Slashdot Mirror
XFree86 4.4: List of Rejecting Distributors Grows
Bootsy Collins writes "Yesterday, we
discussed
Mandrake's
decision to
revert their release-in-development from XFree86 version 4.4 back to version 4.3 because of issues with the
new XFree86 license.
To update this, the list of OS distributors opting out of
XF86 Version 4.4, and future releases, based on licensing concerns continues to grow.
While Fedora seems to be
"preparing to support multiple X11 implementations",
Red Hat has explicitly stated
that they have no plans to ship XFree86 v4.4
under its current license. Also add to the growing list
list
Debian,
Gentoo,
and OpenBSD."
On its face, the new license seems both reasonable and fair -- however, it also seems to create a lot of questions regarding how it should be interpreted and this is causing all the noise. My guess (and sincere hope) is that a clarification from, and possible minor re-write of the license by the XFree Project, Inc will clear this all up.
Money for nothing, pix for free
Short answer, yes, for binary distribution it is that bad. For more than you want to know ..
except if you notice that openbsd and gentoo, both source only compiles won't include it. The problem is the licence is simlar to bsd's but requires extra credit to be placed all over the distro. Anything that links to the xfree code that is gpled or uses a bsd licence is breaking the gpl/bsd licence because of the extra stuff they need to do in order to link to the libarys. In order to make a distro with 4.4 included, you would need to rewrite every app that links to X to the X licence, else you are breaking the gpl and as a distro maintainer, you'd be responcable for the breaks.
The xfree86 V4.4 license adds
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.
vs.
http://www.gnu.org/philosophy/bsd.html
No, the code in question was sumbitted by its authors to XFree86 under the XFree licence. Remember, the author owns the code and he can licence it as many times as he wishes, every time with different licence.
So it doesn't make XFree86 ``gpl derivative''.
But those people who sumbitted those patches oppose changing the XFree licence on their code to something GPL incompatible. At least I've heard that Alan opposes, but I don't believe they asked all contributors if they agree to licence change.
Robert
Bastard Operator From 193.219.28.162
Answer : Yes
More specifically: it is incompatible with clause 6 of the GPL, part of which reads:
You may not impose any further restrictions on the recipients' exercise of the rights granted herein.
Since their advertising clause is "imposing further restrictions" it is incompatible with the GPL.
Even more, Clause 2 of the GPL itself forbids redistribution of GPL'd code as part of a "Program" which also contains parts with licenses incompatible with the GPL.
Since parts of the current XFree86 are actually licensed under the GPL, the Xfree86 guys have two options
a) remove all GPL-licensed code from XFree86
b) get approval from all authors of GPL-licensed code in XFree86 for a re-licensing of their code.
Considering that Alan Cox has already clearly indicated he will not accept relicensing of his code under something other than the GPL, legally the FXree86 people are already obliged to remove all Alan Cox's code from their relicensed XFree86 before distributing it...
The GPL isn't viral, if they release something which contains parts that are GPL'd, and other parts that are incompatible, those incompatible parts don't become GPL, but they do have to either remove the GPL'd parts, or relicense the incompatible parts under a different, compatible license (which may or may not be the GPL itself).
If they don't, they open themselves up for a lawsuit from the copyright-holders of the GPL'd code (which will probably be the FSF in many cases).
And this is why the Distro's don't want to touch XFree86 4.4, as soon as they distribute it, they themselves are doing the same, illegal, thing that XFree86 itself would be doing, and they would be open to lawsuits...
``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.''
Yes, if you have end-user documentation it asks you to have a tiny line about how there's code from the XFree86 Project.
So where you get the idea from that it should be spit out during boot is beyond me. Yes, I know it can also be done in software. But that's done where normal attributions are normally kept, say an about box, or -EEK- perhaps /COPYRIGHT. And note that the software requirement is a MAY, not MUST.
Let me phrase a question back at you lot: "What is against giving credit where credit is due?" Because it looks like some common courtesy seems farfetched with a lot of people at the moment.
And by the way, it is similar to zlib's license, which is not mandatory, granted, but how many of you have actually credited Mark Adler and/or Jean-loup Gailly for their work?
Jeroen Ruigrok/Asmodai
The GPL, for instance, also has a "credit where credit's due" clause, but you'd have to actually go out of your way to modify an already GPL'd program to break it:
By comparison, the XFree86 license is requiring that redistributors of XFree86 modify their documentation and/or files containing credits. That can affect you even if all you plan to do is put a copy of XFree86 on a CD for someone else and include a note saying "Here's how to install it".I'm not 100% there's no workaround that satisfies both parties, and it's notable that XFree86 is a stand-alone package, so it can safely be distributed on the same CD as, say, GNOME (though it may be necessary to include an additional X server that's unambiguously GPL compatable in addition to XFree86 4.4) without breaking the licensing for GNOME.
