Xfree no longer free ?

[Eamon Nerbonne]

Some people seem to suggest that it's partially an FSF or more general OSS attitude problem. The fact that the Apache lisence is GPL-incompatible isn't the same as the fact that the new XFree lisence is incompatible since Apache already always was incompatible... therefore there aren't (supposed) to be any lisence violations by upgrading: none of the GPL-ld programs ever could link against Apache-lisenced code anyhow (and vice-versa).

Furthermore the incompatibility that Apache 2.0 introduces is one which GPL 3.0 might address anyhow (namely that of patent lisencing, I can't recollect the details).

The new XFree lisence is incompatible with the GPL because it restricts redistribution; it requires the program and or documentation to be modified as a condition of distribution; modified to acknowledge the XFree86 consortium. The old lisence did not do so; it merely required you to acknowledge the copyright and not to claim that you wrote the software; which isn't the same thing.

I'm sure politics plays a role too; but the issue isn't necessarily the open-sourceness of the new lisence - it's simply a practical horror. The fact that this doesn't seem to interest the xfree organization doesn't exactly help though - on there website AFAIK they simply don't address that issue (they talk of the new lisence, but not of how they see it's GPL-compatibility and what all the poor souls linking to Xlib are supposed to do all of the sudden when the lisence changes).

I really don't think that _any_ blame falls on the FSF here at all. First of all they do acknowledge other lisences; but that doesn't make them GPL compatible, and pointing out such legal issues is important if you want the lisence you use to have any meaning at all and the OSS environment not to totally fall into a legal spiderweb. So even if their self-important attitude bothers anyone, they're _definitely doing us all a great great service by their actions - esp. as I don't see too many normal developers jumping to bother with legal issues _at_all_.

[IvanHoe]

Perhaps I'm missing something; but again, there's absolutely nothing in the XFree86 1.1 license that requires you to even mention the XFree project if your software links against Xlib. I don't see how it effects the GPL or software that uses the GPL in any way. Perhaps Stallman should actually read the XFree lincense.

[Roguelazer]

just to make this thread more useful:

Version 1.1 of XFree86 Project License.

Copyright (C) 1994-2004 The XFree86 Project, Inc.
All rights reserved.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicence, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

1. Redistributions of source code must retain the above copyright notice, this list of conditions, and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution, and in the same place and form as other copyright, license and disclaimer information.
3. The end-user documentation included with the redistribution, if any, must include the following acknowledgment: "This product includes software developed by The XFree86 Project, Inc (http://www.xfree86.org/) and its contributors", in the same place and form as other third-party acknowledgments. Alternately, this acknowledgment may appear in the software itself, in the same form and location as other such third-party acknowledgments.
4. Except as contained in this notice, the name of The XFree86 Project, Inc shall not be used in advertising or otherwise to promote the sale, use or other dealings in this Software without prior written authorization from The XFree86 Project, Inc.

THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE XFREE86 PROJECT, INC OR ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

[ecatmur]

And because people have obviously failed to {RTFA/%A/T}:

Part of the GPL:
6. Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties to
this License.

"3) Where is the derivative work boundary ?

The problem is further muddled by the place where the boundary for
something being considered a derivative work. The GPL, contrary to the
LGPL, considers that everything linked with a another binary is a
derivative work of it. I believe that this is mostly done so that
someone could not modify or extend a GPLed library by putting the
modified work in a wrapper or in the binary itself, which the LGPL
allows for dynamic linking, and for static linking with some additional
work. In our case, the problem is the opposite, since the XFree86
libraries may impose their further restrictions to the GPLed code, even
if it is the GPL here who cross the boundary."

[sgtrock]

sgtrock wrote:
As more and more people decide that they want their contributions recognized, the list quickly becomes far too long to be easily displayed. It also becomes a real nightmare to make absolutely sure you've identified everyone that you need to acknowledge.

Ian Goldby wrote:
That's not what the new Xfree license says. You don't have to acknowledge every contributor. This is the acknowledgement that is required:

"This product includes software developed by The XFree86 Project, Inc (http://www.xfree86.org/) and its contributors"

You don't even have to display it. You just have to include it in a readme file somewhere in the distribution.

Go back and re-read my post. You missed two key points there. First, not all contributors to the XFree86 project have agreed to this license change. I specifically mentioned Alan Cox because I recognized his name from his many contributions to the Linux kernel. I didn't even know that Alan had contributed code to the XFree86 project. It turns out he wrote some drivers for some old video cards, and is still the current maintainer.

