The Power of X

An anonymous reader writes "The license changes in the last version of Xfree86 have caused many distributions to reject the project in favor of the forked X.Org X server. As X.Org prepares to release the second version of the X.Org "monolithic" X Server (dubbed version 6.8), Ars Technica investigates the future of the X platform, as cooperation between X.Org and projects like GNOME and KDE begin to take take hold at freedesktop.org. Already host to an impressive array of projects, it appears that freedesktop.org will become the hub in which other Free Desktop projects can collaborate. Daniel Stone, release manager for freedesktop.org, gets into the details on how it's all going to work, in conjunction with freedesktop.org's upcoming platform release."

Comments

So...

[Anonymous Coward]

The next X.org release is X, free, 6.8?

Re:So...

[Anonymous Coward]

The next X.org release is X, free, 6.8?

And yet there was an "Xfree68" for Motorola 68000 family machines. Back when the "Xfree86" code really did only run on the 80x86 chips.

Re:So...

[Fry a Lad Up]

Now you can get even better windowing with X-free UNIX: MacOS X. Huh!? -- Some years ago, I convinced a friend to get toe-clips for his bike. A week later I was raving about my new clipless pedals. He felt betrayed. :-)

Progress

[Zorilla]

Looks like the original XFree86 project was going nowhere fast. The distros making the first move to X.org want to make some progress to making Linux (and other Unix-types) ready for the desktop. Hopefully, X.org is the first sign of progress to a backend which will eventually be able to do things a modern desktop will need to do.

Re:Progress

[Anonymous Coward]

One of the things that has always bothered me about XFree86 in the past 6 years I have used linux is XFree86's kind of lag in new releases... development seems to move at a snail's pace, and let's be frank, it's almost the same as it was back in the good ol' unix days.

I for one enjoy X.org and a windowing system that can hopefully be kept up to date and have more active development.

But my question is... how many more forks will we have?

James Carr

Re:Progress

[Zorilla]

But my question is... how many more forks will we have?

Think of it this way, Linux is like a fancy dinner, just remember that this fork is for the sala- NO, NOT THAT ONE!

Re:Progress

[jusdisgi]

I disagree. You hear a lot of bullshit from a lot of people bitching about all the things "wrong with X", but rarely from a well-founded technical basis. More often it's from either a "why is X such a bitch to configure" or "gee those XFree guys are a bunch of assholes." In the distro community I have not seen real dissatisfaction with the technical side of XFree86 in the last few years.

That said, it has long been true and well-supportable that those XFree86 guys have definitely been a bunch of assholes for a long time. They maintained a really closed community which gave the appearance of complete disdain for what anyone else wanted out of X. Whether their actual behavior was in that mode is arguable (recall the massive enhancements of XFree 4), but they certainly didn't like to "play ball" with the rest of the community.

Then of course this license thing was the last straw, and that's what forced the distros' hands...they couldn't build their systems at all anymore when core components were GPL'd and either linked to XFree stuff or used its code.

In other words, I'm not sure how much this will impact the technical progress of X...but it's certainly good to get a broader base of people working on it, and a more open development in general.

Re:Progress

[Anonymous Coward]

or the lame technical arguements built on things the people making the argument could never comprehend.

im amazed at how fast the X.org crew is coming along.

excellent work by them. Keith was working on things prior the "official" x.org releases (or even their commitment. but still it is impressive.

i am very pleased that some team took up the task of bringing X into the new era of computing. by the time longhorn even ships X.org will blow them away. and OSS will have an awesome windowing system that is completely open

Re:Progress

[Anonymous Coward]

I have not seen real dissatisfaction with the technical side of XFree86 in the last few years.

Have you been listening?

(A) There's apparently a ton of longstanding problems with the XFree codebase that are only now being addressed, both in fixing the current codebase and in a longer-term massive redesign/rewrite. The response for years has been "Well, it works..."

