Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×
X GUI

XFree86 joins X.Org as Honorary Member 93

Virtex writes "According to a press release at X.Org, the XFree86 group has been made an honorary member of X.Org, which is an organization of the Open Group. This will allow them a seat at hearings and the ability to vote on the executive board. "
This discussion has been archived. No new comments can be posted.

XFree86 joins X.Org as Honorary Member

Comments Filter:
  • Wow. So all those years spent developing XFree86 just to keep a free (as in Open Source free) X implementation was worthwhile. Which raises an interesting question... does this mean that it's worthwhile to insist on creating free clones of proprietary software, or should we concentrate on creating new, Open Source software (ie., not cloning existing software)? If it's worth the time and effort to clone something just for the sake of keeping it Open Source (ie. eventually we reap the benefits when the Open Source development model overtakes what was being cloned at first), then perhaps we should start cloning things like QT and a host of other things with licenses that people are complaining about...

    I used to think that efforts like OpenSSH and XFree86 are doomed to play catch-up forever. But now I think I was wrong... it may not be such a bad thing to clone. Perhaps the Linux story generalizes to other things as well -- after all, Linus started by cloning UNIX to run on i386, and look at what Linux has become now.

  • You make a good point - X has got some serious problems. I think that pushing stuff like DGA or (what was that other thing? GGI?) seems a better idea than just scrapping X altogether and Doing It Right, because I don't think the Collective Mind of Unix can support such a fragmenting action.

    There's already enough banter about fragmentation of the desktop what with Gnome and KDE and some other distant also-rans. I would think that scrapping X would merely serve to dilute Our (Unix Community, here) Application Base.

  • Hmmpph, is it just my own myopic vision, or does the XFree86 project seem to be the only organization actually DOING something new with X?

    Sun's basically dropped the ball at X11R5 and The Beast That Is CDE, HP does - well, nothing, so far as I can tell. Digital is effectively in the midst of death shudders under the Loving Hand of Compaq, IBM - look, who else is actually rolling out new servers and working on X?

    The members of X.org ought to pony up to XFree86's bar, and get X going again (and thus, their own miserable stinking excuses for Unices, May the Hand of the Moderator be Gentle).

  • 2 years ago the Open Group declared X would be licensed only for a fee and the XFree86 team would no longer be able to use the Open Group's code. Now not only did they reverse that decision but they're really trying to pick their balls off the floor with this XFree86 pledge.
  • I would have to respectfully disagree that XFree doesn't need X.org. Only because the vast majority of developers insist that the X standards be adheered to - including XFree developers. The catch-22 occurs because there has never been a feedback mechanism back into the SI from XFree, so anything that XFree code-forked would have to be re-implemented on new X releases. Not a big deal on the driver side.

    What was bad about this situation, is that when people start to talk about innovations that require major changes to the SI, we must develop them using extensions for fear of breaking with the SI.(Some things just can't be done using extensions, though) When people start using extensions then traditional application developers are weary of using them because there's little reason for workstation vendors to support these extensions. Most developers always aim for the square middle if possible.

    This all changed because X.org realized that the commercial ISV had virtually disappeared. Almost all the commercial work being done was for in-house development and narrowly focused niche markets. The other 99.9% of X developers were using linux and *bsd!!!

    I think some innovations have been (until fairly recently) stiffled because of X.org's unwillingness to accept feedback. Features that are just now being implemented or discussed that could have happened 2-4 years ago. (Insert laundry list here)

    Another factor is that most X developers really thought X was dead. Heck, even Jim Gettys told me at the last ALS that he was really supprised that X was still around (and pleased that it is!). Knowing how "well documented" the SI is (remember that X is larger than the Linux Kernel?) there was quite a loss of real expertise on X internals.

    What you see here is very exciting. You see the ability of XFree to affect the SI. You see a standards organization accepting the true standard. You see a resurgence of expertise in developer resources on the XFree team. You see organizations supporting *very* good things like Precision Insight, SGI, Mesa, and MetroLink.

    And you ain't seen nothing yet.

    This isn't your mama's X server!!

    Roger
  • I would have to respectfully disagree that XFree doesn't need X.org. Only because the vast majority of developers insist that the X standards be adheered to - including XFree developers. The catch-22 occurs because there has never been a feedback mechanism back into the SI from XFree, so anything that XFree code-forked would have to be re-implemented on new X releases. Not a big deal on the driver side.

