Among the major Linux distros, Slackware was the only one using XFree86 4.4.0 (in its -Current tree). On Sunday, Patrick Volkerding dropped XFree86 and moved to XOrg too after receiving input from his users. This change leaves XFree86 without any major distro using it.
Slackware Goes X.Org
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
David Dawes brought this on himself and the rest of the XFree86 developers. There’s no-one else to blame. Good job the open source movement has put together an alternative so quickly in X.org, and even more interesting things are coming via the FDO XServer soon.
> This change leaves XFree86 without any major distro using it.
I thought NetBSD is still using it, isn’t it?
Is when do we get a graphic installer? Slackware installation is just too cryptic for me to figure out.
NetBSD is not a Linux distro. When saying “distro”, we are generally talking about linux. For BSDs, they mostly call it “flavor”.
I’m using Xfree 4.4.0 on Gentoo despite no support. I used X.org but Xfree is a better way in my opinion. This is a suck fork sponsored by many “great” corporations…
I used XFree86 4.4.0 on my Gentoo box as well and now I switched to X.org and I don’t see why XFree86 shoul be “a better way”. Can you explain?
“more interesting things are coming via the FDO XServer soon.”
yep, sure am looking forward to these things… does anyone know how things are progressing?
why should slackware need that – it’s not a distro for newbies – if you want a quick and easy install then use one of the great alternatives (xandros, suse, fedora…).
> I used X.org but Xfree is a better way in my opinion.
If you want to use _non-free_ software, I won’t stop you.
> This is a suck fork sponsored by many “great” corporations…
Don’t mix them up. Xorg is the fork and XFree86 is sponsored by David Dawes’ “great” company.
Is when do we get a graphic installer?
Hopefully not any time soon. Many find the Slackware installer easy and reliable. I happen to be one of them. Maybe one day a graphical installer will be needed to abstract the complexity of installing an OS, but not yet. Heck, it’s not even close.
My question is when is 9.2 coming out?
> If you want to use _non-free_ software, I won’t stop you.
Is it necessary to post things without knowing something about it? Please read this:
http://www.xfree.org/4.4.0/README2.html#2
And please take a look at the XFree86 License 1.1:
http://www.xfree.org/4.4.0/LICENSE4.html#7
And don’t think there are problems with the GPL, the xlibs have the XFree86 License 1.0 like before and programs are not linked against der xserver.
I hope he was just kidding. (Score:5, Funny)
Seriously though, you only need to install once. So what good will pretty pictures and smiley faces do?
A little Googling or a visit to the freedesktop site would have revealed that Xorg provides a couple of tools that automatically probe your video card and build a basic configuration file.
A script, xorgconfig, which corresponds to xf86config, is alos available. You’ll need to know your monitor’s refresh rates; the identity of your video card and the amount of memory on the card; what kind of mouse you have, and the resolution(s) you want to use. Knowing that, you can configure your server in about 5 minutes.
Frankly, if you don’t know those kind of details about your hardware, you probably ought not to be using Slackware. Slack does minimal monkeying with the code in the software it provides. A lot of us like that, because it makes things much simpler to maintain (no need to worry about distribution-specific tweaks breaking things).
I’ve never found the Slackware installtion program to be especially cryptic or difficult. (And, of course, it is not command-line based, something you might know if you’d ever actually used it.)
> Is it necessary to post things without knowing something about it?
Okay, maybe I’ve used a very strong wording, but IMO something isn’t very free if it forces you to insert attributions into the docs.
> And please take a look at the XFree86 License 1.1:
> http://www.xfree.org/4.4.0/LICENSE4.html#7
A classic 4-clause BSD license with the (in)famous advertising clause. (Yes, I know, not exactly non-free, sorry for that.)
Why not drop this shite, and develop a new graphics core, something similar to OS X, or Longhorn?
http://www.xfree.org/distro-support.html doesn’t mention Slackware anymore, and I guess four more Slackware-based distributions will disappear from this list soon.
It’s not like XFree86 was making any money!. Either they fork XFree86 and make it a proprietary binary-only product to compete with X.org or just give up their fight.
If David Dawes is listening, I would suggest that they should forget about doing opensource because the train’s already left the XFree station and has stopped at X.org. The only way XFree can be relevant is to create a binary-only version ala XiG or MetroX and compete on features and ease of use.
Installers for X are simply pathetic and XFree86 guys can work to fix up all these things.
http://www.y-windows.org/
Where I can find more information about the technical diferences (present or planned) between X.org and the XFree86?
Spooky, the “X needs to go” people have been banging that point around for a long time. The important thing is that linux has been a server os for a long time. X is great because it allows you to run individual GUI programs from the server the same way you can run command line programs. This is good, this is innovative, and this should stay.
The “X should die” crowd is kinda quiet recently since Keith Packard has recently proven that we can have X and have all the cool stuff OSX and Longhorn have. I suggest you check out the “new” Freedesktop X Server.
>>>>>NOW FOR MY QUESTION<<<<<<<<
I subscribe to the FD.O xserver mailing list, and it seems like the plan is to integrate FD.O into the X.org server. Does anyone know how this is going to happen? I believe Keith’s server is based on Kdrive, which is a great project that is X11 compatible, but doesn’t require any configuration files. My understanding of X.org is that it’s a free implementation of Xfree86. Maybe i don’t totally understand how the “X Windowing System” works, but it seems like these projects have somewhat different goals.
I usually update my slackware installation with swaret to the current tree. now, how is this going to work with the new X server? is swaret going to uninstall xfree86 packages as if I’m upgrade them?
Is that supposed to Read XFree 4.4 ? Which other distro have switched to X.org for stable release ? Just curious
Glad to see Slackware returning to sense. As one of the oldest Linux distros it was a shame that they failed for so long to see the writing on the wall. It is simply irresponsible for any Linux distro to ship with X!NOT!Free86. If someone wants to use it on their own fine-but no one should be distributing it.
