Slashdot Mirror
Mandrake Blocked By XFree86 4.4 License
Linzer writes "A mailing-list message posted by Mandrake Linux's main developer on the Cooker mailing-list states that the development version of the distro is about to revert from XFree86 4.4 to the 4.3 version because of XFree86's recent license change. Mandrake contributors have started asking for justifications from MdkSoft. Many point out features of XF86 4.4 [an 'an open source X11-based desktop infrastructure'] they can't live without, including support for some not so uncommon hardware.
A later Cooker mailing-list post extends a bit on the reasons."
As far as I can tell, all Mandrake would need to do is include the new text in with the rest of the copyright/liscense info and they'd be in compliance? Why is this a big deal? Or is there some subtle legal thing at work?
Too bad they didn't give freedesktop.org people a little more time to develop a viable alternative.
But your point is well taken.
It appears to my uneducated eye that this is a very slight modification which shouldn't make any difference to mandrake beyond the typical publication of copyright notices.
If Mandrake takes it seriously enough to revert to 4.3 I must be wrong? Anyone have an explanation?
Wouldn't a good solution to be what Mozilla did to ensure GPL compatibility? Cross-license XF86 under its own liberal license, the GPL, and the LGPL. This way, companies like mandrake could easily use it under an "approved" license, hassle free. -- What to keep away from dogs
But how is this license change is big problem?
#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.
From the looks of the problematic clauses, it seems that all that needs to be changed is some documentation.
EVERYDAY IS CATURDAY
Can someone explain why Mandrake is the only distro blocked now?
The Yasashii Syndicate ||
Trustworthy sources tell me that Red Hat, SuSE, and Debian are reacting similarly. The license change was announced as a fait accompli, and after being urged to reconsider, David Dawes went ahead with it any way.
This might be the sort of thing the freedesktop.org people are talking about when they say XFree86 (the project) doesn't have any accountability to the community. They seem to have a problem working cooperatively with others.
Freedesktop.org not only has a couple of big-name figures from the glory days of X involved (Jim Gettys and Keith Packard), but they also have actively involved various third parties and stakeholders in the X Window System technology -- not just the Linux distributions, but leading developers in GNOME, KDE, and Mozilla to name just a few, and some other people who were kicked out of the XFree86 project.
XFree86 does not seem to have been able to make the transition from the small hobbyist audience that it served in 1993. Maybe David Dawes and the few remaining participants in XFree86 will be happier producing a custom version of the X Window System for themselves and a tiny minority of others. Maybe they didn't lack the skills to be a large community project: just the motivation.
I think Stallman would remind he foresaw this situation many years ago:
The X Windows Trap
If people like you weren't so busy misrepresenting his views you'd see that.
an ill wind that blows no good
It is perfectly fine for all "free" software to require the attachment of the GNU license on everything yet it is highly offensive for X86Free to require "free" software to state that it came from them?
hmmmmmmm......
First, XFree is an open source, community driven project. Hence, in many distributors eyes is wasn't merely "someone else's graphic system". The real problem was that all of these distributors assumed that XFree would remain GPL-compatible forever. In fact, many of the distributors contributed to the XFree project (see above). Beyond that, XFree was and is the standard, so it was only natural to use it.
XFree's sudden change to their license was a suprise that many people never saw coming.
Jeesussss....... all this over a BSD'ish clause in the new licensing. Can someone give me a rational explanation as to why the GPL is so problematic in this area? What in hell is wrong with giving credit where credit is due (i.e.: I create something based on a BSD 1.x/MIT/X/Hi I require you to give me credit for being the basis of your creation license, and so , being the upright person that I am, I responsibly give credit.) How does this preclude software, any software created under these conditions from being free, unless the original licensing of the 'base' product I used to create my widget isn't going to allow me to give derivitave works away under my own licensing terms (as in free libre/beer, and if it wasn't going to allow me to do this, I'd drop it like a hot bullet and find something else to use)?
I am so glad that I use the *BSD's. Pretty much avoid mess's like this altogether.
Sometimes people just have to learn and adapt to change, it is one of the requirements of being a living thing.
"Find one who'se license is compatible with your own" is far more efficient.
If you have a BSD-licensed product, you shouldn't feel a need to build your own if you find appropriate BSD-licensed components.
If you have a GPL-licensed product, you shouldn't feel a need to build your own if you find appropriate GPL-licensed components.
If you're making something proprietary, well, I guess yeah, build your own.
I noticed in the first link that they specified that they were remvoing Japanese fonts from Mandrake 10rc1. I happen to use Mandrake because I was impressed with their foreign language support, specifically Japanese. Does anybody know why they are removing Japanese fonts and if there is anything that can be done about it?
freedesktop.org already has replacements for pretty much everything in xfree86. The new license change has just sped up the need for it to work now. They recently released their new xlibs, and Keith Packard is still working on a replacement xserver. The only major problem left is that since the new xserver is a redesign it will need new binary drivers from ati/nvidia.
http://www.freedesktop.org/Software/xserver
>You are still allowed to modify and redistribute the code to your heart's content, as long as you acknowledge the original authors. Wouldn't you want your work acknowledged?