(B) There's been an emormous amount of criticism of X protocol design, going back 20 years. But with the rise of OSS frameworks, this reached a breaking point. The X response has been that the Toolkits serve the Windowing System and "Well, X11 is X11..."; rather than the more sensible attitude that the Windowing System should serve the Toolkits (and user programs). This is probably the biggest philosophical difference giving rise to the XFree fork.

The licence change was portrayed as a legal issue, but it was really just the final case of the "XFree Guys are Assholes". If the project wasn't already forking, there would have been no legal shit stirred up.

Re:Progress

[Performer Guy]

If you're talking about the DRI then I have to comment that direct rendering is nothing like DirectX 3. The DRI is an underlying mechanism to get 3D drivers to interface with hardware efficiently and in a modular way, it doesn't affect the higher level ABI in any way what soever. Something like the DRI is essential for efficient hardware acceleration on modern hardware with complexities like high speed DMA from the card. It is completely invisible to applications that don't care. It is not a hack and it's nothing like the crimes Microsoft committed with D3D.

Re:Progress

[jusdisgi]

I have not seen real dissatisfaction with the technical side of XFree86 in the last few years.

I wish you wouldn't remove the In the distro community from my quote like that...it's pretty integral to my meaning.

There's apparently a ton of longstanding problems with the XFree codebase that are only now being addressed, both in fixing the current codebase and in a longer-term massive redesign/rewrite. The response for years has been "Well, it works..."

And this is pretty much precisely what I meant when I said, You hear a lot of bullshit from a lot of people bitching about all the things "wrong with X", but rarely from a well-founded technical basis. If you are out there someplace thinking about how your argument was from a well-founded technical basis...I hate to be the one to tell you, but.....

Re:Progress

[jusdisgi]

Please see previous comment about bullshit pseudotechnical arguments from people who don't know what the fuck they are talking about.

Thanks

Re:Progress

[Midnight Thunder]

development seems to move at a snail's pace, and let's be frank, it's almost the same as it was back in the good ol' unix days.

While I love open source, sometimes the fact that it is done for nothing is one of the things that ensures it is developed slowly. Unless you are a full time student, most people are working a day job to put food on the table. Without the cash motivation it is not always easy to spend the time and effort necessary to make a great project. I am not saying the money is what is important to them, though being comfortable, being able to buy a workstation and not living on the street is.

I don't know how many /.ers actually give a donation to projects that they use a lot, but don't contribute to? Maybe a poll is in need?

Re:Progress

[Anonymous Coward]

No, it was supposed to be pointers to technical arguments, thats all.

Maybe you could explain what you mean by the "in the distro community" part then.... Of course they can make a package out of it, but certainly part of the Distro world wants to deliver a better desktop to their users, which is why they are funding X.org, Freedesktop.org and others who actually addressing technical issues.

Re:Progress

[cjpez]

Direct Rendering is nothing more then a hack

How is DRI a hack? I've set up hardware acceleration on nVidia, ATI, and Matrox cards (a long time ago for Matrox, though) with no problems. I can code up a simple GL program in minutes and have it running in hardware acceleration. Where's the problem?

Re:Progress

[RAMMS EIN]

I am afraid I have to disagree with your post on about every point.

``You hear a lot of bullshit from a lot of people bitching about all the things "wrong with X", but rarely from a well-founded technical basis. More often it's from either a "why is X such a bitch to configure" or "gee those XFree guys are a bunch of assholes."''

Actually, I think the most persistant complaint is that X is slow. Whichever way you turn it, and whether or not this is actually XFree86's fault, this is true. X apps feel slower than GUI apps on different platforms, even when both use unaccelerated VESA. XFree86 wastes quite a bit of time on expose events, rather than remembering the contents of windows. Finally, even when client and server are on the same machine, display data typically needs to be copied, making performance about half as good as it could be.

``Then of course this license thing was the last straw, and that's what forced the distros' hands...they couldn't build their systems at all anymore when core components were GPL'd and either linked to XFree stuff or used its code.''