    What was bad about this situation, is that when people start to talk about innovations that require major changes to the SI, we must develop them using extensions for fear of breaking with the SI.(Some things just can't be done using extensions, though) When people start using extensions then traditional application developers are weary of using them because there's little reason for workstation vendors to support these extensions. Most developers always aim for the square middle if possible.

    This all changed because X.org realized that the commercial ISV had virtually disappeared. Almost all the commercial work being done was for in-house development and narrowly focused niche markets. The other 99.9% of X developers were using linux and *bsd!!!

    I think some innovations have been (until fairly recently) stiffled because of X.org's unwillingness to accept feedback. Features that are just now being implemented or discussed that could have happened 2-4 years ago. (Insert laundry list here)

    Another factor is that most X developers really thought X was dead. Heck, even Jim Gettys told me at the last ALS that he was really supprised that X was still around (and pleased that it is!). Knowing how "well documented" the SI is (remember that X is larger than the Linux Kernel?) there was quite a loss of real expertise on X internals.

    What you see here is very exciting. You see the ability of XFree to affect the SI. You see a standards organization accepting the true standard. You see a resurgence of expertise in developer resources on the XFree team. You see organizations supporting *very* good things like Precision Insight, SGI, Mesa, and MetroLink.

    And you ain't seen nothing yet. This is just my opinion, ofcourse. I havn't been involved until recently (and in a very minor fashion) But I've been watching XFree since they were X386.

    This isn't your mama's X server!!

    Roger
  • DOOHH!! Sorry about that. I'm using Mozilla M11 now.. looks like I found a bug! (Hitting the submit button twice?)

    Roger
  • X is probably the best reason to use Linux on the desktop. I store all my porn on one machine and then have it display running on my 486/33 in my room with a projector. Its fun. Seriously though, X makes rmeote administration practical. I advise anyone who hasn't use X across their LAn for displaying stuff to do so.

    BTW:: What is the difference between X.org X and XFree86?
  • I quote:

    "XFree86 possesses considerable engineering talent, innovation and energy that X.Org is
    well positioned to leverage...."

    Doesn't that just sound someone in marketing who's just discovered something that'll save their bacon.....


    Score: -5 (You're such a cynic!)
  • What many people are missing is the significance of "Honorary Membership".

    Membership to the X Consortium is /far/ from free. The Open Group has given XFre86 a voting seat on the X Consortium for /free/. That is the significance of this move. AFAIK, this is a first... I believe all other members of the X Consortium are paying members. (Someone correct me if I'm wrong)
  • Progress is still progress. Let's not be too upset when we are making strides forward, even if it is not as fast as we would have liked.

  • after all, Linus started by cloning UNIX to run on i386, and look at what Linux has become now.
    Yeah, Intel's worst nightmare. Why? Sure, it runs well on Intel chips, which is good news for Intel, but, unlike MS-Windows, Linux also runs well (or is learning to:) on everybody elses chips: not so good for Intel. Due to the nature of Linux (GPL), there is nothing Intel can do to stop Linux from running well on other hardware, other than make it difficult for developers to get hold of alternative hardware. Linux is not as good for Intel as Windows has been.

    All that said, I know what you mean. Linux started out as a clone and is now becoming a respected member of the Unix community and one of the few with any real active developement (or so I've gathered).

  • ...how many misconceptions are out there.

    For one, what goes in XFree86 does not become a standard. X.Org is still a governing body for the X standard. I think it was Metro Link who initiated the entire thing of inviting the XFree86 to become an honorary member.

    Also, when the X licensing policy changed with R6, it was Xi who pushed for the change in policy in hopes of isolating the XFree86 and hindering their ability to comply to the standard. We all know what a fiasco that turned out to be.


  • and the link works just fine.
  • I actually meant to say something like this, but I forgot to. The reason Linux is Intel's worst nightmare is Intel is damned if they do and damned if they don't. With the rising mainstream popularity of Linux, Intel will rapidly loose sales to other chip makers as people realize Linux works well on non-Intel hw, especially if Intel Linux starts to lag. However, many improvements in Linux for Intel hw will, in general, also be benneficial to other platforms.
  • The other part of the problem that while Motif was cloned for a good reason (a lot of software, both free and proprietary, uses it), there is no possible way to convince free software developers to clone CDE unless some of them are hell bent on passing the certification -- CDE is inferior to free software (KDE, GNOME, any decently configured window manager other than twm), and isn't mandatory to run any existing programs.
  • I've been programming Motif for a couple of years, ever since Sun dropped OpenLook and my then employer switched to Solaris from SunOS.

