David Dawes seems a little suspicious of the underwhelming support for XFree86 by Linux distributors: “I have heard privately that some vendors were planning to move to an X.Org release even before this licence issue came up. That probably makes business sense for the vendors given that X.Org is a vendor-oriented organization sponsored by hardware and software companies, while XFree86 is an independent group of volunteer developers. I suspect that the licence issue may have affected the timing, but not the end result“, he concluded.
XFree86 Gets Underwhelmed by Linux Distro Support
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I had the bad feeling that this kind of destructive ignorance isn’t very helpfull for open source software generally.
Why would Redhat, Mandrake, Suse care whether the X server is sponsored by hardware and software companies or not?
If it’s taken David Dawes this long to become suspicious about people wanting to shift away from the XFree86 codebase, then he really hasn’t been paying attention to what’s going on. XFree’s core developer team overlaps very little with the people involved in the “Linux desktop” push, and their aims are not entirely aligned. Given the choice between code of similar quality provided by XFree or freedesktop.org/X.org, then it should surprise nobody that the distributions are going to ship code that’s provided by the team who work more closely with the major desktop environments. Closer communication between the distributions, the desktop environments and the X server benefits everyone with any desire to see Linux becoming mainstream. The license change emphasised that XFree weren’t interested in integrating themselves into that axis, and effectively confirmed that a break was more sensible – up until that point, there was still the possibility that XFree’s prevailing opinion would change.
Forks duplicate work, and they’re not something that people undertake lightly. But XFree’s license change confirmed that XFree wasn’t going in the same direction as everyone else was, and so the results should be unsurprising.
He needs to reconnect with reality.
They changed the license to make sure that the forks will not use code that was added to stock XFree86 after the fork happened IMHO.
That is the only logical explanation I could find for the change. The fact the David Dawes is so adamant about it implies that the license change means a lot to him. He claims that the change was a no big-deal change but I think it was to him.
Yeah, Dawes is out of it.
His team screwed up, and don’t know what to do now. They can’t very well go back on the license change. You take your ball and go home, then get upset when nobody plays with you anymore…..
Well, he’s still got that entreanched desktop user base in NetBSD and FreeBSD, but those will last only until the competitors are better.
Hi
None of hte Linux distros will ship it for legal reasons. OpenBSD didnt accept the license for pragmatic reasons. netbsd is more a server release attractive to the embedded market. FreeBSD is the only one remaining that could accept the new release.
freedesktop.org has two x servers now. the kdrive release with several important enhacements and adoption of xfree86 before the license change. Thats where active development will continue. David should have been very careful about a license change even when its not harmful
regards
Jess
I find it a bit ironic that SUSE is claiming to reject the latest incarnation of X based on the license changes, considering the terms of the YaST license, hardly GPL-worthy itself I would think… BTW, running 4.4 here on a RH box, not bad, seems to work nicely.
Well, he’s still got that entreanched desktop user base in NetBSD and FreeBSD, but those will last only until the competitors are better.
They cannot use this as well, since they cannot run GNOME/KDE/XFCE/ROX/etc… legally on this XFree version due to it being GPL incompatible. This is because the XFree license is more restrictive with the advertisment clause.
PLEASE dont let this turn into weather GPL is good or not flameware. The GNOME/KDE/XFCE/ROX/etc… developers decided that that is what they want to use, if you have problems take it up with them, and the other majority of the OSS community
Nah, it is not legal reasons, but the fact they are to lazy to include any acknowledgements if they choose to mention it…
Dude, go read the GPL. There is nothing preventing GPL code from linking to non-gpl code.
The problem exists in that some ppl are spreading FUD about the GPL and other licenses.
I find it a bit ironic that SUSE is claiming to reject the latest incarnation of X based on the license changes, considering the terms of the YaST license, hardly GPL-worthy itself I would think
I don’t think that they have moral objections (successful companies seldom have), but rather pragmatically ones. The xlibs have to be linked with GPL’d code, YaST on the other hand can coexist with GPL’d software.
“BTW, running 4.4 here on a RH box, not bad, seems to work nicely.”
Yes, but can you tell the difference between 4.4 and 4.3?
