XFree86 Core Team Disbands

mumumu was among the many to write with this news: "XFree86's release engineer David Dawes has announced that "a majority of the XFree86 core team has voted in favour of my proposal to disband the core team". XFree86's News Headline has a short message about it. Why, all of a sudden? What is the successor of the XFree86? Xouvert? freedesktop.org?"

Why a successor?

[__past__]

Why would a successor for XFree86 be needed? As I understand it, this is only a change in the "political" structure of the project, not its end.

Core Team Disbands

[Anonymous Coward]

Sounds more like the "core" team weren't actually doing the development anymore, and that they felt it was unfair to be the "core" team when they weren't doing the work.

Nothing to see here folks, keep moving.

Re:Why a successor?

[Anonymous Coward]

That's what I think too.
I think Xfree86 will remain the X11 implementation of choice.

Re:Why a successor?

[Kethinov]

If the core developers stop working on it, innovation, for all intents and purposes, stops. Not that there's been anything all that innovative in xfree86 over the last few years anyway. I think this is a good thing. Gives xouvert (french, stands for x open) and freedesktop.org a chance to really shine. (Hint hint, be more like osx.)

Re:Why a successor?

[Anonymous Coward]

Because, basically, every software project needs to evolve or it will die. And there is a lot of room for improvement in X11 ! Apple has developed a very nice system (Quartz) and even Microsoft is constructing a very modular and IMO quite interesting Avalon system. There are some good techniques in there that will benefit the entire X11 community.

Re:WOW

[Cyn]

read.
the.
exceedingly.
short.
article.

"Core Team" models need to die.

[mcbridematt]

Face it.

"Core Team" Development models are out-dated and sound more M$'ish than Open Source'ish.

While several projects continue to use the "Core Team" model, like FreeBSD, in my opinion, the politics involved ain't worth it.

For XFree86, it's time for change. Hopefully, in years to come, we will see a more efficient graphics subsystem for Unix (MacOS X may be an example) weather it be by a XFree86, XF86 Fork, or some other system (NOT framebuffer because fb doesn't work well with some hardware)

Slashdot trolling?

[rpdillon]

After comparing the /. headline with the actual content of the email, I wonder what exactly /. *does* check on before they post these...I feel like they're trolling for a bunch of misinformed readers to overreact.

It may be newsworthy, but considering the length of the message, why not just post the original email and be done with it?

Re:Why a successor?

[Xtifr]

Yes, and when it says, "the core team was no longer representative of the active, experienced and skilled XFree86 developers," it actually sounds like they might be opening up the project a little more, rather than disbanding it. Given some of the negative comments I've heard in the past about the rigidity and bureaucracy of core team, this could well be a very good thing for XFree86 overall.

Re:Core Team Disbands

[GAVollink]

"keep moving" aside, I actually do believe this to be a bad thing. While the core team was not active in the development they did still help steer direction. These are the folks that would say, "that will break things" - when it otherwise may not be obvious that "n" change could break things. This is a loss of experience, but the core team obviously feels that there is enough checks and balances to keep things from breaking.

Re:Related to the Cygwin blowup?

[Anonymous Coward]

I think it's related to the "firing" of Keith Packard from the core group, when he was one of the few people trying to move X11 into the 20th century.

Re:Why a successor?

[cshark]

I really wish these kinds of announcements were a little less ambiguous. Judging by the post, we know the core team is disbanding. Great! Now what?

There is nothing in it about the future of X86, which would be mine and many others big concern.

It's all Slashdot speculation right now. Unless someone can provide us with more information on the subject.

Any Xfree86 developers out there?

read the "Insightful" article

[jbeamon]

I'm going to forego the opportunity to use my moderator points today on this story because every odd-numbered post in the list is already "Score:5 Insightful". There's just a wealth of wisdom here, and I have precious little to add.

In all fairness to those who questioned the future of X, I was momentarily confused by the announcement, too. It appears this little group of developers has finally just gotten out of the way. I'm hoping there's still a person or two to moderate code additions while the rest of the community keeps up the project.

Re:Full text of email & analysis.