    Motif has an intersting history, as it was partially funded by none other than Microsoft. They saw it as a useful way of testing UI design prior to the big overhaul to Windows 3.1 that resulted in Windows 95.

    As I've always been a little bit partisan about Motif (XView was far easier to program with), I've often wondered if Motif is best seen as an Windows 95 GUI prototype. Motif certainly needs an overhaul in places - the file dialogs for instance.

    And as a sidenote, I recall a former Open Group programmer saying that they toyed with the idea of making Motif open source a few years back. Damn shame they didn't, as I'm sure both the Lesstif and Open Group teams could come up with something far leaner and meaner if they pooled their experience.

    Chris Wareham
  • But, if someone develops an alternative that exists on a wide range of platforms, enables the productivity that X enables, and runs something as nice as Enlightenment (not to mention WindowMaker/FVWM/etc.), and something as minimal as twm, then you might find me using it as well.
    Indeed. There are a host of relatively-failed graphical systems:
    • MGR, which had a dramatically anti-commercial license
    • Sun NeWS, which was too proprietary when X was less so
    • Smalltalk GUIs, that work iff you're using the same Smalltalk implementation
    • NeXtStep/OPENSTEP, which are no longer deployable because Adobe is reluctant to license the necessary DPS layer (which would also cause problems with Sun NeWS)

    Berlin, BeOS, NanoGUI, and such, all may be interesting options, but until they exist on a wide range of platforms, provide the flexibility of X, and actually have substantial bodies of code that will run on 'em, they can't realistically be considered as more than wishful alternatives.

  • Come on, please stop with the extrapolations - I'm no OSS bigot. I do see that you're right about Sun - should've done a 'man Xsun' before posting.

  • It's not "proper."

    The only reason using just the domain name to get to a website works nowadays is to allow the lazy or the ignorant to leave off the hostname.

    Some hostmasters are old fashioned, and don't provide an A record for the domain itself. Can't say I blame them really.
  • The talk about XFree86 being the standard is that X is _the_ X server that most people use on x86 UNIX machines. Also, a lot of proprietary x86 X servers use the XFree86 code base. Those two facts show that XFree is actually the standard in actuality, even though it is not so on paper. What makes a standard is people who use it and write to it, not some silly document.
  • It's also nice to note that XFree86 is included in Netware 5.0. It's been turned into an NLM. I find it interesting that Netware finally has a GUI.
  • This comment is WRONG.

    You first have to go back to the roots of the X Consortium to understand how things worked. The important standard for X was the protocol, not the particular implementations of the client or server.

    Before the (then independent) X Consortium ran out of money (and sold the remains of itself to The Open Group), it had prepared plans to change the license for the reference implementation to add a commercial redistribution fee. This was to give the Consortium an additional income stream, which until that time had mostly been funded by sponsor dollars and contract dollars. Interestingly enough, those contract dollars mostly came from the Open Software Foundation, which outsourced the development of CDE 1.0 (after having ceased all in-house development of Motif, along with DCE and OSF/1). In fact, many engineers from the Motif group ended up at the X Consortium working on the CDE/Motif contract).

    But, the X Consortium ran out of money and sold it self. Most of the engineers left and a few folks hung on at as part of the X Project Team at TOG. Since TOG had emerged from the OSF and X/Open merger, this meant that some of those engineers now had the chance of being layed off from the same building a second time. Some were (but I get ahead of myself).

    The XPT ran on at TOG with sponsor dollars, but, near the end, still needed more funding.

    The planned change in the X license was resurrected.

    However, to the core developers of X, this was NOT A PROBLEM because the licensing only affected the "reference implementation". As long as clients and servers all followed the same X11 protocol, then the world would be served well by the elimination of a monoculture in the implementation.

    In fact, if you go back to the earlier vendor releases of X (by HP, DEC, IBM and later Sun), you'll see that each had their 'proprietary' additions to the clients and servers, but all interoperated by working against one standard protocol.