This is my beef.. improving hardware compatability and stability are good and very important of course, but the lack of feature growth has led me to believe that XFree86 development is stagnate.
I am all for a fork, assuming of course the new development model will be more open and ‘free-er’. Maybe then X can become a truly modern windowing system.
I really hope developers will continue to work and have a fork start off 4.4rc2 (which is the last release before license changes).
For CJK Asian font support, the merge of freetype and xtt backend is important, and of course the first official support of Big5-HKSCS TTF in XFLD font rendering (which shows up in Mozilla 1.6):
http://www.xfree86.org/4.4.0/RELNOTES2.html#12
There is nothing preventing GPL code from linking to non-gpl code.
…unless something in the non-GPL license prevents it. Many people think that this is the case with the new XFree86 license. It is a problem with this certain license, not with non-GPL code in general. No FUD here.
Hi
“Dude, go read the GPL. There is nothing preventing GPL code from linking to non-gpl code.
The problem exists in that some ppl are spreading FUD about the GPL and other licenses.
”
go read the xfree86 faq. it clearly says its incompatible. want to know why?
read this
http://lists.debian.org/debian-legal/2004/debian-legal-200401/msg00…
follow it up.
regards
Jess
No it is FUD. The license change was a clairification of the old one. There is nothing saying you can not link against it in the license. Check for your self.
http://xfree86.org/legal/licenses.html
Thanks, but I’ve all ready read it and it does not change that fact that this anti X kick is FUD.
Nothing in the licenses prevents it. The incompatibility exists in the ideology, not in that they can not coexist.
Hi
“What about GPL-compatibility? Is this license GPL compatible?
No.
The 1.1 license is not GPL-compatible which means that the two programs cannot be combined into a larger work unless the authors of both works agree to this arrangement. That does not mean it cannot happen, it just means all parties must agree to it happening. We refer you to the GPL Faq for an in depth discussion of this issue.”
so would you mind reading the link i passed to you earlier?
regards
Jess
Basically, there are two reasons given for the fork, and for not using 4.4.0:
1. The slow nature of XFree in developing and distributing new features. (I am all for a fork under these conditions.)
2. The new license. From everything that I have read and heard, the GPL does not forbid using XFree under its new license, but that it would be in bad taste to do so. The only practical reason for Distributions (other than Debian) not moving to 4.4.0 is that they don’t want to ‘disturb’ the moral loudmouths that would complain about using it.
Out of curiosity, isn’t there BSD code located in the kernel? Which, of course, would require an ‘advertising clause’? Does this then require a fork of the kernel, one GNU/GPL, and one just GPL? (How much functionallity would be lost?)
Read it… contians nothing any more useful than what I posted, btw I did post a link to the exact page you are quoteing…
I personally rack it up as FUD and reactionary.
Hi
”
Out of curiosity, isn’t there BSD code located in the kernel? Which, of course, would require an ‘advertising clause’? Does this then require a fork of the kernel, one GNU/GPL, and one just GPL? (How much functionallity would be lost?)”
BSD code is there in the kernel but it uses the new revised BSD code which doesnt contain the ad claus and is compatible with gpl. gpl incompatibility isnt about bad taste. gpl doesnt allow additional restrictions on licenses linked to it. it can be gpl,lgpl, revised bsd, mit,python or the old x11 license.
regards
Jess
Hi
“Read it… contians nothing any more useful than what I posted, btw I did post a link to the exact page you are quoteing…
I personally rack it up as FUD and reactionary.”
you didnt even bother reading. xfree86 accepts that it is incompatible. it is incompatible due to the adversiting claus. david of xfree86 and rms of fsf has accepted this?. you know more than them?
what the heck?
Jess
There is probably BSD code in the kernel, but it would be licensed under the new BSD license. It is perfectly okay to include code licensed under the new BSD license, because it does not have an advertising clause.
There is nothing saying you can not link against it in the license. Check for your self.
The new license contains an advertising clause which the FSF considers incompatible with the GPL. The license of a dervative of a GPL’d program must not impose additional requirements for redistribution – but this is the case with the advertising clause. You can’t link XFree86 with GPL software without violating one of the licenses.
