Slashdot Mirror
Fedora Prepares For Xorg Instead of XFree86
ZuperDee writes "I noticed in the development branch of Fedora today that they appear to be in the process of creating new xorg RPMs, and from the looks of the changelogs in those RPMs, it looks like their ultimate plan is to switch from XFree86 to the XOrg Foundation's implementation of X11. Anyone else here think this could signal the beginning of a new trend in Linux distributions, and that XOrg could end up becoming the new de-facto X11 implementation?" (See this earlier story,too.)
I don't know enough about Xorg to know if this is good or not.
Is the driver support there? Will NVidia's and ATI's binary drivers work with the Xorg server? It could be a real problem if FC2 won't be able to do accelerated 3d under NVidia or ATI cards.
Could somone go over the diffrences between X11 and Xorg? Is it just a license issue, or are there other differences?
Thank you.
52 Weeks, 52 Religions with John Hummel
That's one of the reasons I like the open-source market. There is no de facto, its pragmatic.
At least, in theory.
Just when Linux was getting to the point where it could overtake Windows, even in the desktop (www.mandrake.com) environment, XFree86 changes its license. Now, Fedora is switching its Graphical Display. No matter the security, the stability, etc. the average home user will probably remain with windows. He wants his program to be work with his computer. It may not be that simple once more distro's use more widely varied XFree86 implementations
http://www.beyourowneviloverlord.tk
http://www.frozenchickenthrowing.tk
http://www.killercamel.tk
Is this the same thing as Xouvert, or something new?
Can someone give a ten second summary of the differences in the goals and developers of XFree86, Xouvert and Xorg?
Fedora switching just means we have more choice. This is a good thing, just like KDE vs. Gnome is a good thing.
Most people will settle for whatever comes with their distro, so maybe this will give an impetus for the X group to clean up the licensing issue :-)
People ! I am bulding You are building They are building So ... let's give them the credit for offering alternatives.
I think the XOrg codebase is pretty much the last pre-license-change (4.4rc2) release, plus work done by the folks recently run out of XFree86 by Dawes.
While I don't think the X world will turn on its ear just because Fedora may start using Xorg, I think the fact that one or more distributions are currently/going-to try it out is A Good Thing(tm).
No, I doubt it too. What WILL it take to wake them up?
Both NVidia and ATI keep their driver sources and hardware-level programming information closely guarded secrets. This means unless NVidia and ATI decide to support the new X server, we're gonna be stuck with lousy 2-d drivers, maybe with accelerated blitting if the mfgrs decide to throw us a few crumbs.
Meldroc, Waster of Electrons
X is not just a Linux thing. A major free version of X should be designed to work on BSD as well. BSD users do not want to have to put their codebase closer to the GPL than it already is.
Plus, I like the idea of standardizing on MAS. In some ways inferior to Jack, but anything that gets a lower-latency sound daemon to be a standard i'm for.
Of course, Gosling was never an X architect. Those were Scheifler, Gettys and Newman. Gosling was the architect of NeWS, a competing windowing system that ultimately lost out to X. Yes, IHBT. Thank you and good night.
"The invisible and the non-existent look very much alike." -- Delos B. McKown
http://www.freedesktop.org/Software/XserverFAQ
Anyone else think that these personal comments at the end of news posts are irrelevant and should be marked -1 Redundant?
Of course this marks the start of a new trend, Red Hat just beat Mandrake to it. After the announcement last month about XFree86's license change and the very negative response for everyone, this was expected. It's only surprising because it happened so quick
"Some things have to be believed to be seen." - Ralph Hodgson
It's X, X is for the most part X whichever X you run. If feature y on server z of X doesn't make it the standard, what make anyone think license clause w for server v will?
Having two equally used Xes would be better I'd think, after all they follow the same X standard.
I'm sick of following my dreams - I'm just going to ask them where they're going and hook up with them later.
just as Red Hat got Gnome and KDE to look exactly the same... Merge the wrappers, split the libraries
My Karma is so low that even my own postings are beyond my current threshold
That's what the XRender and Xft are for! They are full replacements for the old rendering model and font subsystem.
XOrg could end up becoming the new de-facto X11 implementation
It's a little early to make that kind of prediction. However, the key is compatibility. If XORG maintains full compatibility such that it's still X11 and we can just a recompile and go on our merry way, then anything is possible. Personally, I don't think people care which code base their X server uses so long as it's an X11 server. Reality is that the XF86 group will wake up an smell the coffee sooner rather than later, they're expendable, they just don't know it yet.
Good people do not need laws to tell them to act responsibly, while bad people will find a way around the laws-Plato
"Every time I look at the X window system, it's so fucking stupid; and part of me feels responsible for the worst parts of it."
"The color model and the fonts is aggressively stupid. It works, but oh my god. It's awful."
--James Gosling, X Architect
If he feels so bad why did e make an even bigger mess with Java?
The Register may be stupid and generally wrong, but it is good for lines like "Gosling is a good-natured Canadian, and he set about the competition with the same relish that his countrymen have for clubbing baby seals."
What I'm listening to now on Pandora...
C'mon now...
Besides, if you never read the articles, and just look at the exceprts, you'd never know about the asparagus. What asparagus, I hear you ask? My point exactly.
You're special forces then? That's great! I just love your olympics!
Well, that was an informative two sentences.
Can we get something more substantial than 'plain sucks'?
I know that X11 is a protocol, defining an abstraction of a user interface, "mechanism, not policy", so that there a metric booty-load of libraries wrap it and expose it to various other technologies.
Apparently, the XFree86 core fascists ran out of cool points, so the Open Source community has been fragmenting at a rate that must have certain Monlopoly Sycophants oozing glee in the Northwest.
Can someone with insight explain what the issues are with X11 and the XFree86 group?
It's a good thing that there is a leader like Linus in kernel-land; while there are enough alternative patchsets in circulation to keep the ideas flowing, there is no doubt as to the HMFIC on the project.
Or GCC (Is that RMS personally? If not, indirectly) over in compiler-land.