You already broke your idea!
Where's the:
(TM) - This post includes "IP" from Hayes, Inc.?
That's why advertising clauses suck. *EVERYTHING* we know of is a dervative of something. Sometimes it'd be nice, though, because it would force companies like Disney to face the music. But most of the time it sucks because you waste more ink thanking dead people and companies than getting work done.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
Keith Packard from FD.o will resolve the issue to become DFSG-free though how isn't decided:
.pcf.Z font file support shouldn't be a significant issue; .pcf.gz files for many years.
n -l egal-200402/msg00116.html
"Around 14 o'clock on Feb 11, Branden Robinson wrote:
> The DFSG-incompatibility of xc/lib/font/fontfile/decompress.c is more
> serious.
Disabling
XFree86 has used
We can either stub-out the functions in decompress.c or delete their
references in fontfile/fileio.c and remove the files completely.
I'll do the latter for the freedesktop distribution; I have a strong
desire to make the released software from that source completely DFSG-free.
-keith"
http://lists.debian.org/debian-legal/2004/debia
Just out of interest, what is new/changed in 4.4? I looked through the site and didn't find anything. Is it just new hardware support, or more substantial things (ie: proper XRENDER (think that's it anyway) extensions, hardware gl support, rendering of transparancy....)? Anyone got a changelog or brief overview?
just a few corrections:
;)
If you have a GPL-licensed product, you shouldn't feel a need to build your own if you find appropriate GPL-licensed components. Or find some BSD-licensed components.
If you're making something proprietary, well, I guess yeah, build your own. Or find some BSD-licensed components.
Free as in mason.
Now the license has changed for 4.4, they are sticking with XFree 4.3. They have made their own bed, they will sleep in it.
If the XFree 4.4 license is really an issue, then some GPL xfree replacment will mature at a rapid pace while the distributions stick with XFree 4.3.
Mind you I like XFree 4.3, however, if I had to go back to XFree 3.3, I could live with it. I am sure there will be no problem living with XFree 4.3 and the necessary workarounds for new hardware till a suitable replacement is availble...assuming one is actually needed.
vi +
I'd suggest that Mandrake people could not be forced (at law) to go back to older versions than what they already use, if they don't want to conform with the new licence and its extra notification conditions.
If anybody has already used and relied on the latest version (or latest -rc) and its associated licence conditions _before_ the recent statement of licence change, it seems likely there would legally be an estoppel to prevent the XFree86 people succeeding if they try to retrospectively enforce a tightening-up of licence terms -- though I guess they can use new terms freely for their own future releases. [btw, of course this is legal debate not legal advice - CYOLA]
It looks, regrettably, like the kind of action that could make a fork viable and even necessary.
-wb-
I could not find it easily on the freedesktop.org page; so what is the license of freedesktop.org? Is it GPL or BSD or the old XFree license or something else?
--- guns don't kill people, people with guns kill people ---
Can't someone fork the 4.3 version and just continue to use the old license?
There a small handful of files (main issue is old unix compression for fonts) in XFree86 before the licensing change which are GPL incompatible. After the change the entire XFree86 system is GPL incompatible. Rather than simply saying "stuff you XFree86, you're not that important", you want us to bow down to them and try and figure out that legal quagmire? Here's a wild conspiracy theory for you, MS (and SCO) are paying XFree86 to relicense in this way so they can attack every distribution out there, with 4.4, for not following licenses, Open Source = Pirates.
If somehow it was as simple as you would like to make it out (write XFree86 on packaging) then people might grin and bear it (I doubt it), but when they are making a vast array of programs (GPL) unusable do you really expect just follow along?
Never underestimate the dark side of the Source
Jeesussss....... all this over a BSD'ish clause in the new licensing. Can someone give me a rational explanation as to why the GPL is so problematic in this area? What in hell is wrong with giving credit where credit is due
A) If you're a nice person, you'd already do that.
B) If you're not a nice person, it can be deeply buried in some obscure reference somewhere.
And if you then start to spell it out in detail exactly how it must be placed so that it *is* visible (like the original BSD licence on advertisements, XFree in documentation) it's bound to be either vastly ineffective against people that aren't nice, or so restrictive it becomes really annoying to people that are. That's what happened with the original BSD licence. I suspect this one is just going to be ineffective.
I imagine the reason for this whole crap is that under the BSD licence, the source doesn't need to be released, so in a proprietary app noone will know about it. While with a GPL'd program, everyone can read the source and see if they're using other people's work, and give them bad karma about it.