@ Anonymous (IP: 61.95.184.—) – Posted on 2004-03-09 20:36:17
Right, and my reaction to that is it is FUD and reactionary. Like I said. And BTW go back and read what I bloody wrote and read what I linked too… it agrees with me in that…
1: There is nothing from preventing GPL program from making use of X.
BTW where the does it say in the GPL you can’t link to non-gpl code? Yes, this is what the FSF claims, but I have yet to see some one step up and point out where this exists in the GPL.
Yeah, but where does that exist in the GPL?
Hi
”
BTW where the does it say in the GPL you can’t link to non-gpl code? Yes, this is what the FSF claims, but I have yet to see some one step up and point out where this exists in the GPL.”
from 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.
an advertising claus restriction as imposed by the new xfree86 license is precisely incompatible because of the 6th claus is gpl which says that you are not allowed to impose any additional restrictions over that of gpl. you are free to create a more liberal license like the revised bsd. fsf doesnt claim that you cannot link to anything other than gpl.in fact they recognise and allow revised bsd,lgpl,gpl,mit,x11 and python license to be linked to gpl.
clear?
Jess
AFAIK this the new XFree86 1.1 license only covers distribution of XFree86 software and not linking…
The xfree86 team agreed to not license the xlibs using anything GPL-incompatible. So you should still be legally allowed to link stuff like gnome or kde to xfree86 xlibs.
A bigger problem is one of drivers. There are GPL drivers that many distributors add in to xfree, that they no longer can due to license changes. This code will never appear in xfree86 proper because the project would never accept it. I mean, even the DRI project had to practicly fight to get anything in.
umm… I think the synaptics driver is license incompatible, for one example. the wacom driver might be… I dont know if the 3dfx driver still links against glide. If so, that’s now impossible with the license change. I’m not familiar with the sis and savage drivers, but those might also be suspect.
BTW where the does it say in the GPL you can’t link to non-gpl code? Yes, this is what the FSF claims,
This is not what the FSF claims. You’re spreading FUD yourself. The FSF says that the GPL is specifically incompatible with the advertising clause, not with non-GPL licenses in general.
Point 6 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.
Point 3 of the new XFree86 license:
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.
You see the problem?
>I don’t think that they have moral objections (successful >companies seldom have), but rather pragmatically ones. The >xlibs have to be linked with GPL’d code, YaST on the other >hand can coexist with GPL’d software.
Which libraries do the xlibs require which are GPL (not
LGPL) and are not used by YaST?
Freedesktop.org is destined to become the default focal point for X. The people working at the various projects which are accessible under freedesktop.org understand so much that Xfree86 never did. This had to happen. It’s a good thing. Me, I’m most excited about two things – 1) the standalone X server and 2) Cairo. X is valuable to me because it’s a remote display system. As networked smart displays gradually replace networked PCs then the value of X as a transparent networking layer really begins to shine. The politics is really just an irrelevant sideshow at this point. And it is a good thing.
…is what Nvidia and ATI think about this whole thing. I, for one, could care less what RMS or distro vendors think about this, because if I have to I’ll just compile XFree86 myself. Can you intall the binary ATI and Nvidia drivers for X.org and have it work? If not, then any fork is basically useless until Nvidia and ATI are on board.
Right, that covers the program, not what it requires to run.
Yeah, they do not want to include to acknowledgement.
AFAIK the problem lies with including GPL modules for the server and not the libs…
BTW where the does it say in the GPL you can’t link to non-gpl code?
The issue isn’t linking to non-GPL code, the issue is linking to code that contains additional restrictions.
GPL Section 6: “You may not impose any further restrictions on the recipients’ exercise of the rights granted herein.” – http://www.gnu.org/licenses/gpl.html
If program A is GPLed, and program B allows you to reuse the source but places additional restrictions, you cannot combine them into program C without violating a clause of one of the two licenses.
Basically, you have to use the stricter of the two licenses, but the GPL doesn’t allow you to add restrictions, so you cannot comply with the second license unless it is less strict than the GPL.
This all assumes that you don’t own the GPL’ed program – if you’re the only copyright holder on A, you can link it with B, but you can only distribute C under B’s license, not under the GPL.