    X is intended to be a standard of communication, but a single implementation. Even with the license change, the X protocol would continue to be an open standard. An "open standard" doesn't mean that source is free, but rather the definition of the standard (and implementations of that standard) are unencumbered. OSF was founded on the principal of defining an operating system (OSF/1) as an open standard -- OSF/1 was defined by a book (called the AES) that dictated the interfaces. OSF also provided a reference implementation that was used by others (like DEC) to creatte their own operating system.

    But, I digress.

    At the time the X11R6.4 license change went into effect, several things were cooking. First, a freely distributable source distribution of Motif2.0 was planned. It, too, would have commercial redistribution restrictions.

    However, TOG was in serious money troubles itself. Barely weeks after R6.4 hit the streets, it was announced that the OSF portion of TOG was being shut down. Within two months, almost every engineer in the building was gone. The final move to revert the license was made because at the time, it looked like X.Org was dead.

    (and the Motif release never happened because the people - engineers and management - responsible for X were gone).


    The group that is "X.org" today is composed of the former sponsors of the X Project Team. None of the engineers remain there. Any ongoing work is done by sponsor organizations. If anything, XFree86 is 'the' hotbed of X server development these days.

    To think there is some ulterior motive at TOG in regard to X licensing or X.Org is silly -- the remaining folks at TOG are just the X/Open folks in the UK, and they are just doing what X/Open ALWAYS did ("nothing"?).

    The "they" in "not only did they reverse that decision" just doesn't exist.

    The truth is that the part of The Open Group that was responsible for X (and Motif, and OSF/1, and DCE) is gone and buried.
  • Hopefully they will continue to work together, this can only mean good things for both projects, I wonder if they will share code, and ideas or just vote on the directions they are going in. How much a "part" of the opengroup will they really be.
  • that with XFree86 often being synonymous with X, that they would have been invited long ago, if not present since the beginning of the Linux revolution. I'm disappointed at the lack of recognition that XFree has received, even until now.
  • Always glad to hear about it when Xf86 and TOG are getting along. Let's hope that nothing else goes wrong in this relationship in the near future...

    Anybody got any progress reports on XF86 4.0?
  • Fun facts from the press release:

    Folks funding the X group!

    • Compaq
    • Hewlett-Packard
    • Hummingbird Communications
    • IBM
    • Sun Microsystems
    I'd like to state that I like this list! I think almost all of these companies have made other efforts to promote linux as well. It feels good to know that we've got such nice backing for our X server.

    And my other point, a quote from the About The X Window System section:
    . Further, thousands of software developers provide X Window System applications, and with the emergence of Linux, the number of users is growing exponentially
    I like this, a group with some good backers (see above) publicly stating that their platform is expanding exponentially due to Linux.

    I've never felt better about my OS.

  • Yippee! No more X-clusion for the open source X!

    Now if only they would serve free beer and offer free limo service...man that would be X-cellent! :)
  • I would think that XFree86 has one of the largest installed base of X-Servers. Let's see... Of all Linux and BSD installs which include some kind of X-Server, XFree86 is probably installed on 99.9%.

    Not to be a party-pooper, but maybe the article shouldn't have read "XFree86 joins X.Org," but instead, "XFree86 joins X.Org... FINALLY!"

    Just my observations on the matter...
  • by Enoch Root ( 57473 ) on Wednesday December 01, 1999 @10:58AM (#1489189)
    The amazing thing in this is that XFree86 has not been invited before. I think most geeks wouldn't even differentiate between XFree86 and X. It's like creating a group on Open Source OSes and not inviting Linux people.

    At least people got the wake-up call. It just goes to show that establishments exist even within the Open Source community.

  • I think it should say, 'X.Org joins XFree86'. :)
  • I was under the impression that network solutions disallowed single letter domain names. Is x.org grandfathered in as an older domain, or are there certian exceptions to the rule?

    -Adam

    Best ways to cut a meeting short, #3:
    Your wife is on line 1, your girlfriend is on line 2. Should I just let them talk to each other while they wait for your meeting to finish?
    -- TweetyBert
  • XFree86 has served the open source community by organizing, coordinating, and supporting a massive effort to guarantee open source, freely available implementations of the X Window System. Their work has been the cornerstone of desktop operability on millions of computers worldwide, making them a true industry standard for graphics. It is very heartening to finally see the XFree86 Project be recognized for their tremendous contribution and their years of dedication.
  • X.org was registered before that restriction came into being.