If the GUI people, from X11 through Gnome/KDE, could behave in as unified a fashion...
why not keep XFree (old version) for a while and work on an X alternative, something like the front end for Mac OS X...
Why did I lurk so long before registering for a Slashdot account? I could have had a Slashdot ID of less than 100000.
OK ... for the benefit of those of us who don't hack X in our spare time: is the Xorg implementation the same as the "freedesktop.org" implementation (at http://www.freedesktop.org/Software/xserver) or are they separate and distinct? (Or maybe just separate?)
Either way, how about brainstorming-up a better project name? I personally like "Product X" but that may already be taken.
N
-- Watch the REAL Jon Katz.
Are X.org and Freedesktop.org the same or are they different distributions?
-clueless
Doesn't really take a genius to predict that this would happen :-)
Alright, there's been a shitload of ignorant posts here.
First off, this new server is a snapshot of XFree86 just prior to the licence change. Basically a fork.
Second, it basically has nothing to do with X.org - I don't know why they call it that, most likely due to the licence.
Third, X11 is the protocol that X servers speak nowadays. X version 11 release 6.6 to be more precise.
Fourth, nvidia and ati drivers will work.
I hope this clears it up somewhat.
Ok kids, here's the quick summary to get everyone up to speed.
XFree86 and FreeDesktop.org's X server are both X11R6-compatible X servers. The FreeDesktop.org server (herein known as XOrg) is a fork of an old XFree86 project called the KDrive.
The KDrive was a tiny X server implementation originally designed for PDAs and such. When you compile it the binary comes out to about 700kB and it requires hardly anything else to function. The author of the KDrive took (read: forked) it from XFree86's tree and started adding onto it, and it became XOrg.
So X11R6 applications and libraries work almost exactly the same under XOrg. The XFree86-specific extensions to drivers and shit need to be ported but most apps don't use those.
Gentoo, RedHat, I think SuSE and Debian and soon to be more Linux distros are all slowly switching to XOrg. Until then they'll be shipping XFree86 4.3.99.902 and below as those are the ones without the evil licensing changes.
This has been in the works for some time people, so it's not a rumor or a guess.
Note: XOrg isn't the real name of the server, I just call it that cuz im lazy. XOrg is the name of a foundation that puts out this FD.O Xserver. Info here.
This is an important milestone in the growth of Linux as a whole. Using xFree86 felt like wearing very heavy shackles due to its recent adoption of a horrendous draconian licensing scheme.
xOrg on the other hand avoids these issues by literally borrowing the xFree86 code, preserving it the way it was before the licensing scheme changed for the worse.
We should all take note of this process. I for one, think that the latest version of Mozilla is too bloated. That's why I've taken the source from (current_version - 1), de-bloated it, and altered the license so that I can make a lot of money from it. I've even changed the name to Zila (pronounced like the "zilla" in "Mozilla") to match that of our company.
Once xOrg gets mature enough to mess up their licensing scheme, I'll drop right in and make a tidy profit again.
I guess that is why MS and others are still replicating one of X' most important features, being usable over a network to provide a remote desktop. Don't get me wrong, it is definitely time to kick out some outdated stuff or at least bring in replacement for many of the things X does when used locally, but generally X is very usefull and there is no reason to throw that away.
I know that X11 is a protocol, defining an abstraction of a user interface, "mechanism, not policy", so that there a metric booty-load of libraries wrap it and expose it to various other technologies. Metric booty-load? Is that more or less than an English booty-load?
Why is this marked troll?!? It makes some good points!
Umm ok
First its based on XFree86 4.4 just before the change, with the non-contaminated further changes added and other stuff not in XFree 4.4
Secondly it has a _lot_ to do with X.org. The wheel has turned full cicle from when years back OpenGroup/X.org tried to change the license and XFree basically told them to go away to today where X.org is doing the same thing the other way around and keeping it free. X.orgi is part of this now.
NVidia and ATI drivers may work. The Nvidia ones at least are reported ot do so, although they have chronic problems working with the preferred kernel build settings like 4K stacks.
Anyone knows if this will make it into Core 2?
Opus: the Swiss army knife of audio codec
Funny, but it seems that really don't need bad reviews, freezes, or crashes to end the life of your software project. All you need it a sucky license.
As I understand it, Xfree changed their license to make sure more credit is given to its developers. But who gives a crap when no one will use because of the license itself.
The only thing necessary for Micro$oft to triumph is for a few good programmers to do nothing". North County Computers
that because they are very similar the few widely used binary drivers out there (read: nvidia) will either work with both, or will release a version for both.
-Adam C. Greenfield
I think that this is silly if it's for Licensing reasons alone. I predict that more and more people who actually make a living making Open Source software are going to move away from the GPL and it's permutations, for Open Source licenses that actually make sense, that are not based on RMS's gigantic ego.
"Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
I'm waiting for Y-Windows, personally. They've been making great strides on their core widget set. They plan an initial X compatibility layer, but other than that it's a completely rewrite abandoning X all together.
JeanBatiste Emanuel XOrg
When will we get the Y?
Achille Talon
Hop!
And who modded you up as Informative?
X11 is a standard, but the implementations can wildly vary all they want. For instance, one might implement drivers in a completely different way than another.
This forced switch is a bigger deal than you're making it out to be.
Ther is at least an option in the X marketplace, uptill now (or whenever) there was no option to Xfree86 for all unix based distro's.Now with the xfree86 messing up the license theres some hope for a Mandrake and his buddies, don't you think? ;-)
Lord of the Binges.
Wonderful. Add yet another layer of cruft. Count me out! Paying for a Mac and OSX seems a lot less painful, and cheaper than the psychotherapy bills working with X would incur!
but slightly.
Could linux itself exist under a process that enforced focus? The code which is stable has become stable through an evolutionary process... tortoises have overtaken hares, and the competition of a software ecosystem is crucial.