Right, but this applies to distributing of the XFree86 software…
“Version 1.1 of XFree86 Project Licence.
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.” – http://xfree86.org/legal/licenses.html
>Freedesktop.org is destined to become the default focal >point for X. The people working at the various projects >which are accessible under freedesktop.org understand so >much that Xfree86 never did. This …
Reading the mail archive at freedesktop.org today, I
noticed that one of the people who’s responsible for it
was recommending (along with 2 other options), copying
ongoing changes from XFree86, removing the license.
Odd, but no one objected to that.
Hi
Thats illegal. removing a license wont happen anytime
we may not agree with the license but we cannot ignore it
regards
Jess
I don’t know why someone marked this as abuse, it’s certianly not like he/she’s lying, XFree hasn’t done much in a long time when it comes to features, and if it continues this way, Linux will never succeed on the desktop because Joe average is looking for the goopy features like that found in OS X and the upcoming Longhorn. XFree has only recently with in the past 1 -2 years ago gotten half way decent font rendering, and then Keith Packard was pretty much left before he was going to get kicked out. A fork would be a good thing, and I’m glad there are a few of them.
Still on this…
The Xfree86 project, developers and founders, deicide to change their license to something they thought was needed.
After reading both clauses Old BSD licence (OBSDL) and the XFree1.1 Licence (XL1.1) and GPL for the matter. You see clearly that OBSDL is GPL incompatible and the new XL1.1 is not.
And is not why? because its clearly stated, and here comes the funny part, that “only” if you advertise others you should do the same to XFree86 project and contribs.
And whys that ? because of the total RIPoffs many linux developers did from BSD based licenses projects, even without the proper Crediting needed by law.
XFree86 got tired of this procedure (ask Linux devFB guys) and took actions to protect even more their code.
Now, here cames a question for all you,
Would be better if XFree86 picked up and closed their code with a propertary license ? or would be better to have a BSD based license that protects their interests? OBSDL is much more restrict than this one, and many pieces of code based on OBSDL exists in XFREe86, like the same way they exist on many other projects, FReebsd, openbsd and netbsd.
Or either would be better to keep the current trend and see their work, manly hardware drivers, beeing used by others without any ref or credit deserved to original authorS?
MAny contacted refer that GPL doesnt allow, or does not recommend prior acknowledge because those list would increase each time they pass from hands to hands. If you have no means to enforce this on court you are screwed and this was in my opinion a cleaver way to counter that.
Hail Xfree86 project!!
On the technical point you’ve raised, you are correct and most people are not responding to your very narrowly-defined point.
Yes, there is nothing in the GPL that prevents you from linking and running GPL-licensed code with non-GPL licensed code (the other license is another issue). The GPL does not cover use of such licensed code. Linking and running are considered usage of code.
However, the GPL does restrict distribution of the newly linked code. GPL-licensed code may only be distributed in binary form when it is linked to code that is GPL-compatible. The new X11 license is not GPL-compatible in that it explicitly defines a new (albeit, probably trivial) restriction. This makes the license explicitly non-GPL-compatible and code linked against new X11-license code cannot be distributed in binary form, linked against GPL-licensed code.
So, the end result is that you and I and everyone else can compile, link, and run XFree86 4.4 on any machine we desire, but we cannot legally distribute those binaries. In practice, noone will really care if I send those binaries to a few friends.
However, a responsible business or organization will not distribute these illegal binaries. Make no mistake, unless there is legal consensus that the new XFree86 license is legally compatible with the GPL, then such binaries are illegal to redistribute.
Its not FUD or nitpicking or anyone’s desire to not attribute credit where credit is due. Its a sticky little legal problem, nothing more, nothing less. Now, the panty-waving may be attributable that it is said (I’ve not personally confirmed this) that the new license only covers the XFree86 server, but the libraries are covered under the older license. So, the end effect is that most programs can still be legally distributed in binary form so long as they do not link directly into the server. If this is true, then there is some nitpicking going on about style and preference in using the new license, but nothing that prevents business-as-usual. However, such questions are really left to the lawyers and courts.