Anyway, if the beef was with proprietary apps, there would be a really really simple solution. EITHER acknowledge in documentation, OR distribute source where the copyright headers would be the acknowledgement. That should be fully GPL compatible, and provide a way to verify that BSD code was used under all circumstances. It's not like the exposure would be much different, very few read either the acknowledgements or the copyright headers...
Kjella
Live today, because you never know what tomorrow brings
This will be the death of us all.. Too many restrictive/conflicting licenses..
Will end up where no one can do anything with out stepping on someone elses license/patent/copyright.
---- Booth was a patriot ----
Try spending 40 hr/week looking at the 16bpp1024x768 the 4.3 frigging *vesa* driver cooks for you on a brand new i865G card!
:-)
Gagh! My eyes are bleeding so badly, I might as well have read the Windows source...
I've been waiting for i830 driver goodness all winter.
But sure, you go ahead and clear up the licence issues first.
I think I'll develop a jesus complex
No sig to see here. Move along.
From the analysis I've seen in Debian lists, the new license wouldn't really be a problem if it just applied to the Xserver. The problem comes with the X client libraries (xlib and friends) that have to be linked with GPL (and other, the GPL is not the only problem here) programs.
Now, when it comes to the users, most of the new features they want have to do with hardware support, which is an Xserver feature. So it's possible that, as an interim solution, systems could be shipped with the new, ugly-licensed Xserver, but with older-but-sanely-licensed xlibs. This would seem to address everyone's issues fairly well.
I've always felt it was a bit of a mistake to have the client-side and server-side of XFree86 tied together anyway. They are pretty much independent, and I think it might make the most sense for XFree86 to abandon the client side, and just focus on making Xservers, while Freedesktop could ignore the server side (at least for now) and focus on the client libraries. Would make both parties jobs easier.
Who cares? It works well enough. At least it's stable and works.
Windows is no better. MacOS, yes. But is MacOS a Free operating system that runs on any piece-of-shit computer you throw at it? No.
The way I look it is like this: you can fix it, or not use it. Pick one, and stop complaining. Is your post on topic, even? Does it have anything to do with the license? No. Hmm.
BTW, it's fine on my GeForce 4 card. YMMV.
My other car is first.
From the looks of the screen shots on their web page these projects seem like an interesting alternative that can potentially put the Linux desktop at a stage not achievable via X alone - A true alpha blended desktop!!!... haven't tried it out cuz the driver support seems lacking... is there any way to port drivers from X into this project without causing all sorts of license problems?
Reading this YADAX (yet another discussion about X) here and problems with same, I remembered that a while ago a bunch of people set out to write a replacement, first called "Berlin", later Fresco. But the "latest news" on their web page is about ten months old. Is Fresco dead or just resting after a prolonged squawk?
As usual, RMS penetrates to the core. The X Consortium has made this software non-free. Ultimately, right or wrong, they have the control.
But here's the issue - the last time this happened we ended up with GNONE (not that I dislike it) and KDE (not that I mind so much). Further fracturing the desktop (not that it's GNOME OR KDE's fault).
Can you imagine what will happen if the general consensus is to dump Xfree86 but NOT be able to agree on a successor? Then you'll have even more problems with compatability to deal with!
Ultimately KDE assured the community with a broader license but we all still live with the aftermath.
BTW, do you think that this licensing in any way affects or is partly because of Apple?
"...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
scripsit Be-Fan:
Test performed with a 2.4 kernel on a stock Debian Sarge box (XF86 4.2.1). Hardware is 1999-vintage PIII/450 with ATI video. Result: Some slight smearing (maybe 0.2- or 0.3-sec lag) dragging Galeon windows over each other.
There's sense in that: I can drag as many xterms, gvim windows, xmms, etc., over each other as I want without a hint of smearing. Only Galeon shows any smearing.
I'll refrain from commenting on the extent to which Qt `rocks', though ;)
In principio creauit Linus Linucem.
Told to me from one of the fdo guys:
In two weeks the Freedesktop.org guys will release X11R6.7.
The short term plan is to use the FDO Xlibs with the OSS XFree driver architecture. This will give compatibility with existing drivers (particularly the binary NVidia / ATI drivers) and many of the features of the fdo X server, apparently including compositing.
Long term, though, there'll be a better driver model, and more communication between the guys writing your X server (fdo) and the vendors (one of the main beefs with XFree86 is that there wasn't much communication with vendors, who often waited up to a year for their drivers to get into XFree).
Ahhh, but the freedesktop X server does not have this problem as the windows are drawn to their own little bit of off-screen RAM and then composited onto each other, thus eliminating the need for application level redraws.