Luckily the layout of files in X11 is really starting to change-in the comming months those who still choose to support X!NOT!Free86 will be faced with more and more difficulties-files won’t be where expected, things will fail to compile etc. David Dawes, as head of X!NOT!Free86 had the respect of the entire FOSS community-he blew it and now his project is doomed to obcurity. I wonder how long X!NOT!Free86 is still even going to be around-3 months ? 6 months? With no new developers and but a fraction of those which they once had the only way left for them is down.
Actually all said and done the fact that the community didn’t split up when confronted with this is a testimony to the strength of the community.
My argument with XFree86 had nothing to do with the licensing really. However, I think this is a great move as I see X.org having a more open development and progressing a lot faster than XFree86 would have.
“I subscribe to the FD.O xserver mailing list, and it seems like the plan is to integrate FD.O into the X.org server. Does anyone know how this is going to happen? I believe Keith’s server is based on Kdrive, which is a great project that is X11 compatible, but doesn’t require any configuration files. My understanding of X.org is that it’s a free implementation of Xfree86. Maybe i don’t totally understand how the “X Windowing System” works, but it seems like these projects have somewhat different goals.
Report a”
the plan is to drop xfree86 base completely and use a compatible configuration file for compatibility reasons. xlib based stuff would work but the drivers particularly proprietary stuff would need some changes. all the fancy effects thats supposed to be in GUI’s 2 years later is already being coded in here. however all this would take around 6 months time or so. probably can be expected in fc4 as an alternative. fedora plans to allow swapping of multiple implementations
Every major and most not so major distros have already made the switch to XORG X11. SuSE, Redhat, Mandrake, Gentoo, Debian, etc.
in fact the only Linux distro’s to support X!NOT!Free86 are:
“Linux® based distribution
* Conectiva Brazilian-based distro with a world-wide following using RPMs.
* Arch Linux a Canadian distro; i-686 optimised.
* Ark Linux an EU Redhat-based distro using ISOs.
* Lycoris Linux a desktop friendly environment for novices.
* Magic Linux when native Chinese-support is desired using ISOs.
* OneBase Linux a meta distribution.
* OpenNa Linux when security matters.
* Peanut Linux an 85MB base.
* Rubyx Linux object-oriented ruby is its scripting language.
* Source Mage a source-based distro aimed at linux magicians (sys admins) with a social contract.
* Sorcerer Linux a source-based distro aimed at linux wizards (sys admins).
* Yoper Linux a new down-under (New Zealand) distro.”
* i subtracted the slacker based distros-they too will have to change if they have not already done so.
And now we know who to boycot.
“I’d like to take this moment to thank the XFree86 Project for all the truly amazing work they’ve done all these years, and to wish the project the best of luck. Slackware owes the XFree86 Project a debt of gratitude and will always include the XFree86 acknowledgement, even if we are no longer shipping XFree86.” –PV
The Man said it as it should, despite the unpopular decision of the new leadership. So its only NetBSD that is using it.
I heard they were fixing bugs on X.org, but so far, xorcfg is dung, sets impossible modes. The curses mode segfaults. Xcalc is a jumble of 4 little squares with a zero in the middle. Xorgconfig works if you know your hardware and can probably get a working config. Xorg -configure seems blind to anything but /dev/mouse. Trivial nit picking but a real pain in the arse when you are trying to build an all knowing self configuring distro. This is just a cursory glance too, who knows what else is broken, I’m not real impressed and more than a bit disappointed.
Anybody got a clue if anyone is working on a cygwin port of X.org? I know, it’s not a ‘distro’ but occasionally, I like to test X Windows apps in XP… and it would be better if cygwin was using the same x windows as my fave linux distro’s.
Now, on that topic. What’s this nonsense about no major linux distros ‘using’ it? This begs the questions:
1. What’s using it?
2. What’s major?
To my knowledge these distros still ‘use’ it:
Redhat
Debian
Suse
Libranet
Mandrake
a host of lesser known distros
and these are still committed to ‘using’ it:
Conectiva
Lycoris
a host of lesser known distros
read about 4.4.0 support here:
http://www.xfree86.org/distro-support.html
Clearly, support is leaving XFree86, but it is still deployed much more widely than X.org.
I don’t have a personal preference, either way, but – man! How quickly folks abandon a project that has given so much to the community. Think just how few people would have tried Linux, if it hadn’t had XFree86 – no windowing system – no real acceptance. No flames from the console crowd, please!
Later,
Will
X!NOT!Free86
Please stop this childish behaviour. Those of us reading this board are not five years old.
Just to clarify for you and the rest of the immature crowd, XFree86 *is* free software. It’s just not compatible with the GPL. In fact, from reading the mailing lists a while back I understand that changes were made so that it IS legal to link GPL programs against XFree86, only some very specific core portions of XFree86 are under the new *free* but incompatible license. The Xorg fork sounds like it’s for two main purposes: a) spite b) incorporate new technology that the old XFree86 board was reluctant to.
Preserving the ability to convert the whole program to GPL is probably a third loose reason for the fork, but not a good reason in my opinion.
Now, if you people would be so kind as to stop spreading FUD, maybe we can could have an adult conversation.
We support Xfree86 4.4.0 and we will continue our support with this license.
Though we use Xorg officially, you can switch to xfree86 4.4 anytime.
olm -s | -b xfree86
There is however once clause: you cant perform binary installations or create binary packages. (however if you switch back to xorg then you can)
Hmm,
Having read a bit more on the background of the forking of X.org from XFree86, I will grant that there is merit to switching to X.org from XFree86 on the basis of licensing. However, the statements by the author that no major distros are using XFree86 is still exaggerated.
Properly chastened,
Will
Thanks, I’m tired of the FUD too.
Since when is Debian using X.org?
Every branch is still using XFree86.
yea indeed, but Debian doesn’t use 4.4 , it’s using 4.3
They are merging features from the X Server into the Xorg server. XDamage is already in CVS and working alright. The composite manager could be in there now; if it isn’t, it will be soon. I asked how long this merging is expected to take on IRC and was told it will happen “soon” – withing the next 1-2 releases. Realy eye-candy on Linux is not far off.
“Since when is Debian using X.org?