“indifferent to the business decisions of these vendors”
When these vendors basically account for 90% of the users on your product I would hope you are not indifferent to their decisions.
There is something seriously wrong with that logic. Sadly this flawed logic and obvious attitude will most likely mean the long term death of the project.
As far as the arguement that they were probably going to X.ORG solutions anyway……. yeah right. Thats even a tough sell to the village idiot when all of these vendors are currently running your product and have so much time invested in it with their own products. The fact alone that they are not immediatly replacing XFree with something else but rather sticking with older versions proves that this is nothing but a pile of stinking mass.
It is sad.
The license says “must include”, not “only…supposed”. This is an explicit restriction. I do agree that it is much more narrowly defined than the OBSDL advertising clause, but it still makes the license non-GPL-compatible.
Unless lawyers from both the FSF and whoever represents the interests of XFree86 (X.org ?) say otherwise, this is the truth.
The license says “must include”, not “only…supposed”. This is an explicit restriction.
thats the problem (it seems not your case in part) but i bet that 99% of the ppl bitching didnt read the license.
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.
its a condition not an implication
if you give credits to others then you must
if not then you dont need to include
imho if the authtors wants that way, its fair. And dont forget the new XFree86 1.1 DOES NOT apply to client side. Once more, no prob shipping KDE binarys with linux distros.
afaik, FSF haven’t made any kind of comment on the desirability or otherwise of the current situation or any kind of X server. They were asked to clarify whether, in their opinion as originators of the GPL, the new X license is GPL-compatible, and stated that they believe it is not. That’s all.
That’s all.
RMS gave his opinion not once but more than one time in the xfree ml
>RMS gave his opinion not once but more than one time in the >xfree ml
which one? (I didn’t see it).
http://xfree86.org/pipermail/forum/2004-February/thread.html#3974“&…
Richard Stallman [email protected]
Sat, 07 Feb 2004 04:01:10 -0500
The effect of the XFree86 license change is not as big as most people think. The new XFree86 license is GPL incompatible but it is only applied to the server code. The X libraries are staying under a GPL-compatible BSD license. This doesn’t make any practical difference for preventing GPL code being linked with the XFree86 server because the server alread contained code with the advertising clause BSD licenses.
The Xfree86 project, developers and founders, deicide to change their license to something they thought was needed.
And whys that ? because of the total RIPoffs many linux developers did from BSD based licenses projects, even without the proper Crediting needed by law.
Would be better if XFree86 picked up and closed their code with a propertary license ? or would be better to have a BSD based license that protects their interests?
How exactly has the hard work of the XFree86 developers gone unnoticed in the Linux community? Love it or curse it, we have used XFree86 knowingly for years now. XFree86 is synonymous with strengths and shortcomings of desktop Linux. If the desktop is slow we curse X. If we need to connect remotely from home, we bless those developers at XFree86. How many people start their day with Xinit?
Saying that the XFree86 developers aren’t getting the recognition they deserve is like saying that nobody knows who Michael Jackson is. Everybody knows of what the XFree86 developers do.
Now, if someone is stealing hardware drivers from the project and not giving proper credit, that is a problem. How exactly would adding another licensing clause rectify that if they are already ignoring proper use of the first license? It’s a non solution to the problem. If people were breaking through your window and stealing your stuff, would you glue your door shut? In this case, the XFree people glued their door shut to Debian and many other distros. It’s time they re-opened that door an apologized to everyone for the mistake.
What is this clause and how is it different to open source code that has the authors name in the source code.
Is it like the “about box”, where someone has to acknowledge that this desktop environment is KDE based. (FOR EXAMPLE, of course).
Reading the mail archive at freedesktop.org today, I
noticed that one of the people who’s responsible for it
was recommending (along with 2 other options), copying
ongoing changes from XFree86, removing the license.
Odd, but no one objected to that.
Oh, freedesktop’s X is GPLed. So they have to rip the MIT license off and MIT X license allowed that.
ongoing XFree86 only changes the core parts to the new license. Most other parts still remind MIT X.
Thanks for that link, it was a good read.