:D
It also means we now have the ability to do TRUE transparency. Soon we will be able to have a movie playing underneith an Xterm at 20% opacity! and finally anti-aliased edjes to our window manager skins
Sadly, the other AC is correct. You do find some very opening minded people here, but you also find some zealots of every flavor, including Pro Windows (thats my bit to not get modded down;). But there are all kinds here, mainly good.
While a bit brash, I agree with your main point, that the problem exists. Desktop Linux as a whole is sorely lacking in smoothness. Its not a lack of built in stuff, there is more than enough for a basic office or home system. There IS a lack of production applications, of the proprietary flavor, for Linux. At least from a small enterprise point of view. But the lack of smoothness on the Desktop, in general, is what is holding Linux back in the enterprise. This and total support for OS and apps, but maybe IBM will fill that void since RH seems to be dropping out of the low to mid end.
Fortunately, Windows XP took away the faster GUI in 2K when they added all the useless, ugly eye candy in XP. Of course 2K was kinda clunky in some ways, but pretty smooth. Linux could be smoother, but it appears the programming necessary isn't very sexy, hense the credit grab. I don't care to have MS go out of business, but we would all have better choices if Linux had, say, 20-30% of the desktop market because of the obvious competitive pressures.
As to forking the code, I have no idea the impact. I'm not smart enough to just know that, I'm not a programmer, I'm a heavy user. I'm dying to move to totally GPL OS with mixed applications, if this moves that possibility along, then I would consider the fork a good thing. For now, I use it where I can, and wait.
Tequila: It's not just for breakfast anymore!
Why does their FAQ say this is impossible?
When I heard (from you) I may be able to play my games and use FDO I was hyped and ready to check it out and see what I would need to do to migrate.
Then I read one of their faq'a here is the relavent Q and A
-------------
Q: Couldn't we just write a wrapper for XFree86 drivers and use them?
A: Essentially, no. There are a large number of calls from XFree86 drivers into XFree86's DDX layer. Furthermore, XFree86 drivers don't support acceleration in the same way, so offscreen pixmaps wouldn't be supportable as far as I know.
-- EricAnholt - 23 Nov 2003
---------
Now maybe they are doiong it with something other then a wrapper, or maybe they changed their mind from 3 months ago, but it is also quite possible you are a bullshitter, so someone please clarify.
Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
They are indeed doing something other than a wrapper. The specific comments about using fdo Xlibs with XFree's driver infrastructure came from Daniel Stone and seem to be corroborated by reports from the FDO planet site.
It would be *really* hard to write a compatibility layer for XFree drivers. I looked into doing that myself awhile ago, and while it would be possible, it would involve implementing many XFree86 DDX calls in kdrive. The NVIDIA driver, for example, makes 400 different calls into XFree86. If something like that were to be done, it'd be a long term project. Most importantly, there is no mention of this in the mailing lists.
Maybe what you're thinking is that they plan to release XFree86 with the FD.O Xlibs, which would be doable. It would only include compositing, though, in the sense of supporting the Composite extension, not full window compositing.
I could be entirely wrong though!
A deep unwavering belief is a sure sign you're missing something...
Okay:
Bottom window: 1600x1200 Xine window playing Hellsing Episode 6 (awesome anime!)
Top window: ~1000x700 Konqueror window opened to Fark
Moving the top window results in barely noticible amounts of black areas on the video window. The underlying video doesn't stutter one bit.
This is with NVIDIA's binary drivers, XFree 4.3, and KWin from KDE CVS. Oh, and kernel 2.6.1.
A deep unwavering belief is a sure sign you're missing something...
Now comments are like "What do you know, it works. Now where are the p2p apps...."
---- MISSING MISCELLANEOUS DATA SEGMENT --- [sigdash] trolololol
I looked at the thread regarding Xfree on Cgywin and was amazed at the attitude of the core developers, especially David Dawes.
As many others have said, David Dawes is a liability - no matter how good the coder, if lead developers have poor attitudes (or in this case, APPALLING), the project is doomed whether it be closed or open source.
Congratulations to the various distros and developers such as Keith Packard for saying enough is enough - in the long term we'll all benefit from this.
"Okay, every time we talk about fixing it and proposing ideas,"
Uhm no, you get flamed because people don't propose (sane) ideas. People only complain, people insult developers personally, people insult the software, people say things like "REPLACE IT!!!". None of them are ideas.
What really saddens me is that people think they are contributing ideas while in reality they are only insulting other people/things, and they don't even realize that.
[...]
Various games: ~500MB, ~5CDs
[...]
Games: Included in distro
so what?
ut 2003 included in redhat?
halflife included in redhat?
baldurs gate included in redhat?
sorry but something is wrong with your calculation...
ps:
win2k with office2k, iis, sql server, whatever: just below 1gb
so NOW WHAT?
(before starting to flame me for my biasedness: i use linux whenever i can)
-- Karma: beyond good and evil - mostly affected by posting political