As good as more focus would be for some things, I see no way open source can work that way. Since I do also agree with your point, maybe I'm saying there is no solution... but I still feel there is a ballance and room for the distributors to provide that kind of focus when it's important.
-pyrrho
I think he meant x.org vs. Xorg. x.org is the site hosted by the Open Group for X11 releases and development. Xorg is a much newer group.
This man blows goats. I have proof.
So how many developers are going to continue working on the newly licensed XFree86 project vs jumping to this new forked version?
What distros will continue to use XFree86? Any?
The man who trades freedom for security does not deserve nor will he ever receive either. - Benjamin Franklin
OK, there's been a lot of questions and I got lost following all of them. I haven't seent he answers yet that I was hoping for, so I'll muddy the waters a bit more ;)
:)
It seems like there's 3 faces to compatibility: X11 protocol, driver, and library. I know XOrg is X11 compatible (otherwise it's not an X server), but how about the driver side and library? The driver question seems the most critical for a drop-in replacement to XFree. What about the libraries though? I know an app just has to talk X11 to work, but most do that through a library (and a lot do it throught the xlibs that come with the server). Are those libs going to be affected any? anyone have an opinion?
And, if I'm lost in my thinking, mod me out of sight. I'm used to it
AB HOC POSSUM VIDERE DOMUM TUUM
Anybody else think Xorg sounds a lot like Zorg?
(yes, I know that X is hardly Linux-specific...)
Al Qaeda has ninjas!
***Rick and Roll eats his shorts***
As if Fedora has controlling market share to sway any level of adoption preference by the entire community.
absolutely, except where equal. /. editor, in wich caase yer speling shood ujust accordionly.
next time, close your italics, unless you're trying to get a job as a
I don't take Fedora too seriously. I'd be much more concerned/interested if say SuSE or Mandrake were doing this.
As with most memories, much of the prehistory of X is fading into oblivion.
:-)
...
Probably for the best.
found at
Why do people call it X-windowS?; written by Jim Fulton on Sat, 2 Jan 1993 02:22:21 +0100 ?-!
CC.
TaijiQuan (Huang, 5 loosenings)
% grep -r "Lesser General" xserver % grep -r "Library General" xserver % Sorry for the misinformation. Spread a rumor apparently started on /.
If I were a CVS developer I'd be really pissed off right now:
> a POS excuse for a revision control system like CVS
Sorry? what happened to good manner? CVS is still the work horse of pretty damn near 100% of all free software projects and God now how many non-free ones. It may have a few issues but it works as advertised and it is still very very useful, pretty much bug free and has been for many years. The CVS issues that Arch is designed to fix don't come up very often in real-world projects with a couple, or even a dozen of developers, and at the moment maybe Arch is stable, maybe it isn't, and it certainly hasn't seen the kind of usage that CVS has, so people may have very good reason not to choose it over CVS.
When you have contributed something of the order of magnitude of usefulness as CVS, maybe you will have a leg to stand on.
In the meantime people have work to do. Good day to you.
This is clearly a troll. None of the code in question is covered by the GPL or even LGPL. Besides, the move shows a clear trend towards developers moving in the more-free direction, not in the less-free direction.
Is it good, or is it whack?
I, for one, find Tet offensive...
I think they'll follow whoever the biggest commercial Linux distros, and today that means RedHat and Novell/SuSE.
Why support XFree86 if the big distros are dropping it?
The reason that devleoping for Windows (and OS X) is easy is that you KNOW what libraries are there. If you require Win98 or higher, you just target those APIs. When NT4 was big, you would see things like NT4 SP3 as the requirements.
The problem with releasing software on Linux is that there is nothing to support. A RedHat 9 installation can have all sorts of libraries, so you say glibc x.y, kdelib x.y, etc., and end up with dozens of requirements.
The exitance of KDE and GNOME is irrelevant.
If RedHat version X including in the installation a core set of libraries (regardless of server or workstation configuration) then you could just say "supports RedHat 3.0 or higher).
The problem is the dependancy issue, which would NOT be an issue if the System installed the libraries by default, regardless of applications.
That is the "Linux" problem, you CAN'T release commercial software easily because the support is a nightmare.
The KDE vs. GNOME isn't really a big deal, as a Corporate Desktop (the first step in a desktop market) can easily specify roll-outs. That's Linux's (and OS X's) advantage.
Windows was NOT built in a network-centric way, and roaming profiles & home directories are handled FAR worse than the Unix way.
The problem is that you can end up with a nightmare because distributions don't have standard libraries.
Alex
I good explanation for it that I read at osnews.com was that the XFree86 and the Distros (commercial and community alike) started to increasingly have differences in priorities and culture. The license change was a like message from XFree86 to the distros that they didn't care one way or another for their support. The distros response is logical. Additionally while most of the distros have pleny of software incompatible with the GPL, it is not ideal to have something as central to an operating system as the X server to be GPL-incompatible.
D'oh. :)
Have everyone create spin-off projects or re-implementations. Give them all similar names and/or purposes. Watch as even the 'techies' on slashdot can't decide which is which. Kiss the whole thing goodbye.
Someone needs to sort this mess out, before any more harm is done.
There's a big difference between having the option of a remote desktop vs. paying a performance penalty all the time.
If it doesn't have accelerated support for video card X (and forking the tree will have that effect as development resources get divided), I don't care how open it is.
Does is matter how far you can open the hood of the car? I'd rather be able to open it three quarters of the way open to see a nice eight cylinder 450 than to be able to open it compeletely and see the hamster and his wheel.
Agile Artisans
What does this mean for existing binaries, as well as trying to recompile existing sources.
Will this require porting/recompiling or can I just run an binary for XF4.x with out issue?
---- Booth was a patriot ----
For the user, OSX it is a dream. But for developers, it's a wet dream. Creating slick interfaces is simple, the PDF-inspired graphical model is a breath of fresh air, and the interfaces inherit impressive functionality automatically. Because its code-development process leverages effort powerfully, perhaps more so than for the comparable GNOME/KDE tools, I think OSX offers good potential for the open-source movement, given well-fashioned attitudes and licenses.