Alan has stated categorically he is opposed to the license change because he wants his code to available to the largest number of people possible. If Alan has not assigned his copyrights over to the XFree86 project (likely, as he has always maintained his copyright for his other open source work), then the XFree86 project probably cannot legally include his drivers. They certainly won't be getting any additional support from him.

"So what?" you might say, "It's only some old video drivers no one cares about."

Well, multiply Alan's contributions by some unknown number of other contributors who have already taken the same position, or who might in very near future. All of a sudden you see a project that will be dead in the water very quickly due to a lack of solid contributions.

The other point that maybe I could have stated more clearly is that the *BSD folks found that it wasn't identifying the contributions of just ONE project or individual that got them into trouble. It was the contributions of eventually HUNDREDS of contributors and organizations that did. The XFree86 project attempts to re-open that entire can of worms. Sure, it's just one project. Today. What about in 5 years? We'll be right back where the *BSD people were.

The other point that I wanted to address was this comment of yours:

Ian Goldby wrote:

Just as a parenthetical node, what really irritates me is that the GNU people seem to consider the GPL to be the daddy of all open-source licenses. Effectively their clause 6 (stating that you can't put any other restrictions on GPLed code other than that required by the GPL itself) makes all other 'compatible' licenses subsets of the GPL. What arrogance.

Look, if you don't like the GPL for code you write, then don't use it. Use the XFree86 license. Use the MIT license. Use the BSD license. Use a direct copy of Microsoft's EULA for all I care.

There are a bunch of projects that run on XWindows that happen to use the GPL as their license. No one twisted the arms of the people in those projects to use the GPL. They were just as free to choose a different license as you are. So far, not ONE of those projects has announced that they want to change their license to accomodate the XFree86 project. Actually, several of them have come out in opposition. If someone else (not necessarily members of the FSF) choose to use the GPL, then you are legally (and hopefully ethically) bound to honor their wishes. BTW, very few of those projects (with the exception of the core Gnome distribution) have any affiliation with, as you put it, "the GNU people".

So, where does that leave the good people who are actually packaging all this stuff up for the rest of us? No distribution can legally deploy a version of XFree86 with this license AND deploy all the GPL'ed apps that sit on top of XWindows. They understand that under international copyright law, doing so would open them up to HUGE penalties. U.S. law, for example, provides for damages up to $150,000 per copy distributed if someone does decide to sue.

The people creating those distributions are also, by and large, honorable people. They care passionately about respecting the wishes of the people who actually wrote all of the software included in their packages. Even if they weren't staring at the potential of lawsuits brought by hundreds of people in dozens of countries, they probably would choose to include either the applications or the XFree86 package. Not both.

Maybe an example would help illustrate the issue. Let's say you're a guy who wants to create a brand new distribution. You want to make it as useful as possible to the largest number of people. You also want to show respect to the people who have actually written the code you will be including in your WhizBang Distro From Heaven (tm).

You are immediately confronted with a choice. You can choose to use the new v1.1 licensed XFree86 package and whatever small set of apps are compatible with it, or you can choose to use a slightly older version of XFree86 AND all the GPL XWindows apps that are out there.

Man, for the distros it's a no brainer. I'm surprised it took more than a minute or two to reach the obvious conclusion.

Nope, I think you've misidentified who is being arrogant. IMNSHO David Dawes is the one who is out of line. He thought that he would be able to unilaterally change the XFree86 license and everyone would just naturally accept it. What's happened instead is that he has singlehandedly managed to kill the XFree86 project. Even if he immediately withdrew the acknowledgement requirement, no one would ever trust him not to add it back in again, or do something else equally pigheaded. The only question now is how long the body that used to house the soul of XFree86 will be thrashing around in its death throes.

[Roguelazer]

sgtrock just managed to perfectly summarize and explain the entire issue. /me bows.

[Robe]

sgtrock just managed to perfectly summarize and explain the entire issue. /me bows.

LOL.. you posted that before I could. In complete agreement. Thank and well done sgtrock .

When I started this thread a few days ago I was hoping that a post like that would occur. Now I too completely understand.

[Voltago]

Is there any official reaction of the XFree project to the announcement of several distros to discontinue the distribution of XFree86?
Is this Dawes guy friggin' mad?