No applictions (X clients) have been affected by the license change. See this discussion between RMS and David Dawes, specifically thi message (which gives a good overview) and this one, in which David Dawes writes:

In the meantime, I
will defer applying the licence change to any client-side library code.

In short, RMS was expressing concern that the new license would disallow GPLed applications to be linked against XFree86 libraries. David Dawes offered to install a policy that would ensure GPL compatibility for those libraries that applications link to. He also pointed out that XFree86 holds the copyright on only a fraction of the code, and the rest is under various licenses, including the original BSD license.

It's also worth repeating that the reason that GPLed code cannot be linked with code under the new XFree86 license is not in the XFree86 license, but in the fact that the GPL specifically disallows this.

Regardless, I'm still happy that many distros switched. fd.o was doing some truly great things with their X Server, but without massive adoption it would probably have taken a long time for these features to stabilize and become available to a large public - in part because the code was incompatible with the XFree86 base.

Re:Progress

[jsebrech]

One of the things that has always bothered me about XFree86 in the past 6 years I have used linux is XFree86's kind of lag in new releases... development seems to move at a snail's pace, and let's be frank, it's almost the same as it was back in the good ol' unix days.

Which, together with the license change, is the reason people have given up on xfree86. X.org 6.8 will include all the flashy cool new stuff people have been talking about for years, like translucent goodness a la mac os x.

I for one enjoy X.org and a windowing system that can hopefully be kept up to date and have more active development.

But my question is... how many more forks will we have?

Given that X.org is the original X foundation that has been maintaining the X11 codebase XFree86 split off of, and all the non-xfree86 X projects are now basically working under the X.org umbrella, I wouldn't say that we're seeing all that many forks.

Re:Progress

[be-fan]

Do you have the slightest idea what you're talking about???

A) The Direct Rendering Infrastructure is hardly a hack. It's a design that is optimized to how modern hardware works.
B) The DRI is nothing like DirectX 3. DirectX 3 allowed apps to directly touch video memory. Modern graphics cards don't like that. That's why both DRI and new versions of DirectX don't let you get a pointer to video memory.
C) The lack of responsiveness in some X apps has little to do with the raw speed of drawing, and more to do with dumb applications and poor schedulers.

Re:Progress

[be-fan]

A) If people say X is slow, and it turns out not to be XFree86's fault, then the people are wrong. They can say that OpenOffice is slow (it is) or Mozilla is slow (it is) or GTK+ is slow (it is), but saying X is slow is wrong (it's not, as proven by apps that *do* run fast on X).
B) Only OS X remembers the contents of windows. Most other OSs do not. So it's not a problem limited to X, and in general it's not a problem that would be a big deal if dumb applications didn't handle EXPOSE events so badly.
C) The buffer copy would only make things twice as slow if the *only* thing the server needed to do with each command buffer is copy it to the graphics card. Of course, the actual processing is much more complex than that. In practice, unless you're doing something stupid like drawing single pixels at a time, the overhead imposed by the X protocol is dwarfed by the actual cost of drawing.

Re:Progress

[TheRealMindChild]

A) Sure it is.
B) Again, sure it is. Maybe not in the way you are thinking, but that is a bad assumption you made yourself. Sure, you dont use COM interfaces to deal with it, but if you ever worked with either, you can, without issue, see quite clearly how underneath they are damn similar in concept.
C)Again, bad assumumtion on your behalf. I said X is not as responsive as a desktop windowing system should be. I said nothing as to why. If it has to do with something unrelated, then I think that the windowing system needs to take an approach that overcomes such.

Re:Progress

[RAMMS EIN]

``The buffer copy would only make things twice as slow if the *only* thing the server needed to do with each command buffer is copy it to the graphics card. Of course, the actual processing is much more complex than that. In practice, unless you're doing something stupid like drawing single pixels at a time, the overhead imposed by the X protocol is dwarfed by the actual cost of drawing.''