I make these remarks with some trepidation, since I think the fragmentation across GNOME and KDE dilutes developer momentum. Also, I make these remarks to evoke discussion by those more technically-aware than myself.
yes becuase you can prove how well a mono culture works, see ms. you fools, it's about time a fork of X appeared, i have been itching for an alternitive for ages now. and for the clueless, this doesn't mean we will need to convince developers to write new drivers, see fork, see compatable, see becuase it's not closed source we can do this.
If you mod me down, I will become more powerful than you can imagine....
Most projects haven't even begun to support XF86 4.4 and since X.org is is based on a release of XF86 4.4 prior to the license change, it is certainly going to be almost as easy to move from XF86 4.3 to X.org as it would be to move to XF86 4.4...
To be honest, the only difference between XF86 4.4 and X.org (that I can tell) at this time is the new XF86 license...
So, as to how all of this will pan out...it will be left to the individual distros and developers. If they see promise and innovation in the X.org project, they will go with it, and on they other hand, the new license shouldn't cause a problem for any distro that already includes Apache...because the change to the XF86 license is pretty much the same thing as the Apache license requires.
This reminds me once again of when X.org tried to go proprietary to preserve its investments (actually its members') to be hoarded by non-members. XFree86 reacted, and after I put them in contact (though I'm sure other, more important people did that too) RMS offered to help them going copyleft ([L]GPL); they got the idea and almost went with it but XFree86 didn't want copyleft.
Perhaps now XFree86 decided to go GPL-incompatible, some even say non-practical even while free, it would be time to go LGPL or even GPL? Thus proprietary vendors would have to either stick with XFree86 and its advertising clause, or pay and thus help develop (X.org|XOuvert|FreeDesktopX).
Copyleft wouldn't hurt the BSDs, BTW. They already use gcc.
Leandro Guimarães Faria Corcete DUTRA
DA, DBA, SysAdmin, Data Modeller
GNU Project, Debian GNU/Lin
When the information I have at hand changes, my opinion may change. In other words, I base my opinions on facts. Assuming you don't always have all the facts in front of you right from the start, what is your reaction then to new information?
If you think thats bad, try making an SDK for a foriegn device on redhat.
On other distribututions, rpm (yes, rpm) works fine. it doesn't strip binaries it shouldn't touch (arm for instance in an x86 package), it doesn't add depenedencies it doesn't need, it basically just works as advertised
Then we have the one Redhat box. everything is different. a wrapper was needed for strip so it didn't make arm binaries useless, and about 400 lines of rpm config code were needed to stop rpm screwing up the dependencies.
Then there are the problems of installation and use. there is no such thing as a minimal redhat9 installation. all we wanted was to build packages for redhat9, 2gig was as small as I could get the build. And if you are thinking not so bad... well I got debian down to 20 meg once, with all the same things I needed on that redhat box.
There are also the configs, the different command usages, etc. etc.
I do disagree with it becoming an ad-hoc standard though. Nearly everyone I know who is serious about Gnu/Linux stays away from Red Hat, as serious bike riders stay away from training wheels.
Don't get me wrong, Redhat is still a good distribution....
Redhat: new users and coporate users needing a good backup plan.
SuSE: for those who know what they are doing but want a low fuss machine.
Gentoo: for those that know what they are doing and want a optimal machine.
Debian: Perfectionists without time. (if you have time, go back to gentoo, you just have to do all the checking and restrictions yourself).
I speak as someone who has used them all at some point for at least at 6 months at a time.
...the folly of relying on binary-only drivers.
Really suck.
Pressure the manufacturers for code or at least specs. Explain that we're happy to do the work if it means that we're not tied down to specific releases of their code matching specific releases of kernels and X versions for specific distributions and released when they get around to it.
Please refrain from using overly simplistic arguments to support your cause. In my opinion, it wasn't JUST the license change that lead to this seemingly spiralling downfall, but the head developers of XFree86 itself (David Dawes, to be specific).
/.) and the man strikes me as positively arrogant, with no respect for the opinions of others (unless he was actually majorly outnumbered, and sometimes not even then). He has repeatedly ignored input from other people including his own co-developers and loves to portray himself as the righteous leader. His posts are nothing short of antagonistic and he has very selective memory.
I've read and re-read various threads on the XFree86 mailing list (please look it up in archives and past posts on
Would that be sufficient reason for a project to fail? In this case, I would say so. He insists on having and keeping all control of the project to himself. If he had good sense, that wouldn't be a problem, but he's already shown that all he's interested in is recognition and retaining control over the project (rather than the project's welfare).
Past posts have shown that several suggestions and patches had been ignored which led to the project's stagnation. You may argue that the project is successful and works even now, but the point is it could have been so much better under a different type of leadership.
The recent license change is but one manifestation of how callous the head developers are.
well as a user, you are the last to be counted, the people making the stuff useful are the first, they have to do these things. they are the most important.
users dont get a real voice in these situations, unless those users step up and start developing.
not to mention you didnt read anything about the new project (must have slipped your mind right?) or you would know its got those drivers.
you shouldnt post in these discussions because it is out of your league.
sorry.
On other distribututions, rpm (yes, rpm) works fine. it doesn't strip binaries it shouldn't touch (arm for instance in an x86 package), it doesn't add depenedencies it doesn't need, it basically just works as advertised.
Have you used the 'AutoReqProv: no' line in RPM? Works fine for me in preventing spurious dependancies.
there is no such thing as a minimal redhat9 installation. all we wanted was to build packages for redhat9, 2gig was as small as I could get the build.
Have we not heard of "Select individual packages?" I routinely build RH9 and FC1 boxes in the 800-900MB range. Could probably do with less if I really wanted to.
There are also the configs, the different command usages, etc. etc.
When you're using any package management system there are bound to be configs that are placed in automatically. Aside from RH basically not using
Redhat: new users and coporate users needing a good backup plan.
etc...