In general, XFree86 has made a mistake in that most authors should try to use an existing license that's a known quantity and is as compatable with as many licenses as possible. The new license appears to break this principle.
You are not alone. This is not normal. None of this is normal.
For God's sake people, stop asking the same question over and over. Read this:
[forum] [XFree86] Announcement: Modification to the base XFree86(TM) license.
Sven Luther forum@xfree86.org
Sun, 1 Feb 2004 17:48:57 +0100
* Previous message: [forum] Re: Announcement: Modification to the base XFree86(TM) license.
* Next message: [forum] [XFree86] Announcement: Modification to the base XFree86(TM) license.
* Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Thu, Jan 29, 2004 at 11:58:38AM -0500, David Dawes wrote:
> Announcement: Modification to the base XFree86(TM) license.
Hello,
As discussed with David, i am taking discussion concerning the
problematics aspects of this licence change here. I think i understand
somewhat the reasons behind the licence change, but i wonder if all the
consequences of it have been thought of before doing the change.
Also, there are some confusing wording in one of the clause, which i
believe would best be clarified as to what the interpretations of them
by the XFree86 project are.
Also, first notice that my position is actually quite inconfortable,
since i am here mentioning the concerns of wider community and criticize
the new xfree86 licencing, in other forums, i usually do the opposite,
and take xfree86 side on this, so please do not react badly, and let's
have a rationale conversation about this, so that things can all be
resolved to everyone's satisfaction.
1) Possible confusion.
The following clause is the most problematic of all the licence, and as
such it would be nice to clarify it before starting a polemic about it.
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.
Ok, what does this mean exactly ? If there is a end-user documentation,
but it contains no third-party acknowledgement part, do you still have
to put the acknowledgement or not ? Also, is the choice between putting
the acknowledgement in the end-user documentation or the software a
choice that is free to make, or is the second an alternative only if
there is no enduser documentation. And what do you mean by in the
software itself ? If this software is a linux distribution for example,
would a file on the CD which is copied to the disk be enough ?
2) GPL incompatibility.
This selfsame clause is also the one which clashes with the clasue 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.
And in the 'you may not impose any further restrictions' part. Since the
GPL does not force you to add acknowledgement in the end-user
distribution, then the clause 3) of the 1.1 XFree86 licence is indeed a
further restriction, which cause an incompatibility with GPLed software.
Now this is again modulated with the exact interpretation that is given
in the above point.
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
IAAL(ayman)
From the new XFree license (emphasis added):
- Redistributions of source code must retain the above copyright notice, this list of conditions, and the following disclaimer.
- 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.
- 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.
- 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.
The GPL pretty much says that the only restrictions you can place on redistribution are the restrictions in the GPL. You can't add any other restrictions, however harmless they seem -- I guess the argument is that it's the start of a slippery slope, and IIRC some of the old UNIX systems had to have 3 screens worth of notices like that one.I guess I can see RMS's point: if you let XFree require this notice, maybe Apache will want a notice too. And maybe Wall would want a similar notice for Perl. OK, that's annoying but we could live with it, even if we end up back in the UNIX days of multi-screen credit notices.
But then suppose NVidia releases a driver and says that you have to include an advertisement (not just a credit) for NVidia if you distribute it with the kernel -- or rather, they specify a credit message that many people would consider an advertisement. Well, now we've clearly crossed a line most Linux developers don't want to cross, but it's not exactly clear where that line was in the scenario I just mentioned. NVidia would say "all the other developers got to come up with their own credit text, why can't we say what we want to? If you don't let us have our free speech the terrorists have already won!"
And they'd have a point: if you let some people dictate terms to the GPL you don't have much grounds to keep others from dictating terms, and however innocent the first terms may seem, somebody will find a way to screw it up.
So, you have to kick the camel in the nose while that's still all he has in your tent.
All's true that is mistrusted
It is only linking between software with these licenses that is the problem. With most of the licenses above there are ways of getting around it. The problem with the new xfree86 license is that you have gpled software such as gtk and qt that are going to be directly linking with xlib. These plus the various window managers and other applications that link with xlib and are gpled are probably 99% of the use of xlib. I havnt verified this, but Im willing to bet that only 1% of the software that currently links with xlib in debian could keep doing that with this new license. That is why all these distributions will not be shipping xfree86 4.4. They couldnt use it and it would be a waste of packaging.
"We Don't Need No Truthless Heros!" - Project 86