The way I take the GPL is that the “spirit” behind it is that an individual can download packages, change things, etc. to his needs. Then if that is the case, I find a “credit” notice a good thing… heck, if newbie comes along, installs Linux/BSD/WhatSoEver and see’s “XFree86 http://www.xfree86.org” or whatever credit notice, he then knows exactly what package he has, where he can go for support and updates. I think this is rather good to be honest, and I don’t know what all the crying is about.
But lets be honest here, Xfree86 is the backbone of any Linux distro now, Hardware vendors after a long battle are now writing drivers for it… it would be a REAL REAL BAD time to turn around to them and say “Oh, can you write drivers for this version of X please, we don’t like that one any more”… it doesn’t say much for the Linux argument at all, and really gives rise to doubt about the stability of Linux as a project as a whole to the outside world.
This is a terrible thing for Free Software, and I honestly can’t say I blame the XFree86 project for wanting credit where credit is due.
Dawes: See, it wasn’t me! It was vendors. Vendors, I tell you! Not my license decision! I’m vindicated!
“But lets be honest here, Xfree86 is the backbone of any Linux distro now, Hardware vendors after a long battle are now writing drivers for it… ”
I would think it more likely that hardware vendors write drivers for GNU/Linux.. Not specifically XFree86. They would probably continue to (and be more likely to) do so as the market share for Linux grows — no matter which version of X (or successor) they have to write the drivers for.
I do agree though that this whole thing is not good for the free software community, but its better to get it over with sooner rather than later. I am curious if the major distros are sitting back and waiting for something better to come forward or are activly seeking out and supporting an alternative
Well, he’s still got that entreanched desktop user base in NetBSD and FreeBSD, but those will last only until the competitors are better.
Incorrect, fd.o xserver has already been ported to FreeBSD, eventually NetBSD. The net result will eventually XFree86 being depreciated in the ports tree. XFrees demise was iminant.
“Linux will never succeed on the desktop because Joe average is looking for the goopy features like that found in OS X and the upcoming Longhorn. ”
Small nits. Since when has Joe “got my 90%” Average been exposed to the “goopy” features in OS X, or Longhorn?
Second Windows has succeeded on the desktop so far without those “goopy” features.
Maybe the better question is, will anyone continue to succeed on the desktop without these “goopy” features?
Interesting, according to that link, there is no license problem with using XFree86 in any distribution. And x.org’s XFree86 is GPLed? Is that really true?
It almost seems as though Linux distributions are punishing the XFree86 for not choosing or caring about the GPL for their core server code.
…is what Nvidia and ATI think about this whole thing. I, for one, could care less what RMS or distro vendors think about this, because if I have to I’ll just compile XFree86 myself. Can you intall the binary ATI and Nvidia drivers for X.org and have it work? If not, then any fork is basically useless until Nvidia and ATI are on board.
Well, I would say that Nvidia and ATI will be more than happy in knowing that they’ll finely get a XServer which will allow them to write drivers in which the X server exploits the full features thare are available, also, if it means that the new X server can drop alot of the old crusty parts and make drivers even easier to write, vendors will be encouraged even more.
Oh, freedesktop’s X is GPLed. So they have to rip the MIT license off and MIT X license allowed that.
I just checked this actually, and neither of the X Server (new server) or XFree86 (old/current X) seems to be GPL’d, nor was there ever talk on the lists of making them so. I don’t know what you are talking about.
At any rate, it’s still strange that no distribution will include XFree86 4.4 considering there seem to be no real licensing issues (the *client* libraries are in fact fully GPL-compatible). And to all those bashing Mr. Dawes, judging from the mailing lists, he seems to be a fine guy. He even decided not to apply the new license to the client libraries so that it remains GPL-compatible.
Hi
How many times you say it the x server at freedesktop.org is not gpl’ed and its not just client libraries. we have gpl’ed drivers in there which are incompatible now.
talk about stuff you know or atleast take the time to research it
Jess
Hi
” I would also like to suggest that consideration should be given to
allowing the notion of “attribution equivalency” to fall within the
scope of GPL compatibility.
That requirement causes practical difficulties, because a person
making a change in a package–adding in a file that has this
requirement–may not know which other places such attributions must be
added to. A person in learning enough about the code to make this
change would not have had to study everything in the package that
might contain statements giving credit to someone or other.