    On a side note, I wonder how long it will be before one of the new registrars decides to go ahead and issue one-letter .com domains for, say, a million dollars apiece?
  • Is it just me or does X seem big, slow, bloated, and old? It can't keep up with pretty environments like GNOME, and there are more incompatible ways to copy and paste than I can count. There are more regulations and standards than anyone cares to read. We need a new windowing protocol that addresses the need of the modern PC user, not the UNIX nerd sitting in front of a 12 inch monochrome CRT. The continued development of X seems pointless to me - I think we need something new altogether.

  • by Booker ( 6173 ) on Wednesday December 01, 1999 @11:28AM (#1489197) Homepage
    From XFree's website:

    Short delay in releases

    [December 1999]

    The XFree86 Project has missed a few promised release dates in fall 1999, but we are very optimistic to be back on track for 3.3.6 and 3.9.17 to be released before the end of the year. Similarly, we expect XFree86-4.0 to be delayed by about two months and are aiming for a release in mid-Q1/2000.
    ----
  • On a side note, I wonder how long it will be before one of the new registrars decides to go ahead and issue one-letter .com domains for, say, a million dollars apiece?

    Can't... they're already registered by ICANN.

    Too bad... C.com is a waste of a perfectly good domain.

  • Convincing signs of a TOG "change of heart" would include things like:
    • Defining UNIX 1999 or UNIX 2000 in such a way that Linux systems could become branded as UNIX, rather than their current comments [unix-systems.org]

    What changes to their current UNIX definition would be required for this?

  • Someone mark this guy up as informative.
  • Gurlia,
    I don't think it matters how a project starts. If it's a clone to begin with, and it's a worthwhile project (i.e., many people can benefit, and benefit more from the project if it is in open source form), then it will eventually shed any humble beginnings and be elevated to much greater heights than any proprietary software could ever hope to reach. If it starts out in open source form I believe it will follow the same progression, assuming, of course, that the functionality of the program is important and that, as a component of a computer system infrastructure, it is necessary for it to exist in open source form.
    Frank
  • "What changes to their current UNIX definition would be required for this?"


    Here are some requirements for UNIX 95. [opengroup.org]In addition, several truckloads of cash are required.
  • i know only too well that there's more to the web than the web. the point i was making is that i know slashdot as "slashdot.org" which is what i type in (actually it's in my hotlist) but if, for example, i told a friend to "visit X.org" it wouldn't work and he would give up.
    Imagine that amazon.com only worked with the www (i checked - either works) they would miss out on loads of business because loads of newbies and experts alike would type "amazon.com" because that's what they heard on the tv/radio/from friend or whatever. the newbies would give up and the experts might skip to the next online shopping mall they felt like looking at. even if they persisted the time wasted would be annoying them
  • take uk.profiles.yahoo.com/son_of_hoss or uk.search.profiles.yahoo.com. what's the domain there. i can't think of many of the top of my head but i have seen ????.www.??.com
  • there is no possible way to convince free software developers to clone CDE unless some of them are hell bent on passing the certification

    I think the goal of the Single UNIX Specification was to make all systems that claim to be UNIX look sufficiently like one another that software, and users, don't get locked into particular versions.

    If

    1. not enough software uses the features of CDE to make a system lacking it "not UNIXy enough";
    2. the users whom the standard is intended to help don't really care;

    then the case should perhaps be made to TOG that mandating CDE isn't worth the effort.

    Given that CDE is their baby, I'm not sure they'll accept that case, but that's another matter. Perhaps the answer then is to have commercial distributors of Linux, etc. systems package up versions with CDE, etc. and get those certified as UNIX; those who care can buy those systems, and those who don't care don't have to get a system with CDE.

  • Motif was not partially funded by Microsoft!
  • X seems fine to this UNIX nerd sitting in front of a 19 inch trinitron. I like being able to configure things. And the highlight-middle click seems to work everywhere I've ever wanted to try it. Yay X!
  • Open up the documentation for Motif, and who is listed amongst the sponsors? Yup, that's right Microsoft.


    Chris Wareham
  • by crow ( 16139 ) on Wednesday December 01, 1999 @11:29AM (#1489215) Homepage Journal
    I should be noted that The Open Group's X.org needs the help of XFree, not the other way around.

    Remember back in early 1998 when they changed the licensing on X? The idea was to generate more revenue to fund The Open Group's X development efforts. Well, The Open Group dumped X development that summer (along with most of their other development efforts), and was nice enough to change back the licensing.