Every branch is still using XFree86.”
he was talking about xfree86 and not x.org.
fedora core 2 is the distro that has already adopted x.org.slackware has recently switched. mandrake probably will. the next version of redhat enterprise will have x.org
debian will not distribute the new xfree86
http://freedesktop.org/pipermail/x-packagers/2004-February/000003.h…
connectiva thou not well known is a major distro in some parts of the world and its important that we dont ignore it in making such statements.
x.org is just a fork from the xfree 4.4 rc and as of now isnt very different. the xfree86 license is incompatible due to restrictions
comment from rms.
http://www.xfree86.org/pipermail/forum/2004-February/003974.html
debian legal even had discussions whether it would be considered a free license.
more importantly the amount of changes in xfree86 was not satisfactory to a large number of people. x.org is likely to adopt changes faster and things which xfree86 refused to adopt
So…
Are there technical differences? Did the X.org guys simply need a different way of managing contributions or are they doing/planning anything radically different?
Right now there are very little. At least some of the XFREE86 4.4 source was resubmitted by the authors to the XORG team. I think the biggest difference right now is the binary driver support from ATI and such, im not sure they work, though they might
Thank you. I was tempted to respond but you said it quite well.
@Child: A 5 year old who can use prepubescent in a sentence, I am impressed. Now go play with your toys and leave this thread to the grown-ups.
“I’ve never found the Slackware installtion program to be especially cryptic or difficult.”
What’s he’s probably talking about is what a PITA fdisk is. While a graphic installer is probably out of the question, disk partitioning could be simpler.
What about cfdisk?
If slackware is to cryptic for you then that means you should not be using it. Slackware is not for your run of the mill average desktop user IMHO. Stick to Fedora, Suse, or Mandrake.
You have to update your package list then install x11 packages. Once they are installed you can then delete xfree packages. I rebooted and it worked fine but I decided to run xorgconfig to make sure that everything was the way I wanted. My nvidia drivers worked without a problem.
Eugenia (IP: —.osnews.com) – Posted on 2004-05-31 15:04:16 said:
NetBSD is not a Linux distro. When saying “distro”, we are generally talking about linux. For BSDs, they mostly call it “flavor”.
While true, what about the software found at http://www.GoBSD.com? Wouldn’t it be more aptly called a distribution of DragonFly BSD (a soon to be commercial one, no less), as opposed to a flavor?
“A classic 4-clause BSD license with the (in)famous advertising clause.”
This was in fact one of the major bones of contention. David Dawes was adamant that the new XFree86 license is functionally identical to the 4-clause BSD license, yet he refused to use the same wording as Berkeley. Either he went to the considerable trouble of writing a new license from scratch for no reason whatsoever (doubtful) or he was lying.
This was the opinion of (at least) Theo de Raadt (http://marc.theaimsgroup.com/?l=openbsd-misc&m=107696705911864&w=2) and Branden Robinson, the Debian XFree86 maintainer (http://lists.debian.org/debian-legal/2004/02/msg00162.html).
i also wondered why Slashdot readers went apoplectic about the news of the gpl withdrawl. Why the foamy mouthed hesterics. They just want credit for their work.
It’s still FREE !
Alot of these distributions are playing follow the leader.
“i also wondered why Slashdot readers went apoplectic about the news of the gpl withdrawl. Why the foamy mouthed hesterics. They just want credit for their work.
It’s still FREE !
Alot of these distributions are playing follow the leader.”
And I hope they keep playing follow the leader. Funny how you chose not to mention the problems XFree86 development had in leadership. Here is one example, and only one of many where developers were not allowed CVS access:
http://cygwin.com/ml/cygwin-xfree/2003-10/msg00328.html
Please feel free to keep pretending this fork is solely about licensing. In the meantime, I look forward to the explosion of X.org development that is now happening and am extremely greatful for the open development model it will now follow.
Mandrake Cooker already has X.org. Just for information
The fact that it’s “free” is not the point. Gnu/Linux has to be a GPL compatible OS on principle. Even things like Nvidia’s binary drivers are less than ideal because we can’t see the source. The whole point of the GNU project is to have a totally free, open source, modifiable, redistributable operating system that is still technically superior to propriatary operating systems.
As I understand it, there’s little, if any, real difference between the current releases of Xorg and XFree86, at least differences visible to users.
I suppose developers will always squabble about license issues, and attempt to ostracize projects using licensing schemes that meet with their disapproval. But, as a user, I’d rather see less squabbling and better stuff. The annoyances you’ve mentioned have been there for some time. Frankly, I care a lot more about fixing them than I do about the bloody license.
The X speed issues while using it over a slow link? X speed over anything else than Ethernet is a shame.
I’m not trolling, do you not believe me? just try to use X over a modem connection, and then try the same modem connection but using VNC. VNC beat’s X by a long way.
Is X.org dealing with this issue???
Granted, fdisk won’t when any charm contests but it is the standard fdisk that is used everywhere. It isn’t like someone wrote it just to annoy potential Slackware installers.
Writing a GUI frontend to it wouldn’t be that difficult, I suppose. Not like you’d need to write a new partioner; just pass parameters to fdisk.
”
I’m not trolling, do you not believe me? just try to use X over a modem connection, and then try the same modem connection but using VNC. VNC beat’s X by a long way.
”
i do that on a daily basis and x connects much faster than vnc. are you sure you are doing this correctly. x connections are as fast as unix sockets btw
Now why do you think I choose to writ >>X!NOT!Free<< ?
Was it perhaps because of the license changes ?
Was it perhaps because several of the heads (core) of XFree86 had been actively chasing away some of the best developers in the community ?
Was it perhaps because these same people(core) kept dragging their heels over and over again regarding new drivers, patches, and updated drivers-written and contributed by “non-core” devlopers- which were never included or only much, much later -due to their mishandling of it ?
Was it perhaps because one man alone saw fit to change the licensing upon which the work of thousands of others depend-with no communication with the community-no debate-simply fiat ?
Was it perhaps because this same man developed things which should have been part of X but chose to make them part of his companies own propietary Xserver-ie. working against the community?