[Roguelazer]

More food for thought:
FAQ

To summarize:

What about GPL-compatibility?

The 1.1 license is not GPL-compatible. To avoid new issues with application programs that may be licensed under the GPL, the 1.1 licence is not being applied to client side libraries.

Voltago: To your first question, no. To your second question, maybe.


Oh, and a totally different angle about this new license. Now all the other systems that make use of X drivers (the fbdev drivers, for instance), can't! Isn't that great?

Hi David !

I'm no license/legal expert, but do that mean the licence becomes GPL
incompatible ? In that case, that basically means you are screwing up
any effort to make a decent graphics driver model in the linux kernel.

If I'm wrong on that, you can skip the rest of this email.

We rely heavily on the XFree drivers as those are updated & maintained
by the card vendors or people working for them ("nv" by Mark Vojkovich,
"radeon" by Hui Yu).

Losing the ability of porting code straight from these to the fbdev
drivers will basically kill all my efforts to turn the kernel radeonfb
into a decent driver as I need to be able to re-use the code ATI puts
in the XFree version. I suppose the same will happen to linux rivafb.

So right when we are considering a new & saner video driver model for
Linux, you are doing this move which screws up by blocking us from our
primary source of information & support from the gfx card vendors.

Ben.

[Beholders_Eye]

I was wondering, is there some material about GPL 3.0? I would like to know how it's advancing, and how it will deal with PHP Licensing scheme, as well as the new Apache license.

[Ian Goldby]

sgtrock:

You've summarised well the issues, but I think you misunderstood my position. After a bit of misinformed wibbling,

Ok, I get it now.

Clause 6 in the GPL effectively says that the GPL is withdrawn if you impose any restriction on the distribution of a program that goes beyond what the GPL itself stipulates.

The acknowledgement clause in the Xfree license could be seen as such a restriction.

So I understand why the xfree 1.1 license could be seen as a problem. However, as Roguelazer just pointed out,

To avoid new issues with application programs that may be licensed under the GPL, the 1.1 licence is not being applied to client side libraries.

Now this may be my misinformed wibbling again, but doesn't this eliminate the problem? You can link against xfree libraries without incuring the acknowledgement requirement (or any other requirement beyond what the GPL itself stipulates) so there is no conflict with the GPL. To be fair to the distros, I suspect that particular answer in the FAQ appeared only after the initial fuss. But I think it is time now just to get on with what they do best and lay this controversy down.
_____________

I would stop there, but there are a couple of other points you raised that I want to answer. First, the license change is not being applied to any code where xfree have not got the permission of the contributer. Whatever you might think, David Dawes hasn't tried to make any unilateral change. Unless Alan decides to withdraw his drivers, they will still be included under the original license.

Second, the situation with xfree is different to the so called 'obnoxious advertising clause' that was removed from the BSD license. The xfree clause is far less onerous, as it only requires inserting "This product includes software developed by The XFree86 Project, Inc (http://www.xfree86.org/) and its contributors" into a readme file (or man page etc). I see you understand that. So why do you raise the problem of 'hundreds of contributers and organisations' requiring acknowledgements? It sounds like a 'thin end of the wedge' argument. But even if it were true, I don't think it would be very onerous for distros to include a file of acknowledgements. It's not the same as having to put the list on all boxes, posters, mugs, etc, is it?

Third, I have no problem with authors being free to choose whatever license and whatever restrictions they want to distribute their code under. As authors and copyright holders, that is their right. What I object to is the FSF telling us that we should use the GPL. I don't see any such advocacy of the other popular Open Source licenses. They also crafted it in such a way as to make it incompatible in a fairly fundamental way with anyone who wants to impose their own restrictions on distribution of their code. It effectively places an artificial barrier between developers who feel they want to impose some added restriction, and developers who are happy with the GPL. Well, let them. I'll continue thinking they are arrogant. Others will continue thinking they are driving a great social reform. Xfree have at least taken a pragmatic approach by making sure their libraries are not incompatible with the GPL.

[Roguelazer]

Yes, most of our software links to the client libs. But that doesn't do anything for drivers, or anything that directly links to the server....

[sgtrock]

Roguelazer wrote:
To avoid new issues with application programs that may be licensed under the GPL, the 1.1 licence is not being applied to client side libraries.