[NDPTAL85]

The abundance of abandond projects on Sourceforge would appear to disagree with you. Open Source projects are usually NOT the domain of hundreds and thousands of globally diverse developers but rather a very small and very active "core" team. Once that core team leaves (and I'm not talking about XFree86 here) then the project usually dies. Why? Because even though the code is in the public domain there is a lack of willingness to get involved or a lack of skill of those who ARE willing or a lack of time for those who are both skillfull AND willing.

Don't necessarily blame the product..

[Watts]

Any switch can blow up if a transition isn't well-planned or done with the right expertise. In my experience, major shifts like that require a lot of training of existing staff while bringing in a number of consultants that are fluent in the new tech.

True, choosing a product that is a poor fit will make it blow up in your face, but that doesn't mean that sticking with the old code forever is the answer.

Besides, we use SAP at my place of work and are pretty damn successful.

Re:Bumpy times ahead for XFree86 users?

[Anonymous Coward]

This guy shouldn't have been moderated down. XFree HAS forked, so fragmentation could be a real concern.

> make it hard for driver vendors like NVidia to target XFree86's derivatives as a platform

XFree has a standard "driver model" that they use, and dislike of that actually one of the things movtivating the forks. So, new X servers won't use the same drivers, but the argument is that it will be easier to implement better drivers without being hamstrung by backward compatibility.

> Unless they agree on an API or similar framework

That API is X11R6 + extensions. Ultimately it matters little if you use XServer1 and I use XServer2 -- the difference wouldn't be visible to end users except maybe with some eyecandy features like transparency effects.

This is a far less serious problem than (say) KDE versus Gnome, which affects the end user in all sorts of ways, but yet people manage to survive.

Re:Doesn't really matter. Quality stays for years.

[aled]

False. Enterprise financial apps don't depend on changing hardware every year like graphics applications. And "just plain works" doesn't mean is maintainable. And I would doubt very strongly that someone knows 30-year-old-multi-million-lines-apps of financial code in Fortran well enough to be sure that it does what it is supposed to do...

Re:Jesus.....Thank God.

[Bowie J. Poag]

Hey, just sharing what I know..Jesus. Go and throw a hissy fit why don't you.

I spent alot of time with the Xouvert crew. From what I understand, Xouvert was formed largely out of this same frustration -- Neither developers nor companies could even get a word in edgewise with them, with means the whole project sits and stagnates... Well, until things like today's event, that is. :)

The core team dissolving is a good thing, as I see it. It clears the way for XFree to be less Cathedral and more Bazaar.

Re:Why a successor?

[hankaholic]

There is nothing in it about the future of X86, which would be mine and many others big concern.

Did you read the post? It basically said that the people involved in the "core team" aren't the ones driving XFree86 development.

Given that statement, why would you ask them to describe the future of XFree86, which is something over which they explicitly announced that they don't have control?

Re:When open source dies?

[grumbel]

Looking at Xouvert and XServer there are quite a few people interested in maintaining and continuing XFree86. As far as I can see it the whole old structure of XFree86 and the Core-Team was one of the main issues holding the progress back, making it extremly difficult to get code merged into the core tree and such. Sure it will take a while until the dust has cleaned up, but a big clean-sweat is really needed for XFree86, its IMHO one of the main issue that hold GNU/Linux as a whole back from moving onto the cassual users desktop.

Re:This is THE most irresponsible post I have ever

[Skjellifetti]

Yea, like most CIOs get their tech, political, and business news from /. You grossly overestimate /.'s influence.

Re:Core Team Disbands

[Firehawke]

Personally, I'm a little more cynical about the core team-- overall core team competency has been questioned of late, resulting in several branches of the code. I'm not so cynical to call them incompetent outright-- I've no experience with them directly, so how could I say such, but in either case they've decided to let things go in the direction they have.

Now we just need to see how the structure holds up and see where the actual 'power' in the organization is going to be. In plain english, to see who's going to be OKing the executive decisions now.

Re:"Core Team" models need to die.

[stienman]

"Core Team" Development models are out-dated and sound more M$'ish than Open Source'ish.

While several projects continue to use the "Core Team" model, like FreeBSD, in my opinion, the politics involved ain't worth it.

Uh, say again? Are you saying that open source software favors one political structure over another?