How many developers did the XFree86 core walk all over ?
How many good developers left the scene due to the stupid internal politics(core) at XFree86 ?
How many other projects (Cygwin/DRI) were constantly being held back by such stupid internal(core) politics ?
David Dawes afronted the entire community with his little “license change”. But that was just the straw that broke the camels back. XFree86 had been anything but Free -understood as open-for several years now. Some of the old “core” don’t even use, let alone develop XFree86 anymore-but of course they could strip people like Keith Packard of cvs priveledges. The insular little community of XFree86 was hostile towards the community at large-they did nothing to help foster a broader community-in fact they worked against it. The moment it became obvious that people were considering a fork-David Dawes went and changed the license-now why do you think he did this ? why at that point in time ?
David Dawes made a fundamntal mistake- he declared unilaterily that the only then existing X11 implementation for the x86 world was “his” project-to do with as pleased him. His license change was and is a very, very large middle-finger stuck out at all those who would dare and challenge his hegemony.
Actually the change in the license simply codified the non-open atmostphere which permeated that XFree86.
Look at what is now going on with X.org- Sun is using the new xcomposite, fixes and damages extensions in their Look Glass desktop project. Cygwin-X11 has now become a part of X11-as it should have been all along- the pace of improvements illutstrates such clearly. X.org has opened the doors to the community at large-welcoming new developers and hackers- working together with the distributors to create a better implementation. Did you ever wonder why X11R6.6 had been stagnating for so, so many years ? why no work had been done to further develop the standard ?
I am certainly not alone in saying that X11 development was being held captive by the XFree86 core. Over the years that group did many great things-everyone in the open source community owes a tremendous amount to XFree86 for their historic contributions. Linux would never have gotten anywhere without it. But at some point along the line that group became a insular little community with serious problems relating to the rest of the community-resistant to change, hostile towards those who dared to actually try and improve things-they showed zero vision over the past couple of years and failed to integrate themselves with the rest of the community.
As such :
X!NOT!Free is a perfect name for what XFree has become.
They had their day in the sun. Now the times have changed and the future of X11 looks brighter than it has in many, many years.
Hear, hear! Excellent post. I don’t understand why someone would report you for abuse…To whomever reported it: how about coming up with counter-arguments instead of trying to censor posts you don’t agree with?
You pretty much laid the XFree86 licensing issue to rest with this extremely insightful post. If you are not an attorney, you should be.
i do that on a daily basis and x connects much faster than vnc. are you sure you are doing this correctly. x connections are as fast as unix sockets btw
X seems, to me, to be the most bandwidth-hungry of the alternatives. Using ssh -C for compression helps, but I would welcome some streamlining. Optionally moving more processing to the client, perhaps?
yeah its been dealt with. nx is soposed to be able to run x just fine over a 9.6kbps link. they have made something similar to a proxy for X that runs heavy compression. and all of the nice stuff the made is opensource they have some parts that is closed (the nice configuration utilitis and stuff)
iv heard some talk about bringing nx compression to X so we just have to wait and se
Well, I’m not an expert or anything like that, I just dial in onto a VPN using a modem, I’m running eXeed locally on a Win2k box and running X apps remotelly on a Sun box and a Linux one running Debian.
On a modem it crawls to death, but surprisingly TightVNC is usable whereas X is impossible to work with.
Is anything out there that can accelerate X over a slow link that I’m missing???
“Anybody got a clue if anyone is working on a cygwin port of X.org? I know, it’s not a ‘distro’ but occasionally, I like to test X Windows apps in XP… and it would be better if cygwin was using the same x windows as my fave linux distro’s.”
I’m pretty sure Cygwin already uses X.org. I noticed this while installing it again on a different machine.
X on Cygwin seems a bit more buggy than when it used Xfree. In fact, X in general seems more buggy than 1 year ago. Maybe just me getting more perceptive of slight bugs in Linux and FBSD.
I don’t want to sound like a fanboy but NX from nomachine is really amazing. If you have a direct network connection, ie. not a modem/isdn/dsl/cable, running at 10mbps X server-client speed is pretty damned good- at a 100mbps you can’t tell that you aren’t local. But even on my 768kbps up/128kbs down DSL line X is only barely usable.
I had heard about NX before-but had never actually tried until I noticed they had GPL’ed most of their stuff. I downloaded one of their sanpshots about a month ago and all I can say is wow. I now use it from home via my DSL line to connect the LTSP server I adminster. I find it hard to believe myself-but it is almost as fast sitting at one of my LTSP clients locally which is conencted via a 10mbps network.
I really hope that the GPL’ed parts of NX find their way into X11 and furthermore I hope it get’s integrated with LTSP. I am now trying to get the department to move to 100mbps connection for the LTSP clients-but if I could use NX with LTSP we wouldn’t even need to upgrade the router.
A snippet from the Slackware current changelog:
“Something I’m *not* in favor of is dragging around two
nearly identical projects, so XFree86 4.4.0 has been moved to the
/pub/slackware/unsupported/ directory on the FTP site.”
Earlier, he said the vote in favor of X.org is 4:1.
“I’m getting sick of seeing stories like this, Slackware does X, I didn’t see a story when Fedora Core’s development branch switch, or when Gentoo decided do the same.. and so on.”
So your gripe is its not your favorite distro. yawn. lol.
Gentoo gets all the hype it needs (and more IMHO), as does Fedora. Why do you need yet another site doing the same? I think hearing about slack (and I’m not a slack user) is refreshing, especially considering that you’ll see the mandatory gentoo comment in almost any thread about almost anything, no matter how unrelated it might be. sigh.
When it comes to X There are many things about it that still shake my brian but I have to agree that this is a good move. If anything to help speed up the enviroment and stability along with placing in eye candy that we see in Windows and OSX. How moving to X.org will affect me in general with drivers and what not, I really don’t know but I for one agree with the developers that wanted faster movement on the project
Good move Pat!
Over the years that group did many great things-everyone in the open source community owes a tremendous amount to XFree86 for their historic contributions. Linux would never have gotten anywhere without it.