    So now the only organization doing public development of X is XFree. Hence, anything that goes into XFree becomes the defacto standard. By making XFree an honorary member, it makes it easier to keep the defacto standard and the paper standard in sync. Since The Open Group only exists for standards and branding anymore, they desperately need to retain control.


    DISCLAIMER: I am a former employee of The Open Group, but was not involved with X.
  • by PiMan ( 2859 ) on Wednesday December 01, 1999 @11:31AM (#1489216) Homepage
    Hey now, don't discount this guy as a troll so fast. X, although it might be the best we have now, is horribly old and bloated. Did you know that the core code for most X distributions is the original reference implementation? Yep. No one has ever written a real X core. Plus, the 1984 standards really haven't been updated.

    Now, look at XFree86. They're doing modular systems, fbcon servers, and native OpenGL acceleration - things that actually involve changing core code and bringing the standards up to date. While I don't support XFree86 pulling away from X entirely, they have been taking their own direction a lot, and that's a really Good Thing.

    Also, some small hype: The Berlin Project [sourceforge.net] is an attempt to redesign a new windowing system from the ground up, using CORBA and OpenGL. While it's far from complete, if you've got any coding skills and dislike X, donate some time there.

    Also, please do drop the 86. XFree is a great cross-platform X distribution, not just for the Intel line anymore.

  • Yes, it's a really old domain.
  • Berlin [benham.net] is trying to replace X with something more modern (using bits of OpenGL, Corba, ...). I'm not involved in the project at all, but I think it's worth a look if you're interested in alternatives to X. As a former Amiga user, I certainly agree about the 'big, slow, bloated, and old' comments. However, X has its good points (e.g. network-transparency).
  • by Christopher B. Brown ( 1267 ) <cbbrowne@gmail.com> on Wednesday December 01, 1999 @11:39AM (#1489219) Homepage
    It's not clear that TOG and XFree86 are "getting along;" what is evident is that the X committee has decided that they need to "get along" with XFree86.

    It is entirely possible the people at Sun, IBM, HP, Compaq, (and possibly others) decided that as they're supporting Linux from other perspectives, that the needed to tell TOG that it needs to as well.

    Convincing signs of a TOG "change of heart" would include things like:

    • Defining UNIX 1999 or UNIX 2000 in such a way that Linux systems could become branded as UNIX, rather than their current comments [unix-systems.org]
    • Releasing Motif/CDE in Open Source form.

    Shameless Plug: People should help Sponsor XFree86. [xfree86.org] My local Linux Users Group, NTLUG, [ntlug.org] is in the process of soliciting that members help sponsor several free software project organizations, including XFree86.

  • by alexhmit01 ( 104757 ) on Wednesday December 01, 1999 @11:40AM (#1489220)
    I disagree, maybe SOME Linux users don't make the distinction, but "geeks" will. We've read the white papers, the design specs, etc. We understand why X is so powerful. XFree86 does a tremendous job of creating a really solid Xserver, but it is NOT X. (I'm typing this from IE under NT, while my KDE desktop smiles at me from my session under Exceed).

    Until recently, XFree86 was playing catchup, providing an implementation of X designed for people to play around with. However, with the other Unix vendors moving out of the workstation market and into the headless server market, XFree86 is the most compliant server out there.

    For a while, XFree86's interests and The Open Group's interests were NOT compatible. They were trying to develop an interoperable system so that expensive software makers would have an easy time developing for multiple Unixes. However, even more importantly, was the attempt to share research costs. People bought Unix workstations for their power, not their Xserver, so it was in the company's interest to work together on the display portion.

    Now, with Unix moving into the high end, and Linux gobbling up the low end, there is a desire for a strong, inexpensive X-server. Hummingbird probably isn't thrilled, although Exceed doesn't seem to be their major selling point, so perhaps they don't care. The Unix vendors are bringing XFree86 in because the need inexpensive workstations to manage the Unix machines. X's distributed nature makes a Linux machine an ideal platform. Additionally, after watching XFree86 on my friend's Solaris x86 machine, maybe the Unix vendors are planning to get out of the X-server market, letting XFree86 pick up that entire R&D tab.

    Realize that XFree86 could have joined the group at any time, merely by ponying up the cash. Being given a free seat on a Industry board is pretty significant.