Ian Goldby wrote:
Now this may be my misinformed wibbling again, but doesn't this eliminate the problem? You can link against xfree libraries without incuring the acknowledgement requirement (or any other requirement beyond what the GPL itself stipulates) so there is no conflict with the GPL.

I hadn't heard that, so it may be new. I know I haven't heard any reaction to it from the legal people involved with the various affected parties. This may resolve some of the issues. I honestly don't know.

I have to wonder, however, whether Dave Dawes may have burned up all of his karma with a lot of people, though. It'll probably take quite some time before people quit thinking about this whole mess.

Ian Goldby wrote:

Second, the situation with xfree is different to the so called 'obnoxious advertising clause' that was removed from the BSD license. The xfree clause is far less onerous, as it only requires inserting "This product includes software developed by The XFree86 Project, Inc (http://www.xfree86.org/) and its contributors" into a readme file (or man page etc). I see you understand that. So why do you raise the problem of 'hundreds of contributers and organisations' requiring acknowledgements? It sounds like a 'thin end of the wedge' argument. But even if it were true, I don't think it would be very onerous for distros to include a file of acknowledgements. It's not the same as having to put the list on all boxes, posters, mugs, etc, is it?

The changes to the license have been posted twice to this thread, so I won't tie up disk space doing it again. I would suggest that you go back and re-read it carefully, though. Conditions 2 and 4 create a real problem for distros. Taken together, they mean that a distro can't create a boxed set and mention XFree86 without putting the entire notice on the box or getting specific permission from the XFree86 project. It also means that they can't mention XFree86 on their Web site without doing the same thing.

This is a classic 'thin edge of the wedge' problem. Suppose for the sake of argument that every project made similar demands. Gentoo boasts something like 5,000 apps in its distro. I think some of the mainstream distros like Debian and Red Hat boast something north of 10,000 apps. Care to tell me how a distro is going to cram all those acknowledgements on to a box? Or alternatively, get specific permission from every single app project (some of which probably have dead email addresses for contact info)? Alternatively, how on earth are you going to create a Web page of marketing fluff without going through all the same hassles?

Heck, if even 10% of the apps out there made similar demands the distros would probably have an insurmountable obstacle. After all, would you like to track down all the copyright holders who have contributed to 500 or 1,000 apps?

Ian Goldby wrote:
What I object to is the FSF telling us that we should use the GPL. I don't see any such advocacy of the other popular Open Source licenses. They also crafted it in such a way as to make it incompatible in a fairly fundamental way with anyone who wants to impose their own restrictions on distribution of their code. It effectively places an artificial barrier between developers who feel they want to impose some added restriction, and developers who are happy with the GPL.

Well, that's because the FSF has never seen itself as pushing an Open Source license. They see themselves as pushing a FREE (as in speech) software license. Your analysis of the effects is correct, btw. They are there by design because the people who crafted the GPL and the LGPL didn't want anyone taking their work and locking it up in closed source apps. Since those licenses were created, many other organizations with similar goals have adopted those licenses for the same reason, rather than try to craft their own.

One side effect is that it limits some OSS apps from linking to GPL work. As long as everyone understands that and plays by those rules, no problem. (Witness the whole side discussion about Apache)

The GPL has been pushed hard by the FSF because they are evangelists. Yes, their attitude can approach religious fervor. I don't think that even RMS can be regarded as a fanatic, though. If they denounced all other licenses and refused to allow their stuff to be distributed in association with apps that use those other licenses, then I think we could call them fanatics. Heck, just the opposite seems to be true. The LGPL was created in response to some criticisms of the GPL, after all. As it is, I regard them more as a somewhat dotty but absolutely necessary bunch pushing a pair of absolutely necessary licensing options.

Personally, I'm all for the GPL. And the BSD license, the MIT license, and the other certified OSS licenses. I'm also all for all the licenses that call themselves OSS but don't have that blessing, as well as the most draconian EULAs on the planet.

More choice doesn't just mean that the developers have the freedom to choose how to release their software. It also means that those of us who use this stuff can become informed about what works for us. WE get to choose which software we want to use.

Yep, the broadest possible spectrum of licenses is exactly what we need.

[Ian Goldby]

Yep, the broadest possible spectrum of licenses is exactly what we need.

We agree on something then

I also see what you mean about the new clause 4 in the license. Do they really intend this to stop distros putting "Contains XFree86" on the box without specific written permission? I completely missed that.