This is the important point. So because the XFree core team started not to listen to the community and dragged their feet in implementing new features, you find it fitting to go and disparrage the XFree brand name which represents all those historic contributions as well as current core team?
This is nothing but being childish. If you want to make a statement about the current core team, do so, but do not piss on the efforts and time of all the people who developed XFree86 over the years.
I’d also like to point out that the source to XFree86 has been free in the FSF sense for many many years, and in fact it still is. Nothing stopped people from forking earlier if they had problems with the XFree86 team. Notice that up until recently no one bothered to fork? The community was happy to let the core team continue to do all the work despite the poor leadership.
The communities hands were never tied and as a whole the community has no one to blame for lack of leadership other than itself. Quite simply, until now no one cared enough to fork and fix the leadership problems.
Was it perhaps because this same man developed things which should have been part of X but chose to make them part of his companies own propietary Xserver-ie. working against the community?
May I remind you that no one HAS to write free software. They do it out of their spare time and general good will. You have absolutely no right to attack Dawes for not giving away all his code completely free. You again forget that Dawes has donated a lot of his time and code to the XFree86 project, no doubt much of which is still in use in the Xorg fork. We should be gratful for this donation and not attack him for ‘not donating enough’ and wanting a little credit for his hard work. Just think of the number of developers that exclusivly create proprietary work next time you want to attack an open source developer for ‘not giving enough’.
Given all the issues behind the fork, your behaviour is *still* childish and inappropriate. I agree with the fork for technical reasons, but I also value all the work that has been done on XFree86 and do not begrudge the old core for wanting some credit.
“You again forget that Dawes has donated a lot of his time and code to the XFree86 project, no doubt much of which is still in use in the Xorg fork. We should be gratful for this donation and not attack him for ‘not donating enough’ and wanting a little credit for his hard work.”
Dawes is not the only code contributor. Alan Cox, along with many other OSS developers contributed code too & they contributed code to XFree86 under a certain license, which Dawes changed, without even bothering to ask the other contributing developers how they felt about it.
“but I also value all the work that has been done on XFree86 and do not begrudge the old core for wanting some credit.”
And I do not begrudge the other contributing developers, such as Alan Cox, NOT accepting the license change on the code THEY contributed.
Of course, our opinions don’t really matter. XFree86 will wither and die. This is nothing but Evolution in action.
“I’m using Xfree 4.4.0 on Gentoo despite no support. I used X.org but Xfree is a better way in my opinion. This is a suck fork sponsored by many “great” corporations…”
I don’t know if you’d call it a fork. You see, since xfree is essentially dead I think xorg is more of a replacement or a peaceful takeover?
TMK Dawes is the only developer on xfree, is this a myth?
And calling xorg bad (suck) is pretty silly since it’s not really different from xfree just yet.
It’s more like a takeback, than a takeover..
From what I understand, XFree86 itself was a fork of X.org’s X11 a long time ago.
“From what I understand, XFree86 itself was a fork of X.org’s X11 a long time ago.”
1. What is X11R6.7.0?
X11R6.7.0 is the seventh full release in the X11R6.7 series.
X11R6.7 is the current X.Org Foundation release series, based on the XFree86 4.x code base. The first release in the XFree86 4.x series was in early 2000. The core of X11R6.7 is a modular X server. The 6.7.0 version is a new release that includes additional hardware support, functional enhancements and bug fixes. Specific release enhancements can be viewed in the Release Notes.
Come on, now, how is forking X.org “pissing on the efforts and time of all the people who developed XFree86 over the years”?
you find it fitting to go and disparrage the XFree brand name which represents all those historic contributions as well as current core team?
The “XFree86 brand name”? You’ve got to be kidding! The historic contributions are in the code, not in the name.
I’d also like to point out that the source to XFree86 has been free in the FSF sense for many many years, and in fact it still is.
Uh, no. The notorious advertising clause makes it incompatible with the GPL.
I’m sorry, but considering how often you’ve attacked the GPL in these comments section, it seems to me you’re only taking advantage of this issue to push your own agenda. You are conveniently ignoring the fact that a lot of past XFree86 contributors (including some key individuals such as Alan Cox and Keith Packard) have approved of the fork.
Over the past few years, XFree86 development was notorious for being slow, inefficient and – above all – controlled by a hermetic clique. Despite the source code being available, it didn’t take advantage of the FOSS development model. It seemed to many that some features that would put X on the same level as (and beyond) commercial OSes – such as compositing and desktop double-buffering – were being deliberately kept out. It’s as if the XFree86 board wasn’t all that interested in making their software competitive!
Karl’s post was right on the money. The only one acting childish here is you by criticizing a group of talented developers (more talented than those left at XF86, IMO) just because of your anti-GPL agenda.
Dawes is not the only code contributor. Alan Cox, along with many other OSS developers contributed code too & they contributed code to XFree86 under a certain license, which Dawes changed, without even bothering to ask the other contributing developers how they felt about it.
…snip…
And I do not begrudge the other contributing developers, such as Alan Cox, NOT accepting the license change on the code THEY contributed.
May I kindly suggest you keep your mouth shut if you don’t even know what you’re talking about. From David Dawes himself:
No worries. License changes apply only to code where “The XFree86 Project, Inc” is the copyright holder (and a subset of that, in fact). XFree86 contributors remain free to choose which license applies to their code, within classes of licenses that are listed in our LICENSE document.
This was posted in direct response to Alan’s concerns.
So, no, David did not change the license on Alan’s code without bothering to ask. He didn’t change the license on Alan’s code at all
Adam
“This is the important point. So because the XFree core team started not to listen to the community and dragged their feet in implementing new features, you find it fitting to go and disparrage the XFree brand name which represents all those historic contributions as well as current core team?”
If you had bothered to read what I wrote you would not be listing those two things as the primary reason for writing what I wrote. Through decisions that the XFree86 core made several first-rate developer/hackers where excised from the XFree86 development community and many more were simply driven away from even wanting to contribute. You can go and read up about this in the mailing lits if you want. I am not going to do your homework for you.