Which is why X is so fantastically efficient, right?
The communication protocol introduces overhead (even when run locally), which is why it is more efficient to use shared memory. However, even shared memory induces the overhead of one extra copying phase. And something stupid like drawing single pixels is pretty much exactly what you do for displaying movies, rendering text (save for X core fonts), and possibly other tasks.

Re:Progress

[TheRealMindChild]

See http://slashdot.org/comments.pl?sid=119444&thresho ld=1&commentsort=0&tid=104&mode=thread&pid=1007811 6#10078348

I said NOTHING about Direct3d and I said nothing about the ABI. So no, that isnt what Im talking about.

Re:Progress

[TheRealMindChild]

Because you can get it to work, doesnt mean the way it works is a good approach. See Windows and MCI.

Re:Progress

[hundalz]

But my question is... how many more forks will we have?

As many spoons and knifes that we have :D

Re:Progress

[cjpez]

Okay, so in what ways is it broken? That's what I wanted to know, because I've never experienced any problems with it.

Re:Progress

[lscoughlin]

"In the distro community" is a pointless addition, and also kind of silly.

End users tend to X, which means vedors have an issue with it, weather they choose to address it or not.

As to there not being technical issues with X? Well lets start with that wonderful configurability argument. I'm sorry, painful and error prone configuraiton is a technical problem. It's indicative of poor design, and messy code. For example, Xinerama breaks xvideo, and dri, sometimes. When doesn't it? Ask the magic eightball. Then there is XRandR, which breaks all kinds of shit. I got two more fun words: alpha channel. And lets not even get into fonts. Granted anti-aliasing post-dates X's design, which is a perfectly reasonable argument. However, it comes with the point that X's design is out-dated and insufficient for modern display requirements.

X is sufficient for the majority of computing display needs. But the number of requirements it does not deal with in an adequate manor, or at all, is rapidly growing. So to address the future, and hell, even the present, X either needs to undergo some serious restructuring, or be replaced.

Re:Progress

[Anonymous Coward]

And what do you mean by that? There is NO information, so I suspect your handle should be "The no Mind Child".

All you have said "Argh, it's a hack! Windows dows it better! And I find it slow, so it must be the DRI!".

Try using nvidia hardware and running UT2004. Check the FPS. Linux will be faster (if not by much) than Windows. Nvidia uses DRI to access hardware on Linux. Therefore, being slower is not DRI.

Check out Windows95 on the same machine as your WinXP. Notice how slow WinXP compared to that?

Re:Progress

[be-fan]

If you actually benchmark the damn thing, you'd see that it is pretty damn efficient. With a good set of drivers (NVIDIA), 3D is as fast as on Windows, and 2D is within 10% (go to www.rocklyte.com and look at their benchmarks comparing Athene, X, and Windows). My own benchmarks show that with reasonably-sized primitives, you can drive the graphics card at about half it's memory bandwidth, which is very efficient for a general-purpose 2D protocol.
Also, you're obviously not a programmer if you think people use single pixels to draw movies or text. Movies use the XVideo extension or OpenGL, which stream pixel data in batches to the graphics card. Text drawing uses server-side pixmaps of each glyph, and the normal case involves simply sending some commands to string together the glyphs that are already in video memory.

Re:Progress

[be-fan]

1) Sorry, you attempted proof-by-assertion. That is an illegal move in any civilized discussion. You lose, but thank you for playing, try again anytime!
2) They aren't at all similar in concept. If you think so, describe each system and show how they are similar. DRI is much more like modern versions of DirectX than DirectX 3.
3) You said X is not as responsive as a desktop windowing system should be. I pointed out that it is actually quite responsive (as evidenced by the fact that there are fast X apps, like most Qt ones). No assumptions on my part. I think you meant to say that "this or that application is slow or this or that toolkit is slow" which is something entirely different than saying X is slow. Also, suggesting that X should somehow do things to fix broken toolkits just doesn't make any sense. What do you suggest it does? The common answers of "get rid of the network protocol" or "put X in the kernel" will do jack-shit to help apps that redraw the entire scene when a few pixels change.