Blah blah blah. It's certainly as possible to tune a RH system for low-fussness, or for high performance (i386 packages don't make that much of a difference!) just as it is to mis-configure any of the other distros. Use what works for you.
Hire a Linux system administrator, systems engineer,
... as in the Bill Gates type in the _Fifth Element_.
since MAS works over the network (like setenv $DISPLAY) and JACK is local only.
"Steve Jobs said two years ago that X was braindead and would be gone within two years. He was half right." - Dennis Ritche.
I'm expecting that the majority of distros will very quickly follow Fedora.
I know for a fact that Debian, Gentoo, and a few others are specifically NOT touching XFree86 4.4 (i.e. post-license-change), and are looking for alternatives.
X.org sounds like it is currently the most mature alternative, and will likely have the marketshare XFree86 does within months, unless David Dawes pulls his head out of his ass and stops shooting himself in the foot. He doesn't seem to realize that his license change is going to make XFree86 a defunct project VERY quickly.
retrorocket.o not found, launch anyway?
It's been a known fact for a while that most of the major distros have been looking for an alternative to XFree86 thanks to the license change.
I'm predicting that it'll be a matter of weeks before Gentoo, Debian, etc. follow suit. It was a matter of time before someone made this plunge, Fedora just happened to be the first.
There really aren't many other choices.
The fd.org experimental server isn't ready for primetime
Sticking with old versions of XFree isn't a long-term option.
X.org seems to be the most mature branch with a decent license, and it sounds like a lot of former XFree developers have already jumped ship and are working on the X.org tree.
retrorocket.o not found, launch anyway?
What have you done for Free Software?
I'm the maintainer of CSCVS, a tool for breaking CVS repositories into changesets, reporting on them, and importing those changsets into TLA. As such, my familiarity with CVS (and Arch) goes beyond that of the average user.
Now, about the issues that Arch is designed to fix: Revisioning renames and moves is not something that comes up only infrequently. Revisioning metadata is not something that comes up only infrequently. Mutually merged branches are not something that come up only infrequently. Taking forever to do a "cvs update" on a 10,000 file tree because the tree needs to be walked to look for updates is not something that comes up only infrequently. I've had the lead developer at work bitching in my direction because CVS is coming up with spurious conflicts that Arch would ignore.
I have a leg to stand on right now, and if you'd care to stand up and try to argue your position on its merits rather than firing off some angry rant, I'd absolutely love to do so.
Actually, a hundred is a bit of an exaggeration. It's more like 21. Here's the run down of which fork you should use to eat which foods:
"Three forks for Elven kings under the sky
Seven for the Dwarf lords in their halls of stone
Nine for mortal men doomed to die
One for the dark lord, on his dark throne
In the land of Mordor, where the shadows lie
One Fork to rule them all, One Fork to find them
One Fork to bring them all and in the darkness bind them
In the land of Mordor, where the shadows lie"
See, pretty simple....Happy eating.
Ok and what is the performance penalty you mention mr. rocket scientist. Show me proof that a performance problem exists in x because it can work across the network.
Got Code?
Let's see...
- Yellow Brick Road == Yellow Dog Linux.
- Mormonism clearly == SCO unix
So what's the Pied Piper?I use nVidia drivers in Linux for Xfree, will Xorg use the same drivers? are they compatible? I'd hope/assume so since Xorg is a fork of Xfree...correct?
CVS
free ipod and free gmail!
Nobody makes a 8 cylinder 450 so, you will have better luck looking for a hamster and his wheel. Go to the auto store and try to buy parts for an 8 cylinder 450 and they will know that you work on computers :)
There is no such thing as a metric booty-load.
"Booty" is from the nigger dialect of U.S. American English. The rest of the world (which uses metric measures, SI) does not have American niggers, even though we try hard to emulate them in every way we can think of.
That description in the gp is somewhat confusing.
De jure means it is the *official* or 'stated' standard (by law).
De facto means it is the standard by the *fact* that it is most widely used or recognized.
An example: if, in your company, the IT dept. has a 'standard' browser (IE) and refuse to support others, it can be said to be the standard 'de jure'. If, otoh, most users download and use another browser instead (Mozilla), that can be said to be the 'de facto' standard.
"I assumed blithely that there were no elves out there in the darkness"
anyhow, i would like to see another serious option besides xfree86
xfree86 crashes quite frequently for meI currently installed the rpms by Mike Harris from redhat/fedora on my Fedora Core 2 Test install and it works great. I simply had to reinstall my nvidia drivers, which work fine for everyone out there worried they won't work, and it runs fine. I for one am glad they are making this transition. It is time that X be open and maintained by a community with bugzilla. As more and more patches are sent and applied more companies may produce patches for their hardware since they are actually being accepted. This in my opinion is going to do nothing but help and improve a users experience with linux.
For starters, while LSB is terrific as a concept, it's a designed-by-committee kind of situation. It gives you the same problem that POSIX does, it can't be cutting edge.
When Apple or Microsoft put out something new as a standard, it becomes the default immediately. Win32 withered on a vine with NT, but Windows 95 made it real. The new libraries that came with later versions were allowed, to some extent, to be packaged with the OS. The only reason that DirectX took off was that games included the installer so that they could keep pushing the envelope.
LSB is always going to lag the releases, which is a bit problematic. Also, without a GOOD installation process (Apple has one, Linux should IMMEDIATELY adopt sudo and use a GUI sudo like OS X), there is no easy way to install new libraries.
The reality is that vendors are going to support limited distributions. That COULD be 1 distribution, it COULD be 5 distributions. However, there is no way that you can support "LSB 1.3 or higher," as you'd need to test it and explain how to install it in those scenarios.
Realistically, OpenOffice/StarOffice is basically at the point where you can set up Linux desktops for non-power users. Those that don't need the heavy automation that Excel offers will be fine (even if Star Office HAS those features, if it is differently done, it is too much of a pain to retrain), but better interoperability will help.