“what” changed in the licensce was not and is not as important as “how” this change was effected. There are thousands of projects which build on XFree86-far more developers than XFree86 has ever had- who based their projects on the work of XFree86. A bond of trust existed- trust that they could count on XFree86 to maintain the existing licensing arrangement which had been in place for several years. This trust was broken-it was broken by David Dawes.
Sure Dawes later revised his licensing changes to re- enable GPL linking against the xlibs-but when he first announced his changes in the license he did not take into account the work of all those who were dependent upon this freedom to link.
He only later changed it after everyone screamed bloody murder. There is no one who used XFree86 who did not know where it came from and who the people were who were developing-there was never any kind of lack of “brand recognition” or other such nonsense. You would be hard pressed to find somebody who thought Redhat or SuSE wrote “that graphic thingy which puts windows on my screen” in the Linux world. Once upon a time XFree86 stood for something- stood for a free opensource implementation of the X11 protocol. They were very well respected by almost the entire community who appreciated what these developers contributed.
The X11 system is not just any open source project. The fundamental role which it plays in the FOSS world is matched only by the GNU toolset and the Kernel itself. If Richard Stallman was to try to change the licensing of the GNU toolset or Linus Torvalds did the same there would be a very similiar reaction from the larger community. The livelihoods of 10,000’s of people are directly dependent upon such projects-and millions are indirectly dependent upon these projects around the world. With such a fundamental role comes a degree of obligation, a responsibility.
The right thing to do when someone is the head, or is one of the core members, of such an important
project is to step down and hand this responsibility over to others who value it and will continue it-rather than engaging in solitary acts which alienate the community to which it belongs.
I am disparaging the name “XFree86”-those who run XFree86 no longer deserve this name, for they are no longer living up to what is entailed in such a name.
“This is nothing but being childish. If you want to make a statement about the current core team, do so, but do not piss on the efforts and time of all the people who developed XFree86 over the years.”
No, David Dawes was being childish by unilaterily deciding that XFree86 was “his” project to do with as he pleases without communicating his intent and opening up the doors of communication which could have led to a more amicable outcome.
I do not -and did not piss on the efforts that XFree86 developers made in developing a free implementation of X11 for the x86 platform. I used XFree86 for more than 8 years-and was well served by it. I have never had a problem with the quality of the code. I do appreciate the fine work that has gone into the XFree86 project. I do appreciate the time and energy which David Dawes personally invested in XFree86-he has made great contributions.
But the core had been engaging in the politics of exclusion-rejecting much of which was being contributed(new drivers, driver updates, patches etc.) and rejecting first-rate contributors, most notably Keith Packard, from even participating-and had driven many previous developers away. The vast majority of XFree86 developers had no say in the decisions which the core engaged in. The actions of the core do not speak for the majority of the developers. But the core is self-selected- the developers have no voice in the decisions which the core makes.
I trully believe that those who developed for XFree86 will find a new home at X.org-a new structure which has emerged as a result of the cores decisions. They would then have a voice in the decisions being made and could work together far more closely as an integral and inclusive community. I no longer support those who contribute to XFree86. I have no choice in this matter for I do not respect the decisions which the core made and the licenscing decision made by David Dawes.
“The communities hands were never tied and as a whole the community has no one to blame for lack of leadership other than itself. Quite simply, until now no one cared enough to fork and fix the leadership problems.”
No one who worked on X wanted to fork the project. The developers who were expelled from XFree86 went out of their way to make it clear that they were not trying to fork the project- they tried to communicate with the XFree86 project- tried to heal the trust that had been broken. No one in their right mind forks such a fundamental project lightly.
You are correct in one sense-the larger community failed to show an interest in spearheading development of the X system-but I would argue that this has as much to do with the politics of the XFree86 core group and it’s relationship to the distributors and the community at large as it does with the initiave of the community itself. Put simply-the only way things could change was via a fork-and no one wanted the fork to happen because of the nastiness which inevitably ensues once a fork happens.
As a visibile head of one of the most important open source projects David Dawes wears the shoes that fit him. He is the one who drew this attention to himself. He made these decisions. I as a member of the open source community have every right to voice my opinion about his actions. His decisions effect me-I will not quietly let such, with which I so passionately disagree with, pass without commenting on it. I feel an obligation to speak out about this issue. It is part of my responsibility. I never said that David Dawes had not contributed enough-or that he or other developers HAVE to give us anything.
The truest form of payment in the open source world is mutual recognition and respect. I respect what David Dawes and the other XFree86 developers have done, past perfect tense. I no longer respect the Xfree86 project. If their policies change and the core members leave and open up the project again I may come to respect XFree86 again some day. But untils such time I do not feel that I am being childish in expressing my discontent with these decisions and those who made these decisions. If anything I am admonishing them to live up to the name of something for which they should be proud of. They no longer deserve my respect.
”
So, no, David did not change the license on Alan’s code without bothering to ask. He didn’t change the license on Alan’s code at all
”
its important to ask the others who have contributed their opinion and have a dicussion before changing something as important as the license regardless of alan’s code was having the same license. david cannot change alan’s copyrighted code btw.
david went ahead and changed the license single handedly. the last good feature after several years was fontconfig and xandr by keithp and he was prohibited by david recently so its clear xfree86 had to fork regardless of the license change. the license is just what broke the final links
“May I kindly suggest you keep your mouth shut if you don’t even know what you’re talking about.”
Only if you are so damn perfect that you have never made a statement based on incorrect information. I stand corrected.
Anybody got a clue if anyone is working on a cygwin port of X.org? I know, it’s not a ‘distro’ but occasionally, I like to test X Windows apps in XP… and it would be better if cygwin was using the same x windows as my fave linux distro’s.
Yeah, Cygwin was one of the first ‘distro’s to make the switch to X.org, because the XFree86 guys didn’t play very nicely with them. Anyway, looks like pretty much every distro has jumped the XFree86 ship.
You should perhaps follow your own advice.
Alan Cox went on record of having withdrawn his contributions to XFree86-4.4-rc3 because he was not consulted about the licensing changes in advance. Those contributions are now part of the X.org distribution.