I think you've misread about 'cramming acknowledgements onto a box though'. The new license definitely says nothing about that. It is quite specific that you only have to acknowledge them in the same form as you would acknowledge any other 3rd party components. Quite why anyone would not wish to acknowledge all the hard work of the XFree developers is a mystery to me.

[sgtrock]

Ian Goldby wrote:
I also see what you mean about the new clause 4 in the license. Do they really intend this to stop distros putting "Contains XFree86" on the box without specific written permission? I completely missed that.

That is exactly the end result. Whether or not that's what they meant is an important question. Since they have refused to change the wording after several people have pointed this fact out to them, I think we can safely say that is Dawes' intent now.

Ian Goldby wrote:
It is quite specific that you only have to acknowledge them in the same form as you would acknowledge any other 3rd party components.

That is the issue in a nutshell. That's where the potential for every word in every copyright notice to get crammed on to a box becomes a possibility. Either that, or the need to contact each and every project.

Ian Goldby wrote:
Quite why anyone would not wish to acknowledge all the hard work of the XFree developers is a mystery to me.

Set aside for the moment that the XFree86 project is a shadow of its former self since most of the talent have left for various reasons. It's not that people don't want to acknowledge all the work of the XFree developers. It's just that providing that acknowledgement the way that Dave Dawes wants becomes quickly unworkable when it's expanded to cover every other project.

You can't even really arbitrarily draw a line and say that you're only going to provide separate acknowledgements for a relatively short list of, for example, Linux or BSD, XFree86, KDE, GNU, and Gnome. Every project has the right to determine their own licensing, after all.

Let's get back to worst case that everyone has the same clauses as the new XFree86 license and look at what it means from a practical standpoint. Let's assume that everyone, including the Linux kernel developers, agrees to let you advertise their project. So, you can't say you've got a Gentoo Linux distribution. Nope, you have to say you've got a Gentoo Linux-XFree86-KDE-GNU-Gnome-Apache-Webmin-Samba-xine-OpenOffice- ... etc. ad nauseum 5,000 times.

Bad, right? Well, it gets worse.

Now let's say for the sake of argument that Dave Dawes decides he doesn't like the background color that you chose for your boxed set, and refuses to give you his written permission to use the XFree86 logo. According to the XFree86 license, you can still use their software, but now you have to include their license verbiage instead. Let's assume further that just a dozen more application projects make the same decision for any reason whatsoever. Or, that you can't find the right people to grant permission for that dozen projects. Now you've got 4,987 names for your Gentoo distribution and 13 full copyright clauses, all crammed somehow on to your box.

Where on earth are you going to find room to print all of that stuff and still keep it legible? You'd have to provide a high powered microscope for people to read it!

Nope, at this point in time I think that probably the best possible solution for all concerned is to fork the XFree86 project based upon the last codeset before the license change. The next best solution is probably the XFree Desktop project.

[cyfred]

I in no way defend the new license change, I actually think its stupid. Personally I think the general idea behind requiring each dependent package (anything that builds off XFree code under the new license -- which will eventually include parts of Xlib) to give credit really would only work if every X based application made changes to their own repositories that satisfied the license.

Its similar to employment when and employee (XFree) says to the Boss (Open Source)

you need me, I dont need you

Ive made contributions of XFree before, personally I dont care about the code thats there -- I contributed it for the benifet of the colective Open Source situation. Heck we all want a GREAT desktop. That said please read my first line again.

XFree totally asside, its time to move on people, the new license is / will cause more problems than effort is available to resolve. Honestly I reckon X11 itself is a dead duck, but do realise that everything is still tied to Xlibs. I believe a more concerted "Linux Desktop Server" project is required that (along the lines of y-windows (www.y-windows.org) provides server side widgets. This really would do leaps and bounds for both WM design/implementation/configuration and Theming.

FD.o is good aswell, but its still based around X11 which in its entirety I tolerate only marginally, but there really isnt a very viable replacement -- now there is a reason and I cant see why work from fd.o and Y couldnt be brought together to provide a new system. Yes yes I know thats a lot of work

[dmmgentoo]

I'm wondering also, what does a change in terminology have to do with suddenly every distro dropping XFree? Hell with it, I'll make some unofficial ebuilds for the 4.4 series I guess. It would be great if the xserver and xlibs projects were at a point that we could replace XFree entirely, but that's not coming for a while yet. Get to 'er, until then I'll stick with what's available, even if some people have a hairy fit over some stupid wording in the license.