Re:Progress

[imroy]

My understanding (I've never actually looked too hard at this) was that with videos/movies, you use the Xv extension. That allows the app to decode into a shared memory segment (like the old Xshm extension), and leave the graphics card to do the scaling and YUV=>RGB colour space conversion. You certainly do not send every pixel over the X socket. Otherwise my 1600x1200 display would grind to a halt when playing a movie fullscreen. In reality the player only uses a few percent of my AthlonXP/2400 to decode a DVD or XViD video.

Re:Progress

[Anonymous Coward]

One would assume if he was talking about that post, he'd have replied to that post. Instead, he replied to this post, containing " It reminds me of DirectX 3 on windows 95... and even that was more elegant an implementation"

Re:Progress

[Anonymous Coward]

DRI ISN'T essintial for efficient hardware acceleration. NVidia drivers don't use DRI and they are by far fastest drivers available for Linux...

Re:Progress

[Dwonis]

Nvidia uses DRI to access hardware on Linux.

Are you sure? I haven't looked at the NVidia drivers lately, but a year ago (when everybody else had already switched to DRI), NVidia's drivers had not, and there didn't seem to be any plans to port them to DRI.

Re:Progress

[Dwonis]

X11 (the protocol) is perfectly fine. XFree86 (the implementation), on the other hand, is a legacy piece of crap.

Re:Progress

[noselasd]

- Which is why people came up with Xv.

Re:Progress

[RAMMS EIN]

``With a good set of drivers (NVIDIA), 3D is as fast as on Windows, and 2D is within 10% (go to www.rocklyte.com and look at their benchmarks comparing Athene, X, and Windows).''

Ok. That benchmark even seems to use unaccelerated video, so I was all wrong. Sorry to have spread FUD.

Re:Progress

[RAMMS EIN]

``Xv extension. That allows the app to decode into a shared memory segment (like the old Xshm extension), and leave the graphics card to do the scaling and YUV=>RGB colour space conversion.''

Completely right. I was talking about unaccelerated video in my original post. However, be-fan proved me wrong on that as well. I stand corrected.

Re:Progress

[Curtman]

I see you've played knifey spoony before.

Re:Progress

[jasontwarnock]

It's a technical problem with XFree86, and X.org(as of right now), however it is not a technical problem with X, those are implementation problems, X(the protocol) does not define how to configure hardware, and as for fonts; server side(the application) fonts(which permit antialiasing, but take up more bandwidth) are better then the old method(client side; fonts in the X server), which was un-natural and was only that way because at the time of the creation of the protocol networks did not have much bandwidth.

Re:Progress

[ultranova]

DRI ISN'T essintial for efficient hardware acceleration. NVidia drivers don't use DRI and they are by far fastest drivers available for Linux...

DRI is an interface between kernel modules and X drivers. Since NVidia supplies both its own kernel module and X driver, it can (and has) replace the DRI interface with another "NVIDIA" interface. However, this Nvidia interface serves the same purpose as DRI, and does it in a similar way.

Re:Progress

[the_2nd_coming]

probably very few (or more likely some experimental forks like you see in Linux which later if they pan out become part of the main tree) since X.org is very open and welcoming.

Re:Progress

[FooBarWidget]

It doesn't take up more bandwidth. It's been tested. With the old core fonts protocol, the client downloads x bytes of glyphs from the server. With Xft, the client uploads almost just as many bytes of glyphs to the server.

Re:Progress

[FooBarWidget]

Sigh, there we go again...
First, get this: when you see a problem, there can be a whole bunch of causes. The only way to fix a problem is to correctly identify the cause.