    What makes this amazing, is that this is the same group that almost caused XFree86 to fork from the X standard, because they were going to charge beginning with R4... Perhaps they realized that XFree86 was the only group interested in pushing X further, as Unix was too high end for workstations. Linux, however, has a desktop fancy, so it needs a good GUI.

    Alex
  • by Anonymous Coward
    But... GNOME runs on TOP of X...
  • That doesn't answer the question. Obviously there are requirements; what changes to those requirements are required to be "Linux-friendly"?

    In addition, if Linux (or any other UNIX-flavored OS) can't satisfy those requirements, why should the definition, rather than Linux (or any other UNIX-flavored OS), be changed? (There may be legitimate reasons to change the definition, but I don't consider "Linux out of the box can't meet them" - or "{Free,Net,Open}BSD out of the box can't meet them", or "{pick your favorite commercial UNIX} out of the box can't meet them", or "{pick your favorite OS that has a UNIX-compatibility environment} can't meet them - to be, in and of itself, a legitimate reason.)

    In addition, several truckloads of cash are required.

    "Linux", in the sense of the Linux kernel, isn't something to which one would give the UNIX trademark, given that it doesn't implement the full API, or command-line interface, of a UNIX system, so what would be tested would be a Linux distribution.

    One could argue that the vendors of the larger commercial Linux distributions could supply the truckloads of cash necessary to get their distribution tested.

    However, if there is ever, say, a Linux Standard Base "sample" or "reference" implementation, it'd be nice to have that certified (assuming the LSB covers everything that the current UNIX standard specifies), so that perhaps it could be arranged that any LSB-compliant Linux distribution gets the UNIX trademark automatically. If that's doable, perhaps, again, the commercial Linux distributors could supply the cash.

    If not, however, it'd be nice to arrange that it be possible for the non-commercial distributions, or distributions from smaller vendors, could get the UNIX trademark as well.

  • Huh? GNOME depends on X. X does not define policy.
  • In addition, if Linux (or any other UNIX-flavored OS) can't satisfy those requirements, why should the definition, rather than Linux (or any other UNIX-flavored OS), be changed?

    Because requirements include mandatory use of some proprietary software (made by TOG itself, BTW).

  • Yes it's probably just you. At least not me!

    X serves all my needs and even a little more.

    Study the documentation for your window manager a little ("man fvwm2" for example) and you can make it look just as you want it to..

    Markus
    --

  • The Unix vendors are bringing XFree86 in because the need inexpensive workstations to manage the Unix machines. X's distributed nature makes a Linux machine an ideal platform. Additionally, after watching XFree86 on my friend's Solaris x86 machine, maybe the Unix vendors are planning to get out of the X-server market, letting XFree86 pick up that entire R&D tab.

    You're spot-on here, but you're actually understating the case. The Unix vendors are planning to get out of the Unix market too! Most of them consider themselves hardware vendors, not OS vendors, hence Compaq, IBM, HP all throwing development dollars at Linux. The faster Linux gets accepted and is able to run on their hardware, the faster they ditch their software development budgets and spend money where it really counts, designing hardware that they can beat each other up with.

  • of the XFree representative sitting on that board and glaring down at the peon members of the other companies from her throne, then suddenly wielding her terrible power to punish the recalcitrant corporations for their slowness in appeasing the mighty beast that is Open Source . . .
  • Because requirements include mandatory use of some proprietary software

    Such as Lesstif? :-)

    Unless the requirements mandate that you provide APIs that can only (either for technical or legal/patent/etc. reasons) be implemented by software from TOG, I don't consider those requirements ipso facto "Linux-unfriendly". They may require that free reimplementations of the APIs of XTI, and Motif, and CDE, be implemented, but that's another matter - after all, Linux distributions already include a kernel, libraries, and utilities that are free implementations of many UNIX-flavored-OS APIs....

    (Yes, commercial distribution vendors could buy the TOG software and put it into the distributions that they run through the test suite, but if you have to do that, I wouldn't exactly consider that very "Linux-friendly".)

  • Oh, I thought XFree86 was X.
    X without XFree86 seems to be shown as just a letter of the alphabet.