I think we've got the GPL hype to blame. If something isn't compatible with the GPL, it's evil. What the hell? Who says everything on a Linux system has to be 100% GPL or GPL compatible? Even the Linux kernel itself isn't 100% "pure" GPL, because it's got some exception clauses, dealing with distribution of closed-source binaries or some such.

[Chaosite]

I'm wondering also, what does a change in terminology have to do with suddenly every distro dropping XFree? Hell with it, I'll make some unofficial ebuilds for the 4.4 series I guess. It would be great if the xserver and xlibs projects were at a point that we could replace XFree entirely, but that's not coming for a while yet. Get to 'er, until then I'll stick with what's available, even if some people have a hairy fit over some stupid wording in the license.

I think we've got the GPL hype to blame. If something isn't compatible with the GPL, it's evil. What the hell? Who says everything on a Linux system has to be 100% GPL or GPL compatible?

Um, the GPL?

It says that all software that links to the GPL must be GPL'ed.

But to get a fuller understanding, please read this thread. People who understand this issue way better than me have exlained it perfectly already.

But if you don't want to use the GPL on your code, don't. Use some other licesnse. Just as you can not use the GPL, we can use it.

[dmmgentoo]

According to my limited understanding, GPL software is not allowed to advertise or promote in any way non-GPLed software. That's one of Stallman's core tenets - that all software should be free (read GPLed) and no one should be encouraged to use (or even told about) non-GPLed software.

Presumably this is why they are so unhappy about xfree's new licensing condition - that you have to ackowledge the authors in a reasonably prominent place, which in their view amounts to promoting software that doesn't use their GPL.

I personally think the GPL should change. There should be credit where credit is due.

Edit: Sorry, AFAIK, the GPL itself doesn't ban promotion of non-GPL software, it is the GNU philosophy - so the GPL itself probably doesn't need changing. I haven't read the GPL recently though.

Or am I barking up the wrong tree? Is there another problem?

The problem is that we've got two guys with egos, i.e., Richard Stallman AND David Dawes. I hope some middle ground can be reached. The problems stem from the fact that the XFree license emphasizes the author recognition, whereas the GPL emphasizes the overall contribution of all contributors to the code. Hence, in the GPL case, the authors tend to become more anonymous, and the code itself is more important than the author recognition.

[dmmgentoo]

but i think it comes down to the code itself. how much is really needed? what can be thrown away? so on and so forth. in the scheme of things X is a bloated pig, not very fast.. how many of you remember Linus pitching a fit about having to renice the X process so it ran well?

But who's to say any other graphics/windowing system wouldn't have the same problems under Linux? The fact is, with the 2.6 kernel series, the scheduler became much better, and various improvements to the scheduler and VM subsystem have been made such that X is pretty much transparent and responsive without any renicing on the part of the user. I remember XFree86 on FreeBSD always ran very well, BUT, I do believe the FreeBSD kernel team sneaked in some scheduler tweaks to make X run better. So, you could be correct.

[Ian Goldby]

Unfortunately, it's fairly clear that some (many?) people want to see the XFree86 project die, and this licensing issue is a convenient stick to beat it with.

Never mind that they have clarified that the client-side libraries are still as GPL-compatible as ever. Never mind that the new license only applies to code where the contributer has explicitly given his permission. Never mind that they have clarified exactly what they mean by "same form and place as other third-party acknowledgements". Never mind that if you don't acknowledge any other 3rd parties, you don't need to acknowledge XFree86. (True, by the way.)

I've had about enough of this FUD. Just read the FAQ, people, please!

I don't see why it should be my job to be apologist for XFree86, but I'm just a little fed up with the current hysteria and misinformation.

[zOOz]

imo it's wrong to authorise xfree86. it is open source and it have to be so kind of soft. ;-(

[KAMIKAZE_]

www.y-windows.org go go go...

[NuclearFusi0n]

www.y-windows.org go go go...

http://freedesktop.org/software/xserver go go go...

[deepwave]

All I hope is that at the end of all this, there will still be some fairly stable, and modern GUI system. I really don't care who makes it, so long as it runs well and that the graphics hardware vendors will have decent drivers for it.