So he would not find it easy to follow this requirement.
Thus, although that license requirement qualifies as free software,
I think we should try to discourage it.”
see how reasonable that is?
regards
Jess
I’m lovin it.
B.Gates
The new X server is *not* GPL’ed. Its the old MIT X license.
I like that bit most :
————————————————-
He continued by saying: “I think this provides a good opportunity for XFree86 to cut out the middle-man, and go back to a stronger emphasis on providing our software directly to the end user as we did in the early days.”
————————————————–
As if XFree86 had ever been a simple package to compile, install and configure !
Most complex linux software is a bitch to install and maintain from source hence the middle man. This guy needs to get out more or start doing rpms for and deb packages for a living. Maybe that the plan ! lol
First: FreeBSD seems to keep close watch on freedesktop.org developments, as their inclusion of various fd.o extensions suggest, like composite for instance (http://www.freshports.org/x11/compositeext/):
“New port: Prerelease version of compositeext from freedesktop.org:
X Composite extension headers and specification
Testing is encouraged, but please do not use these ports as dependencies until
they are updated to release tarballs and the XFree86 ports have been updated to
depend on them.”
On the other hand, even RMS’s link (thanks for the link) suggests that new licence seems to be ok. The problem is with the word: it seems. I don’t think there is problem with distributing the new 4.4 XFree86 in binary form on the install CD. The documentation (The FreeBSD Handbook) is already in compliance.
As to running KDE on the new XFree? Do we have to modify the sourcecode of KDE to link with the new xlibs … oh pardon, that’s still GPL compatible.. well the server? If not, then that’s mere aggregation, which the GPL permits. (Otherwise you won’t be able to run mplayer on windows via Cygwin, and whole lot of other programs). So before saying things like FreeBSD won’t be able to run KDE because of new XFree86, get a clue please.
I read the GPL back and forth, and I see no problems in distributing XFree86 and KDE on the same media (like the first CD of FreeBSD): SuSE distributes YAST on the same medium it distributes GPL software.
” If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.”
AFAIK, KDE and X and FreeBSD can be considered as separate works. X is not based on KDE, nor the other way around. Linking (ldd) happens when you install these programs after one another. The GPL allows that. What it would forbid is to distribute complete images of a BSD installation that includes GPL code (KDE) linked to potentially GPL-incompatible libraries. What many of you seem to forget is that GPL’s strongest point is forbidding distribution of such programs. Hollywood uses heavily modified GPL programs, and you’ll never see the source for those modifications. But they don’t distribute the modified work, they use it in-house, which the GPL permits.
>And to all those bashing Mr. Dawes, judging from the mailing
>lists, he seems to be a fine guy. He even decided not to
>apply the new license to the client libraries so that it
>remains GPL-compatible.
The license change was done so that most forks would not be able to take advantage of new XFree86 code. Dawes was well aware that the X.org folks (and many others) would not be able to accept a new license with GPL incompatibility for political reasons, regardless of the fact that it makes little practical difference. (Neither the X.org or fd.org/kdrive projects are being released under GPL.)
This was no surprise to Dawes; it was, in my opinion, a very purposeful “scorched earth” policy, ensuring that new XFree86 code would never migrate to most possible forks, in direct defiance of the spirit of Open Source and Free Software. And so XFree86 will fade into irrelevancy, as both developers and vendors migrate to greener (read: friendlier) pastures.
It’s all a horrible shame, of course, when you remember that all of this could’ve been avoided if XFree86 had simply opened up their development process earlier in the game. Before they had managed to piss off just about anyone remotely interested in contributing to XFree86, including virtually all new blood. Terrible shame. Things could’ve been different, and while I certainly don’t blame Dawes personally for this failure, the XFree86 culture and baggage have not served that project well. It’s not like no one saw this coming; the high-intensity hub-hub surrounding XFree86’s shortcomings as a project has been building for well over a year, and there were many opportunities to move forward without forcing forks or increasing the bad blood out there (broadening CVS access to a more reasonable level, for example; broadening the core team, etc.).
But it was not to be. Alas.