So if a core team is bad, what about Linux with essentially a technical dictator^H^H^H^H^H^H^H^Hdirector? And do you believe the MS uses a core team to direct development of their software? They have a simple hierarchy, like most succesful businesses.

While it may be warm and fuzzy to say that open source == no core team, the simple fact is that different political structures are good for different projects during different phases of their life. Linux has gotten too large to be developed by a single developer, so Linus has changed the political structure to fit his needs.

Furthermore, this doesn't mean the end of the core team for XFree, only the end of a core team. They haven't spelled out a change in structure, only a change in personel.

-Adam

Re:"Core Team" models need to die.

[iabervon]

The framebuffer console kernel project has too few developers for the variety of available hardware, and the developers are (from my reading of the mailing list) not especially well organized in their interactions. So a lot of hardware seems to lag behind in API changes and get subtly broken in various ways. That might give this impression.

Also, it's relatively difficult to specify a 3D API, so that you can use hardware rendering, particularly because it means that you have to do 3D in software in the same layer for anything that doesn't support some particular operation (or some set of arguments), which means that you're going to have to do 3D in the kernel, which nobody would be happen about.

Of course, the kernel is really the right place for device drivers, and good device drivers should make the differences between devices irrelevant (aside from performance and output quality), but anything complicated with graphics cards is such a mess that it hasn't been moved to the kernel.

Actually, one thing I'd really be pleased about if the dissolution of the "core team" brought it about: people writing drivers for XFree86 could move to writing them for the kernel instead, and XFree86 could do active development on both sides of the userspace/kernel division. (This is important for things like having the 3D rendering state of your graphics cards restored correctly after suspending to disk).

Re:Full text of email & analysis.

[Tassach]

The abundance of abandond projects on Sourceforge would appear to disagree with you

Do you think, perhaps, that the reason there are so many abandoned projects is because they suck, are irrelevant, outdated, or duplicative? How many half-assed winamp clones does the world need? Do we really need 2000 different email clients, or yet another piece of desktop eye candy?

Open source development is a Darwinian process. The strong prosper and the weak either die off or adapt themselves to survive in an isolated niche. If a project is so uninteresting or so obscure that it can't attract a new maintainer, then it deserves to die. The carcass remains part of the ecosystem -- scavangers are free to pick the bones for anything useful, or someone can come along and breathe new life into it.

Re:"Core Team" models need to die.

[einhverfr]

"Core Team" Development models are out-dated and sound more M$'ish than Open Source'ish.

What do you propose as a replacement? If politics is the issue, I guess you can have a single maintainer (ala Linux). Or are you seeing the corporate control of open source (ala MySQL) as a better alternative?

If democracy is what you are looking for, are you advocating public CVS commit privilages? I certainly hope not!

The core team's job is to design and develop the next version *with the help and feedback of the community.* If they don't listen to that feedback, then there is a problem, but a good core team does the development and facilitates community involvement.

FOr example, PostgreSQL uses a core team model. It is very successful and LIGHT YEARS ahead of MySQL in terms of what capability it has and probably will always be. The core team has been extremely responsive to the needs of the community. THis does not always have to be the case, but it is the way things should be.

Core disbands does not mean the end

[plcurechax]

The disbanding of the current XFree86 core team does not mean an end to the continuing development of XFree86, it means a change of people recongised as being key players.

The biggest remaining question IMHO is whether there will be a expansion of cvs commit access. I think the former core team realises that new up and coming developers need to be added to the project to subtain the continuing improvement and work with others groups such as X.org, and freedesktop.org. To say nothing of expanding access to video card manufacturers so they can maintain and improve open source drivers for their cards (Most companies are at least partial supportive of 2D drivers, the real issues occur over 3D accelation).

I expect it will end up being a good thing.

Re:Why a successor?

[Anonymous Coward]

Considering how ambiguous the release was, to most people...

Um... which "release" to "most people".

This was not a press release or anything intended for "most people", it was a message written for the XFree86 developers mailing list. Most people on that list will know the context.

Now it happens that someone has taken that message, made a misleading comment, and Slashdot has published it to a general audience, but that's not the fault of the message author.

Re:Bumpy times ahead for XFree86 users?

