be-X11-replacements, I''ll wait until I see real results before giving it much real attention. While it may just be personal opinion, I think it attempts too high a level of abstraction which makes it prone to design flaws that are impractical to remove (if the level of abstraction doesn''t count as one on its own). Seems like it made many of the same decisions as Fresco. While some of their changes from the X11-way-of-doing-things are probably good ideas, there''s a couple big ones that just put me off from liking the projects.

Just about the only X11 alternative I give any heed to is DirectFB, but it''s obviously not meant to altogether replace X11 (it just doesn''t have some of the capabilities that X11 does). It is an alternative is some situations though. Mostly by focusing on simpler goals, they''ve been able to make a design that doesn''t bother me somehow . It doesn''t hurt that the GTK+ (GDK) and Qt people seem to give a decent amount of support to DirectFB either.

Share this post


Link to post
Share on other sites
In short, the XFree86 project committed suicide.

Basically, we''re going to use XFree86 until another major project comes to a reasonably mature point. I''m keeping my eye on 3--Xouvert, FreeDesktop.org, and Y Window System, FreeDesktop being my fav so far.

Share this post


Link to post
Share on other sites
quote:
Original post by Promit
FreeDesktop being my fav so far.

Have you used it? I heard it supports hardware accelerated transparency, just like Aqua, so I could have [true] transparent terminals and shadows without the slowdown of software transperecy. Anyway, have you tried it? I dont know if I want to switch to it because someone in irc told me it was extremely buggy.

Share this post


Link to post
Share on other sites
The thing is that XFree86 is soooo stable! In Windows, if one program hangs, you have to press Alt-Ctrl-Del about 463 times, then an End Task window comes up. You press "End Task", and the End task window disappears without any other visible effects! Then the window you were trying to move suddenly moves rapidly, leaving bits of window everywhere. Then, 15 End Task windows appear and nothing happens. If you're lucky you eventually get the BSOD telling you to give up, Windows has won the battle yet again.
X, on the other hand, is so stable (along with the Linux kernel) that if one thing goes wrong (even I admit that its possible, though unlikely, for an open source program to fail) everything else continues to work fine.
What I don't understand, though, is why we can't, as an open source community, just pick up XFree86 and go on on our own where they left off.

EDIT: embarressing spelling mistake

Zorx (a Puzzle Bobble clone)
Discontinuity (an animation system for POV-Ray)

[edited by - clum on February 21, 2004 11:31:14 PM]

Share this post


Link to post
Share on other sites
quote:
Original post by clum
What I don''t understand, though, is why we can''t, as an open source community, just pick up XFree86 and go on on our own where they left off.

There''s nothing stopping anyone from maintaining XFree86 from before it was tainted. That''s what I now kind-of expect of the Xouvert project in the short term, at least. All the code that was under the nicer original license is still under the nicer license.

Share this post


Link to post
Share on other sites
The license didn''t really change, it''s just clearer what the intent always was. I prefer the xfree license, and those used by libraries such as boost.

The license is less restrictive than the GPL or LGPL, so I don''t see how it''s a conflict. GPL would taint the xfree license.

Share this post


Link to post
Share on other sites
quote:
Original post by Magmai Kai Holmlor
The license didn''t really change, it''s just clearer what the intent always was. I prefer the xfree license, and those used by libraries such as boost.

The license is less restrictive than the GPL or LGPL, so I don''t see how it''s a conflict. GPL would taint the xfree license.


You, apparently, have ignored everything that''s happened in recent times, or have "read the licesne yourself" and prefer to trust your own judgement to that of a lawyer on this legal issue.

Share this post


Link to post
Share on other sites
I see the point of the slippery slope argument, but an acknowledgement does not seem like a lot to ask for to me.

It's only required with binary-only redistribution or if you already provide documentation with other third-party credits. No "written"(printed) documentation, no third-party credits, but the source code, means no acknowledgement. Libraries (such as Gnome, KDE, gtk, & qt) are specifically excluded from requiring any acknowledgement.

[edited by - Magmai Kai Holmlor on March 3, 2004 10:07:00 PM]

Share this post


Link to post
Share on other sites
again, you're trusting your own judgement over that of a lawyer's.

at the current time, they're saying this only applys to the server code _for now_.

Putting it in the server causes two problems:
1. Threat of application to the client code
2. Clear signal that David Dawes will do what he damn pleases about what he damn pleases and cannot be trusted in any way.

personally, I think the guy is scum because of what he's done recently.

Technically, server code based on the 1.1 license is only a problem for debian because it is absolutely not DFSG-free. The other distributions have just gotten completely disgusted and decided to ditch the project lead by this david dawes self-appointed leader (EDIT: toned it down a bit)

Most distributions will probably be switching to both a multiple X server capable distribution method, as well as making the X.org X server hosted at freedesktop.org (called Xorg, NOT XSERVER. xserver is keithp's kdrive and isn't... ready) the default X server.

[edited by - C-Junkie on March 4, 2004 9:41:18 PM]

Share this post


Link to post
Share on other sites
You''d probably have to have read quite a bit on mailing lists to understand.

In short:
1. Kicks Keith Packard (one of THE smartest people there are with respect to X11 and graphics in general) out of XF86, for reasons that are entirely obfuscated. (I could probably give you a history, but that''s a sufficient explanation)
2. In responce to public outcry, creates a ''forum'' mailing list to discuss the project in general, which despite excitement about the possibility to fixing problems, results in nothing. (the forum mailinglist is now entirely about the license change)
3. Dissolution of the core team, which would be a good thing, if David Dawes hadn''t apparently taken it upon himself solely to take over.
4. Radical change to license which is both entirely uncalled-for and made without warning, or request for comments.

Share this post


Link to post
Share on other sites