  • You might think, but Intel has shown interest in *helping* Linux, and I think there are a couple of motivations for this. First, although Microsoft software has pulled Intel to the top of the semiconductor heap, Microsoft has been the dominating force in Wintel. Second, reducing the cost of software makes for more machines being sold, which is good for Intel. Intel doesn't just want to sell PCs, they also want to move into the Personal organizer market. Make Linux a viable competitor with WinCE and PalmOS, and Intel has a point of entry. Notice how Linus has been talking about embedded OS machines recently? (The weirdness there is that Transmeta is an Intel micro-competitor -- I don't see them working together.)
  • What a crock. Given that xfree86 is the only future that X has, they should have simply declined the offer and waited a few years for X.org to cease to exist.

  • The XFree86 Project, Inc. was a member of the X Consortium of old, too. At least in 1994 UUNET sponsored [xfree86.org] them and contributed the membership fee. So belonging to a standards body isn't anything new to them.

    Actually, I'm kind of shady on the details of how they lost that membership. Could someone more knowledgeable tell what exactly happened when TOG took over X? I was under the impression that the new X.Org had more or less the same members as the X Consortium. Apparently not...

  • Why do they call themselves "x.org"when typing in x.org into your browser doesn't work?
    Some sites work without the www and some don't but all websites should at least promote a link that works! or else fix there DNS setup.
    That's common sense to me.
  • Modular servers, native OpenGL are all old hat. The sample implemetation may not have had them, but they've been around (discussed at conferences) for 5+ years.
  • The nice thing about Berlin is that they have sought to use some substantively new bits of infrastructure, like CORBA and OpenGL, whilst "stealing" ideas from the (essentially failed) next-generation GUI, Fresco. [tu-harburg.de] (Fresco probably should have been pegged as the X GUI rather than Motif, to peg its chronology...)

    The problem is that Berlin utterly throws away compatibility with the huge bodies of works currently running using Motif, TK, GTK, Qt, Xt, and FLTK (to name the likely "top" GUI libraries used on X).

    As a result of discarding compatibility, Berlin is pretty useless until ALL your favorite software is rewritten to use Berlin. Some emulation may be possible, but this is nontrivial, and it'll literally take years for this to be robust enough for production use.

  • by Roundeye ( 16278 ) on Wednesday December 01, 1999 @12:22PM (#1489239) Homepage
    Is it just me or does X seem big, slow, bloated, and old? It can't keep up with pretty environments like GNOME

    What are you smoking? What do you think X is? What do you think GNOME is?

    There are more regulations and standards than anyone cares to read.

    "Man, there's too many regulations and standards to this Internet thing. Nobody wants to read that."... Again, what are you smoking? Noone has to read a single regulation or standard to use X, or even GNOME :-), but those standards are what makes it possible for me to launch a Netscape on a Solaris box behind a firewall in NY and have it appears on MI/X on a WindowsNT box in Tennessee. That's a simple example of what the standardization allows.

    We need a new windowing protocol that addresses the need of the modern PC user

    You've got one. It's called Windows. What's that you say? Windows is a shameless rip-off of the significant UI advances over the past 20 years? Oh, Windows is poorly written bloatware? I'm sorry. Maybe we should use a standardized system that has been refined over a decade and which allows extensible network-transparent sessions over multiple architectures. Hey, why not make an OpenSource version that's also compatible with the system that runs on Crays SGI's, Unix Mainframes, and PC's?

    The continued development of X seems pointless to me - I think we need something new altogether.

    Your continued ignorance seems pointless to me. However, regardless, why don't you scurry along and work on that dream of bringing a real GUI to the masses. Go sign up for any of a number of projects (browse through freshmeat and sourceforge) in this area. Heck, if you're uber-1337 maybe you can even get PAID for your efforts by working for a company developing such projects.

    Having used X over the past decade on everything from Heavy Iron visualization systems (with graphics hardware the likes of which YOU CAN'T GET for a PC), to mainframes, to Unix workstations, to Linux/*BSD boxen, to win16/32 machines, to wierd-ass closed network devices... I'd say you'd be hard pressed to find a viable alternative to X anytime in the next 5 years. The flexibility, scalability, portability, and customizability (X+GNOME+Enlightenment, for instance is slicker than eel snot) of X beats the pants of the competitors.

    But, if someone develops an alternative that exists on a wide range of platforms, enables the productivity that X enables, and runs something as nice as Enlightenment (not to mention WindowMaker/FVWM/etc.), and something as minimal as twm, then you might find me using it as well.

    So get your ass to work if you're so dissatisfied.

  • by Anonymous Coward
    Look at his posting history.

When speculation has done its worst, two plus two still equals four. -- S. Johnson

Working...