[Anonymous Coward]

Why should Nvidia and ATI support any kind of X server? Doesn't that just make it harder to produce a driver package? Not to troll, but I couldn't even find a good how-to doc for this on the xfree86.org site. At most just a few hints on bare-minimum functions.

Why can't we shove off the X11 and let Nvidia/ATI implement a raw OpenGL driver? Just let them support their hardware. Let them expose OpenGL, shaders, and overlay/MPEG stuff. Wouldn't it be fantastic not to need a new set of drivers for each XFree86 version?

Then if you want legacy X11 functions, just re-implement them to call OpenGL functions instead. There's no reason for an X line when the hardware is built for OpenGL lines.

In this scheme the X11 is sidebanded (networking) and wrappered (graphics calls) and graphics card driver updates are greatly simplified. Dare we hope it may spur more frequent driver updates as well?

Re:Related to the Cygwin blowup?

[Anonymous Coward]

Do you have any basis for these assertions (for example, the assertion that the problems in Linux video driver support are actually due to weaknesses in the design of X) or are you just being an idiot ? If the former is true, and you actually know something about writing video drivers, could you please share with us the basis for these assertions ?

Re:Full text of email & analysis.

[ajs]

You are using your own definition of core here. XFree defined core as a certain set of people. What they have found recently is that the "core" was not in fact at the core of development of XFree, and that those who were were better capable of filling that role. Grats to XFree for having the stones to make that call. THAT is probably the best example of what open source has going for it that proprietary software does not. GCC continues, but companies and steering commities have come and gone. Mozilla continues, but Netscape has come and gone and gone.... Open source endures.

Re:Why a successor?

[coolgeek]

(SMACKS AC upside the head) Looks like a press release [xfree86.org] to me, since it is it the "News" section under "Headlines".

Re:Doesn't really matter. Quality stays for years.

[Chanc_Gorkon]

Actually, 2002 was when the new standard went into effect. COBOL Standards [cobolstandards.com] has the link to the official ISO standards regarding COBOL. Also, COBOL is not dead. Far from it. COBOL can even be used to develop for the web as well. We are using Miami Univeristy's (of Ohio) DARwin product and it's entirely ported to run with Micro Focus COBOL Libraries/compilers and on multiple platforms including UNIX. DARwin, or Degree Audit, allows students to play a what if game comparing what they have taken with the program they initially intended and other programs. This allows students to see what else they have to take in order to get their degree if they want to switch majors. I have not seen software like it anywhere else. Most of the schools in Ohio use it. COBOL still has lots of use in it and it's just morphed in to doing different jobs.

The same wil go for Xfree. So what? The core has disbanded....long live the new open core.

Re:Why a successor?

[Sleuth]

Well, not exactly, since XFree86 also has a board of directors. Interesting, eh?

Congratulations

[johnos]

I think we should congratulate the core team for doing the right thing. Its pretty rare for any institution to volintarily disband no matter how irrelevant it becomes. I can think of a few institutions a lot less relevant than this group that have continued plugging along for generations.

These people are showing maturity and class usually missing in the software industry. Just by taking this action, the team has refuted one of the more subtle FUD points out there, that projects will eventually peter out or be consumed by internal bickering.

X Replacements

[localman]

It seems that the tone of this article is misleading; X development will continue on in good health.

However, I always find myself thinking about Y as an X replacement [ic.ac.uk]. It's certainly not the most mature option out there, but reading throught the PDF [ic.ac.uk] is a pleasure, as the author seems to have struck a great balance of power and simplicity.

Cheers.

Re:Related to the Cygwin blowup?

[osu-neko]

This can be worked around. You move to a new core, and you toss out all the backwards compatibility crap. THEN, you add a backwards compatibility layer or module for the code that still needs it. Surely it would be easier for X than it was for, say, Apple, when they made apps designed for System 6 on an m68k processor successfully run under System 7 on a PowerPC processor. Thus you can get your brand new spiffy clean core with all the latest features while retaining legacy support. You just need to lose the idea that legacy support needs to be maintained at the core...

Re:Don't overreact

[Trepalium]

I think the problem was the core team had become too much about politics, and too little about software. Look at the well publicized event about Keith Packard being outed from XF86. Keith has contributed some of the most radical changes to the X server system in recent years (XRender, XRandR, fontconfig, etc). He was outed because he dared to try suggest to others that there should be a new project started to create a new X server to both encourage XF86 to be more active, and to also try to solve some of the lingering archetectural problems of XF86. It's ironic that he achieved his goals by being kicked out.