Linux will become a corporate desktop, and that will be the start. Linux will become a viable home computer for people that want to work from home and surf the web. The application issue matters less and less, as computers are more about the Internet than anything else.
Ignore the digital toys market, let Apple and Microsoft fight there. That market isn't about functionality so much as experience, let them play there. Then plan to go there in 3 years when it is a well known market and easy for open source.
Let it become like hardware. Those that need the best pay the premium and R&D costs. Then the middle solution exists, 80% functionality, 20% cost, then the generic version (open source instead of Taiwanese manufacturers) comes in.
LSB isn't the solution... at least not yet. The solution is corporate desktops with controlled libraries, and some consolidation in the Linux market. I mean corporate consolidaiton, not hacker consolidation. The open source stuff will run anywhere, but the corporate software market will consolidate on 1-2 solutions.
Alex
I applaud Fedora/Redhat - they have kept a constant standard of only having open source products in their distro.
I hope they keep it up. It is nice to some kind of personal standards/morals in the IT business.
Personally - I have a hard time choosing between Mandrake, Suse, and Redhat/Fedora. They all install so much better than Windows and Solaris they are a real treat to work with.
Comercial software companies could learn something from them.
I run nothing but Linux at home and work and I enjoy working with them everyday.
Everytime I sit in front of a windows machine I just thank the lord I don't have to rely on that piece of crap OS.
X11 was X11 right from the start as far as I remember. The 11 stands for one megapixel (as in a display 1000x1000) and one MIP (million instructions per second). At the time X11 was conceived this theoretical platform spec was thought to be about 5-10 in the future (ie mid 1980s) but actually such a machine was available in only 2 yrs.
X11 is a great example of designing for a theortical platform in the future so you're not tied to hardware constraints which results in sw with good longevity as it is to some extent, future proof (hence X11 last 20yrs beyond it's original design). Games designers have to do this all the time - MS Windows didn't do this in the past hence had to be rewritten from the ground up several times while X11 is (design wise) largely unchanged. MS Windows advantage is better performance generally (because it's is designed each time with the hardware more in mind) but much shorter life between rewrites.
This has also been a drawback of X11 requiring to take advantage of hw technology especially in the last 5-8 years - starting with SGIs GLX extensions and many more since then - some done nicely and some not.
pithy comment
> and that XOrg could end up becoming the new de-facto X11 implementation?
Only if we're lucky. Even if the XFree project comes to it's senses it can only be a good thing for there to be other implimentations of X available.
--
If I actually could spell I'd have spelled it right in the first place.
it would be a nice demonstration of the claim that opensource software can adapt quickly to 'breaks' in incompatable licenses (and unwanted behaviour).
Lets see how flawless this comes off -- will it cause confusion or can RedHat (the leading distro) make the change, leaving others, will it 'fork' and provide two distinct servers or will One Fail?
Should be interesting to watch.
That X.org would do something like 3-D desktop or Microsoft Bob clone :) Seriously though, I recently switched to Fedora Core 1 from Mandrake 9.2 ( previously I switched from Redhat 7.2 to Mandrake 9.0 and stayed with Mandrake until the switch mentioned above ). Anyhow, despite the opinion the FC is just a free beta testing bed for Red Had AS, I do find it does everything I need it to do, except games, but I have a Windows XP box for that :).
If they switch the X11 framework out, I will go support them all the way, one of the main reasons I decided to go Open Source for my main OS ( Windows is on one box of 5 ) is freedom of choice, and the ability to change even the video system out is a good example that freedom.
I can't afford a sig!
It would be a nice demonstration of the claim that opensource software can adapt quickly to 'breaks' in incompatable licenses (and unwanted behaviour).
This is not all (directly) about licenses. Keith Packard has done most of the new, interesting functionality in XFree86 for some time. By going with him, they are aiming for more modern functionality in their X server. XFree86 is very conservative about new functionality.
May we never see th
X11 was X11 right from the start as far as I remember. The 11 stands for one megapixel (as in a display 1000x1000) and one MIP (million instructions per second).
Sorry, the 11 is a version. From "man X":
I'm guessing your megapixel*MIPS was a retcon. Some of us are actually old enough to barely remember when X10 was just passing out of relevance, and I'd imagine a few of us remember before that. Versions before X10 were never really relevant outside of MIT. X10 was 1986, X11 was 1987, and there's been various X11R*s since then. Today, we use X11R6.4, but many programs want lots of extensions on top of it (eg, XRender). Since many of these have only been implemented on XFree86, that's now a de-facto standard.
According to this, four distros are continuing to use XFree86 despite the license change. They are Ark Linux, OpenNA Linux, Plamo Linux, and NetBSD. Does anyone know of any others or is everyone else switching?
Wikileaks, no DNS
People need to spend their mod points more considerately.
Wikileaks, no DNS
"Developed too long ago" - Bullshit
"Widgets and toolkits part of the server" - Bullshit
"X has outlived the usefulness of its design. It's time to move on" - Bullshit
Basically you're talking out of your arse.
Deleted
I don't want to be too PC, but I'm really uncomfortable reading the 'N' word.
Oh, I don't think anybody is talking about having the toolkit intergrated into the "X server" so tightly that it cannot be replaced with something else; that doesn't make sense. It's just a question of disallowing many different toolkits on the same desktop.
HAND.
> X has outlived the usefulness of its design. It's > time to move on.
That's why X is abandoned in favor of Xrender/Cairo.
> 'this_weeks_version_of_something_like_render' not > found"
You're talking nonsense. The mistake that X made was to mix server and client structures. Now the FDO moves all structures to the client and communicates with the server only via cairo/xrender.
Moving all structures to the server and let clients communicate with the server via remote procedure calls as Y wants it [y-windows.org] is nonsense, you constantly replicate the client structure to the server. X11 is already bloatware, a Y server would not even run if only 800K is available.
> [...] side of a 56 kbps link.
Nonsense again. NX cuts the X11 network overhead so that you can run a kde desktop via a 14.4 link (www.nomachine.com)
your wrong.