Alan Cox went on record of having withdrawn his contributions to XFree86-4.4-rc3 because he was not consulted about the licensing changes in advance. Those contributions are now part of the X.org distribution.
What’s your point? What you just said has no bearing on the fact that the anonymous poster was completely incorrect when claiming that David Dawes changed the license on code contributed by Alan Cox.
Adam
The anonymous poster was *not* completely incorrect.
“Dawes is not the only code contributor. Alan Cox, along with many other OSS developers contributed code too & they contributed code to XFree86 under a certain license, which Dawes changed, without even bothering to ask the other contributing developers how they felt about it.”
IANAL. But the way I understand it-correct me if I am wrong, is that each code contribution is releseased by the author with a specfic license and then as part of the whole of the XFree86 distribution it is (also/again)sublicensed under the official XFree86 license. When Dawes first announced his license changes(which were effective immediately) these changes changed the license of the entire distribution. At least this is what everyone thought at the time. Later when pressed Dawes started to delimit that to which these changes applied, finally reversing his earlier decesion as regards the Xlibs.
The salient point in the comment by the anonymous poster was Dawes changed the license “without even bothering to ask the other contributing developers how they felt about it.” That point is true and it is perhaps the most important point in the breakdown in trust which took place-as evidenced by Alan Cox’s withdrawl of his contributions.
Perhaps we are splitting hairs here. The gist of what the anonymous poster posted was right on the money.
“The anonymous poster was *not* completely incorrect.”
No, Adam is right. I was incorrect. I was under the impression that the license for ALL XFree86 code had been changed, no matter the developer. I see now how ridiculous this is, given that no one can change the copyright on a work, except for the copyright owner. I had followed the XFree86 license change story when it first hit. I never followed up on it as it progressed. So, my information was incomplete, thus the statement I made in that post was COMPLETEY incorrect. So Adam is right.
ok-glad things are cleared up now.
…is that no one can hijack a project. Dawes made a mistake and was stubborn about it. His autocratic style has created a more than viable alternative, one that has a more open structure and therefore should progress more rapidly than the ossifying XFre86. End users will be the overall winners.
@A nun, he moos
I never said forking was pissing on the previous developers. I said doing childish things such as calling it X!!NOT!!Free86 or claiming it’s non-free is childish. The historic code and program name is XFree86. If you go making fun of XFree86 as a whole, unless someone is deeply familiar with XFree, they will think you are talking about ALL of XFree, past and present.
Uh, no. The notorious advertising clause makes it incompatible with the GPL.
I said this, but I guess you didn’t read my post. That does not make it non-free. According to the four freedoms of free software you people worship, it looks to me to be free software regardless of compatability with the GPL.
I’m sorry, but considering how often you’ve attacked the GPL in these comments section, it seems to me you’re only taking advantage of this issue to push your own agenda. You are conveniently ignoring the fact that a lot of past XFree86 contributors (including some key individuals such as Alan Cox and Keith Packard) have approved of the fork.
And what agenda would that be? To attack the GPL? Wrong. I’m simply trying to stop GPL supporters spreading FUD about non-GPL projects. I am not ‘conveniently ignorning’ previous contributors. In fact, that was exactly my point. I focused a bit on Dawes because he’s being unjustly attacked. He may be a bad leader, but he still gave more than most people on this forum. But again, you didn’t even read my post, did you?
@karl
I read all of what you wrote and didn’t bother responding to most of it because I knew most of it, put it down as one of the reasons for the fork in my first post, and it has nothing to do with my complaint against you and a few others who seem to be making fun of the XFree86 project.
There is no one who used XFree86 who did not know where it came from and who the people were who were developing-there was never any kind of lack of “brand recognition” or other such nonsense. You would be hard pressed to find somebody who thought Redhat or SuSE wrote “that graphic thingy which puts windows on my screen” in the Linux world.
This is exactly my point, there is not lack of “brand recognition”. People recognize XFree86. When you claim XFree86 is non-free (false anyways) or use your silly little word play, you end up including what people recognise as the ‘historic XFree86’ as well as the current XFree86.
The truest form of payment in the open source world is mutual recognition and respect. I respect what David Dawes and the other XFree86 developers have done, past perfect tense. I no longer respect the Xfree86 project.
That’s fine and good. If you had posted your second post first, I would have mostly sided with you. Dawes is a bad leader yes. But your silly word play did *not* get your point across. It did not voice your opinion. It was vague, and in the context of previous posts claiming that the XFree86 project is non-free, could easily be misinterpreted. Also, for those who do not follow the politics of the situation, it could easily be interpreted as attacking the historic XFree86 which is still in use in many older distributions such as Debian and RedHat enterprise. Hence, it I still stand by that it is childish.
By all means, voice your opinion. But do with with sentences that give information, not silly little quips that no one but you and possibly your friends can understand.
Also my original post on licesening still stands for those claiming the current XFree86 is non-free. It may be politically encumbered, but it is still free software, and the current people working on it still give their time freely. There is no need to continually bash them. They may have been and still be bad leaders, but they are not out to attack free software, they should not be so shunned. I’m sure they’ve written more free software than I have, and more than most posters here too.
According to the four freedoms of free software you people worship
Worship? Aren’t we a bit dramatic?
For the record, the “advertisement” clause has proved so problematic that it was dropped from the BSD license.
It’s not a matter of worshiping freedoms, but of agreeing with them on principle. In the present matter, the new license makes it quite problematic for distros to include XFree86 because of its new license. It’s quite a pragmatic issue, and has nothing to do with zealotry. Your condescending attitude only serves to illustrate your agenda.
And what agenda would that be? To attack the GPL? Wrong. I’m simply trying to stop GPL supporters spreading FUD about non-GPL projects.
You’ve often been critical of the GPL in the past, but at least you admit that you have an agenda. Trying to spin this into yet another BSD vs. GPL discussion is quite pointless when the new BSDL does not, in fact, have the advertising clause which is at the heart of this matter.