Re:WOW

[Trolling4Dollars]

That's what is wrong with the support industry. Lazy asses who don't take pride in their work and have little or no patience. Way back in 94 when I was a Windows n00b, I used to call support and was appalled at how bad it was. Adaptec, S3, Microsoft... they all sucked. Within a year of using Windows (originally a Mac and Atari/Amiga guy here) I became my own support because I knew more than any of the jackasses on the other end of the phone. What really used to irk me was when I would call and have to walk through all the crap solutions that they were reading from a database even though I'd already done all of that. I'd tell them I already knew the outcome but they wanted me to do it anyway. Then in the end they would say "fdisk, format and re-install". Of course I wouldn't do that. I would dig up the info online or form other users and eventually solve the problem. That's when I realized that the fools that work support are just plain lazy, but your comment cinches it.

I deal with stupid users every day where I work, but I still help them even if it's the upteenth millionth time. I don't expect them to understand or to know what to do. Face it. Computers are STILL to hard to use for the average user because they are very complex machines. The only people that have a prayer of being able to use a computer to it's full capabilities are people who are very good at deductive logic and can understand abstractions easily. This is NOT the average human being. It's probably only about 10% of the population. About 75% would be the people who learn by wrote. They just know what buttons to puch when, but don't know why. So people and computers are meeting halfway, but when something doesn't work, it all falls apart. Face it... computers STILL suck for the majority of the population. (Note that the other 15% I didn't mention are comprised of both the guys with the bulbous heads who can do advanced physics calculations at 245 MIPS in their wetware and the other end of the spectrum with the small cranium that has trouble turning on a light and votes for George W. Bush because that's what TV told them everyone else is doing.)

Anyway... I guess a big part of the "American way" (these days read that as the way capitlist countries act) is to be lazy. Make millions while you sleep, yadda yadda...

Re:The rise of project K12 : XouverK

[be-fan]

We are all looking forward to your patch. Just post it to bugs.kde.org and a maintainer will get right on integrating it. Do you think you can have it done before the KDE 3.3 string freeze? The translators need some time to do their work. Oh, and make sure you don't break any of the Qt themes --- the theme developers have enough work to do as it is, dealing with the new KWin API, and Qt 3.3 API changes. At least, don't break Plastik, because that'll probably be the default in 3.3.

Thanks!

Re:http://www.directfb.org/

[chromatic]

... and quickly drop that pesky cross-platform portability it enjoys right now, at least until someone ports the Linux framebuffer device to non-Linux systems.

Re:Why a successor?

[abradsn]

Open standards can lead to better / more accomplishments. Our nature is to want what the other guy has. By that alone, help can be garnered from competition.

Re:Bringing it into the 20th century?

[HeghmoH]

I love how people always compare things to Windows, which is the most backwards, un-advanced OS ever.

If you want to compare GUIs, compare with Mac OS X. OpenGL-accelerated drawing? Check. Incredibly rich graphics in apps? Check. No need to wait until 2006. And of course, by definition, right now X is still where it is now.

If Linux always strove to play catch-up with Windows, it would be horrible. Fortunately, it doesn't do that, except in the area of the GUI. It's no surprise, then, that Linux's GUI isn't very good.

Re:maybe we need a new X server (or two)

[penguin7of9]

The problem with using C++ because it has resource management is that assumes the resource management model that C++ uses is appropriate for the problem space.

C++ doesn't "have resource management", it has standardized hooks for implementing whatever storage allocation and resource management strategies you want.

You wouldn't use C++ resource management for buffer cache handling in an OS, would you?

Using C++ would give you identical performance to what kernels currently do in C, yet it would greatly reduce the risk of bugs.

In any case, more generally, I see no problem using languages that actually have resource management built in. Some of those are badly designed or inefficient (VisualBasic, Java) and are therefore unsuitable, but others are perfectly fine (Modula-3, C#). Automatic resource management (garbage collection, etc.) is almost always more efficient than anything C programmers do by hand, and it is far less error prone.