2 0C HEVY%20ASSEMBLED.htm
http://www.speedomotive.com/450%20CUBIC%20INCH%
OS X ships with an X server by default now.
DNA just wants to be free...
It's not one of the "greatest technologies ever created in computer world." You've got to be kidding me. Then you go into a long advertising spiel on X11.
Anyway, here are the reasons listed in Mark's paper:
"The X Window System [23] is the de facto standard graphical user interface (GUI) system on UNIX and UNIX-like platforms such as GNU/Linux. However, as X approaches its 20th year, signs of its age are beginning to show. Commonly cited problems with X include:
Aside from the user interface inconsistency, the lack of standard components also makes internationalisation difficult, particularly for languages which require a complex input method.
Although the X protocol supports extensions very well, some of the latest extensions have begun to interfere with each other. For example, when Xinerama (the extension which allows X desktops to span multiple monitors) was first released, it broke XVideo (the extension which allows X to use hardware accelerated overlays for video play back). The 'fix' for this was to allow XVideo to only work on the primary display. The latest extension, XRandR (Rotate and Resize), is also known to break many older applications which assume that the screen size will never change.
Further, the internal design of X itself is outdated. Even adding a simple feature, such a stranslucent windows, requires large changes to the server [17]. Because of the requirement to be backwardly compatible, these features must be implemented for everything that X works on, including two-colour displays.
This post isn't flamebait, and it isn't offtopic: It states a legitimate opinion wrt the source control management of XFree86 forks, along with some level of basis for that opinion. Moderators who call something "flamebait" because they disagree are abusing their privilege.
Lets see..
Till about 6 months ago, my fastest machien was a slightly overclocked dual pII 333 machine with 512mb and a gforce4mx graphics card.
On this machine I used to run Windows 2000 and FreeBSD 4.x (both using drivers obtained from nvidia)
How is it that on FreeBSD I had far less performance issues playing hires divx files or playing (identical!) games?
If you don't design a graphics system with remote use in mind, there is very likely no good way to add it later on.
X11 was X11 right from the start as far as I remember. The 11 stands for one megapixel (as in a display 1000x1000) and one MIP (million instructions per second). At the time X11 was conceived this theoretical platform spec was thought to be about 5-10 in the future (ie mid 1980s) but actually such a machine was available in only 2 yrs.
Yeah, I make shit up too....
Can I get an eye poke?
Dog House Forum
Did a bit of scrounging on the net and it looks like you're right. I find reference at http://www.mit.edu/afs/athena/system/ to X10r4 whilst the oldest system I recall was X11r4 which I downloaded over a 2400 baud modem (it took 5 days!) to build for a military project I was working on at the time. Back then Sun had just come out with workstations with X11 or SunView(?) (I think it was) as the GUI.
Whilst Apple did a good job with the Lisa then Mac it wasn't until the Amiga with windows structure very similar to X11 (or identical much of the time, if you look at the code) that things really took off for X.
Good UI for programming started when I saw Compass Pascal under CP/M in 1982 and used it for several projects controlling laboratory equipment (on an AppleII with an 8080 card). Compass (a German comany I think) was bought out by an unknown (at the time) US company called Borland and re-released as pretty much the identical product call Turbo Pascal and suddenly it was easier for every man and his dog to code - largely because edit/debugging/run was so much faster. It was text based but very well done.
Back to the X name though - I'd love someone from the original project to give some insight into the origin of the name. Clearly the story I read 15+ years ago was incorrect.
pithy comment
I'm sorry, you are the flamer, I would have said nothing if you had remained polite and objective.
All I have to say is that CVS is still tremendously useful, and you have no right to call it a POS. It's not remotely polite. I'm not a CVS developer but I've been using it for many years, and I've also used others SCS over time like SCCS, RCS, Perforce and now Subversion.
I've never used CSCVS, I'm sure it's tremendously useful, I mean no disrespect.
In defence of CVS all I'm saying is that it has limitations, but that these limitations are well known and well documented, and that CVS is as bug-free as any piece of software can be. When Arch has 1% of the usage that CVS has had I'll certainly consider using it.
I know very well that CVS doesn't do moves and renames nicely, but you can still do them, and you can still do branching. The point is that for the truly vast majority of software projects out there CVS is enough, and if not Subversion certainly is. Only a few projects really need Arch's features and very few of them should be ready at this stage to trust Arch's implementation.
I'd love to see the proportion of software projects, free or otherwise, that use 10,000 files. Most start with something like 10 and die before they reach 1000. The main problems I face with people who use any version of source control systems is not branching or atomic operations, it's developers who can't be bothered double-checking their changes and only submit a subset of what is needed for the others to compile the tree again. Now if there was a SCS that could force users to do this double checking that would be a real advance.
Moreover Arch might be very nice but it is maintain by a single lone guy who says on his home page that he doesn't have the time or resources to maintain it as he would like to. Would you trust your whole source tree with a piece of software that might go totally unmaintained tomorrow?
AFAIK sourceforge does not support Arch or Subversion. You have to use CVS, full stop. Maybe that means something?
Finally I really can't see why you think a particular project is not worthwhile because it doesn't use Arch as its SCS.
wow, thats some messed up logic. So Linux distros should be at the behest of every other project?
.tar.bz2 it has too download full of patches. And it might download some more patches as well.
i ding-subs tance-abuse-job.patcht tern-die-die-die-v2.pat chl _4.3.0-nv-unresolved-symbols.patch
The issue here isn't the kernel being used at all. I've never heard of anyone complain about XFree86 not being Linux-friendly or it favoring *BSD or something.
You can clearly see XFree86's problem when you compile it in Gentoo. The amount of patches that the Gentoo dev's have added to XFree86 number in the dozens upon dozens. Sometimes Gentoo will have a patch here and there. Sometimes to makefiles and the like, just to adapt a package to Gentoo's way of compiling. Or perhaps some bug fixes that come out shortly after a release. But with XFree86, its clear that something is wrong, there's a whole extra
And they're important (and some humorous) sounding patches too, heres a sample:
0182_all_4.3.0-redhat-xlib-linux-fix-avo
0199_all_4.2.0-die-ugly-pa
0230_all_4.3.0-craptastic-cast.patch
5325_al
Note that they have to have a 4 digit number to keep track of the patches.