You’re right that the advertising clause doesn’t make the software “non-free”, but it makes it incompatible with the GPL and complicates things for distro makers. In other words, it was an arrogant, autocratic decision from Dawes and the fact of the matter is that he has sabotaged his own project’s future.
Finally, you should really consider that while XFree86 is free software, its development process was inefficient because of its lack of true openness. There are worst offenses towards the “XFree86 brand” than karl’s wordplay (which, though inaccurate, did not mar an otherwise insightful argument); Dawes’ and the XFree86 board’s bad leadership has done much more to sulley the XFree86 name than anything anyone here could say.
Well I knew that this would happen. I’ve upgraded all of my Linux boxes which had the XFREE 4.X (and some 3.X) to X and I’ve had no problems whatsoever. The moment the XFREE86 organization changed its license was the day I knew that XFREE in its current form would be over. I wish all the luck to the X Organization.
The only way the installer for slack would be too cryptic is if your a n00b( in this case you get cut tons of slack. pardon the pun) or you cant read.
I dont care which way Pat goes with the graphical portion of slackware….he’s nerver mislead me before.
For the record, the “advertisement” clause has proved so problematic that it was dropped from the BSD license.
I realize that and I am not endorsing the license change nor the advertising clause. I’m only pointing out that XFree86 is in fact still free software.
You’re right that the advertising clause doesn’t make the software “non-free”, but it makes it incompatible with the GPL and complicates things for distro makers. In other words, it was an arrogant, autocratic decision from Dawes and the fact of the matter is that he has sabotaged his own project’s future.
At the same time, Dawes did remove the license from parts of X that would cause problems with GPL compatability. You can legally run any GPL based X program with XFree864.4. The license only covers code very specific to XFree86 itself (ie not in any library linked with GPL programs). So it’s no where near as dramatic a problem as people make it out to be. This doesn’t make it a good thing, but it’s not the end of the world.
There are worst offenses towards the “XFree86 brand” than karl’s wordplay (which, though inaccurate, did not mar an otherwise insightful argument); Dawes’ and the XFree86 board’s bad leadership has done much more to sulley the XFree86 name than anything anyone here could say.
Except that karl didn’t make any insightful argument UNTIL I complained about the silly wordplay, which in the context of previous posts sounded like it was suggesting that XFree86 is not free software, which is a very innaccurate statement. If karl wanted to make a point about X leadership, that should have been his comment. Not silly wordplay which tells us nothing and can *so easily* be misinterpreted in so many ways.
“Now, on that topic. What’s this nonsense about no major linux distros ‘using’ it?”
With ‘using’, they mean new versions of the OS. Currently, you are right, however as planning goes ahead it is as stated. If you want to look at the past (older versions) then RedHat is “using” Linux 2.0. So i suggest that when a distribution has made a new version, we agree they are “using” it after a while. That’s all.
“This begs the questions”
No it doesn’t. (It isn’t a fallacy.)
“Just to clarify for you and the rest of the immature crowd, XFree86 *is* free software. It’s just not compatible with the GPL.”
Just to clarify the nonsens coming from this regular troll, it contains an advertising clause similar to the original (not the revised) BSD advertising clause. This is considered non-Free, and it makes a mess of forking forks.
To fix the GPL incompatibility the xlibs (X Libraries, needed by many X programs which are GPLed) do not have this advertising clause so one could argue those are Free software. All the other new additions from 4.4rc2 and higher are not Free since they have the advertising claus.
Just to clarify the nonsens coming from this regular troll, it contains an advertising clause similar to the original (not the revised) BSD advertising clause. This is considered non-Free, and it makes a mess of forking forks.
Sorry, but you happen to be wrong. Check http://www.gnu.org/philosophy/free-sw.html for the definitions of free software. Also check here http://www.gnu.org/philosophy/license-list.html#GPLIncompatibleLice… and notice that the ‘original BSD license’, the one with the advertising clause, is listed under ‘gpl-incompatible free software licenses’. It is *not* listed under the non-free license section.
The advertising clause may be stupid, but please stop spreading FUD by calling it non-free. Call it stupid if you want, but it most certainly is free software. Sounds like your the troll here. It’s funny how you only go by the definitoins when it suites your fancy. At least I do not spread lies, only opinions you don’t like hearing.
Anyway, it isn’t only GPL incompatible like you want to push it. It isn’t incompatible with other licenses either, and it is also a legal, inconvenient virus which leads to long lists of forcable authors while such is currently done but not explicitly needed.
Here’s a /. post which describes the XFree86 path, accurately (enough) imo:
the XFree86 leadership has been doing a lot to piss off developers, avoid enhancements, and turn XFree86 into a stagnant project for quite a while now.
damn stright! I read the Xfree mailing-list around the time when Keith Packard was kicked out of Xfree. David Wexelblat was flaming Keith like there’s no tomorrow. Now, Keith is just about the best thing that has happened to X in a long time. He was the one who made all those cool new features (like RENDER-extension). He was the one who was driving the developement of Xfree forward. And they kicked him out.
Who is this Wexelblat-guy who was flaming Keith? He’s one of the guys who started Xfree and a member of the core-team. By his own admission, he doesn’t hack Xfree anymore. He doesn’t even USE Xfree anymore. He said that he uses Windows these days. Only X-related thing he does is that he lurks in the mailinglist.
Keith Packard gets kicked out, while useless deadbeats like Wexelblat are member of the Core. I’d say the sooner Xfree dies off, the better.
You don’t have to convince me, I do agree with you. And I agree with the fork for technical reasons, X needs new features. The old X team seemed incredibly slow at adopting new technology.
I was only getting upset over people doing two things: claiming XFree86 is somehow not free software and seriously bashing XFree86 and Dawes as if they are the devil and his minions. It’s still a free software project, the people involved still give their time freely. All in all, they are still being kind and generous people. They may have made many bad moves, but they are still creating free software and their hearts are in the right place. Bashing them so much is entirely uncalled for.
And it is working great.
From what I’ve gleaned from the pages of responses here, it shouldn’t have made a big difference, but on my machine (Slackware -current running on a K6-2 500 w/384 MB RAM) KDE seems to look a bit better and it feels a little faster than before.