Now granted, XFree86 is not just fairly large (though projects like KDE are far larger), but is also very dependent on hardware making debugging difficult. Which is why XFree86 more then any project needs requires the cooperation and to graciously accept the cooperation of the distros, whose users will be on the frontline of finding bugs.
Simply untrue. LaTeX, perhaps, is as bug-free as software can be -- but just because you aren't hitting the CVS bugs doesn't mean they don't exist.
CVS gives different paths to server-side ,v files depending on which command is being called. CVS fails to quote log entries to disambiguate them from end markers, making it impossible to machine-parse output. CVS has a vast array of designed-in pserver security holes. I've seen CVS corrupt CVS/Entries files on more than one occasion -- quite a few, for that matter.
If people chose their operating systems that way, we'd all be running Windows.
No, you can't -- not for any reasonable meaning of the word. When I do a version-controlled move, I expect the following to happen:
No mechanism for attribute #1 is available which does not violate attribute #3. Attribute #2 is not available at all. Thus, for any useful definition, CVS does not support moves.
For some crippled value of "branching". Branching is vastly less useful without history-sensitive merge support, because it's impossible to apply a patch to multiple branches and then later merge them without creating tons of extra work -- so people don't do it as much, even cases where (if it were truly a low-cost operation) it would be useful.
Of course. If you know that renames and moves don't work well, you don't get in the habit of using them much, so you don't "need" them. If you know that branches are a hassle, you don't use them much, so you don't "need" them. If you've never had history-sensitive merge support or distributed repositories, you don't know how useful they are, so you don't appreciate how much power they give you to choose your workflow.
Arch's implementation is one helluva lot more trustworthy than CVS is, or SVN's. Let me explain why:
Arch, unlike SVN and (optionally) CVS, has no "smart server" and no data storage format more complicated than a bunch of tarballs with patch files in them. This means that even if Arch itself were to disappear tomorrow, I could take my Arch archive, unpack the files with tar, apply the patches via a shell script (though Arch's dopatch tool is easier), and have my source control history back. These tarballs, once they're committed, are never changed -- unlike a ,v file that's rewritten whenever new history is created or a database that's likewise constantly modified. This likewise means that mirroring an Arch archive is just a matter of mirroring a collection of directories and tarballs; these can be written to CD or any write-once media if need be, because history is never allowed to change!
In short, Arch's core design is far better built for trustworthy source control management than any of its competitors.
The source control repository at my workplace is quite a bit larger than that -- and we're a fairly small and new company. As for Free Software, I'd look in the direction of Emacs, Linux, the BSDs, GCC, XFree86, and quite a few other p
By the way:
I was something along the lines of employee #17 at MontaVista Software. When I came on, we were using BitKeeper under the zero-cost license (since we were using it to work on free software and could cope with having our changelogs published to the world). Larry McVoy changed the license after that to force us (yes, this was quite explicitly aimed at us) to switch to a commercial license; instead, being a budget-constrained startup, we chose to switch back to CVS.
First, though, we put three of our best developers on a side-project of trying to clean up CVS and adding support for some couldn't-live-without features (excepting, of course, those which couldn't possibly be added due to design constraints). When that project was cancelled (post-cleanup but pre-new features), they'd put together a CVS fork with all the functionality of the original enabled -- with 25% the number of lines of code! The three people I know who are most prone to making profane remarks regarding CVS, its design and implementation, are in fact programmers with intimate knowledge of its codebase. (Want names? Mark Hatle, Mark Ferrell, Paul Mundt. Feel free to do a bit of googling to determine their bona fides, but please don't bother them on account of this thread).
So -- I wouldn't be so sure that CVS developers would be so offended by my statement that CVS is a POS; the ones I know all agree.
all this fork talk has me hungry . . .
forks are good when they lead to replacements, or fold new ideas back into the trunk after compatibility and bugs have been worked out. A great example is mozilla.org.
forks are bad when they result in a splitting of efforts over technical or ideological differences when they cant be overcome and are lost, since, to date, most recent forks off xfree86 4.x have failed to generate any major replacement for, or many improvements to xfree86.
X11 is old, perhaps it doesn't need to be replaced, or re-written from scratch, but it does need a major technical review on all levels on how to clean things up and re-focus for the future.
Mozilla often puts new ideas or redesigns of existing infrastructure on a branch when major changes are made, and once deemed stable they are incorporated into the trunk.
How often in recent years have any major changes/clean-ups been made to xfree/x11? From Moz 1.0 to 1.6 huge changes have already been made in the underlying code, and mozilla is no light project in itself.
People seem to thing change is bad, why reinvent the wheel they think. Well, change doesn't mean the old code was bad, but there are alot of times when new ways can replace old ways (the computer field wouldn't be progressive if we stopped inovating), and do things better. Yes sometimes things break (like 10+ year old un-modified/unmaintained programs) but it doesnt mean the code was written poorly, just the programmer never envisioned it would be in use 10 years later. We have to understand that a 10 year life span for an app is in my opinion an EOL program.
Anyone still using a 10 year old app that breaks with changes in how X fuctions, lets be honest, is probably NOT 1) running the most up to date hardware anyways 2) caring about the prettiest interface 3) updating their system with anything more then security patching often if at all.
Compatibility is neccessary, but often leads to its own caveats of hackishness when its not designed in. Perhaps the technical review of x11/xfree86 could encompass looking at just what backwards compatiblilty can be reworked into a refreshing, what can be sidestepped with re-design, and what can be dropped.
I don't have any easy answers, but to me, the issue is as simple as a bunch of stubborn ppl on both sides unwilling to check egos at door (someone needs to take the lead in X and be the next benevolent Linus) and do whats best to take a NIX GUI to the next level.