Which you are not doing here. "X apps seem to be slower so X must be slow" may make sense from a non-technical user point of view, but that doesn't mean it's correct. In fact, it isn't.
If you benchmark things and stuff, you'll see that the problem is not in X itself: it's in the toolkits. So if people listened to you and ditched X, we'd still have the same problems because the toolkits are still slow.
Try using something like WindowMaker and some non-GTK non-QT apps. You'll find that they usually respond significantly faster (not on my machine though; Athlon 1.4 Ghz here, I don't find X apps slow).
And try playing 3D games. Look at the high framerate (assuming you're using a good card and driver, like NVidia + vendor drivers). How's that possible for a windowing system that's slow?
I've written several testing apps, for Windows and X. On both platforms, I get the same frame rate.

Expose events: Windows does pretty much the same thing. If I wrote an app that only responds to the paint event once, and I move a window above it, you'll see that the window won't be redrawn. Windows in fact uses the very same expose mechanism.
X *does* in fact have a feature which allows you to save the content of the window. It's saved Backing Store and Save Under (I think). QT and GTK don't use it except when popping up a menu. Ask the toolkit authors why they don't, because I don't know.

Data copying: data is *not* copied when transferring pixmaps, which is about 90-95% of the traffic. On localhost, XFree86/XOrg uses shared memory for that.
On localhost, normal X messages are transferred via unix domain sockets (not TCP sockets), which are almost as fast as shared memory (at least on Linux). Remember, this is small amount of data.
No modern windowing system allows the app to touch the hardware directly. Not Windows, not MacOS X, not BeOS. In fact, Windows internally uses the same message-based communication system as X. Windows apps don't draw directly to the hardware.

Re:Progress

[arkanes]

I guess what you consider fast is important. I'm in the process of moving 100% away from windows, and while some things are much faster (a 2d polygon rendering test was 10x faster than the same test, on the same machine, under Windows), things like moving and resizing windows is much, much slower and more flickery. The resizing and repainting in response to mouse resizing is very slow - it lags well behind the movement. Revealing another window by moving one over another streaks and tears. It's annoying. It's a little better under KDE than Gnome but far worse than Windows under both of them.

Re:Progress

[be-fan]

I'll buy that (it depends on your machine), but they have nothing to do with X. Resizing and moving is limited not by the raw display performance of X (which is all X really does), but interactions between X, the window manager, and the toolkit. It does no good blaming X for it when the blame lies elsewhere. In particular, there is no synchronization between the window manager and the toolkit, which means the window manager resizes as quick as it can (100s of times per second), creating an ugly visual effect and starving the app. There is a feature in the current draft of the netwm spec that allowed for synchronzation. On my desktop (2.0GHz P4), a KDE and Qt version with support for that feature resizes and moves windows as fast as Windows.

Re:Progress

[parnasus]

... will include all the flashy cool new stuff people have been talking about for years, like translucent goodness a la mac os x...

As nifty as this is, I would like to be able to add input devices to the system on the fly. If I'm on the road with my laptop, I don't need a chorded mouse, I just use the glide point. When I get to the office/destination, a chorded mouse is easier to use. As it stands, I have to shutdown X, plug in the mouse, and start X again. (And yes, I have a USB definition for the second mouse using CoreEvents). It's a shame I have to end any running app before using the secondary pointing device because it wasn't initialized at startup.

Re:Progress

[k98sven]

While I love open source, sometimes the fact that it is done for nothing is one of the things that ensures it is developed slowly.

Yes. Sometimes, but not in the caes of Xfree86. There, it appears the main reason was a rediculous level of conservatism from the project leaders, including some of which who no longer even were active developers.

There are plenty of stories out there of presumptive Xfree86 developers who turned their backs on the project after being treated with what they felt was an unfair and arrogant attitude. Many of these are now active in Xfree86.

The problem that it's 'done for nothing', (not true, there are paid developers not living on contributions out there) is actually pretty small, if you're working on a project with a sufficently large interest. For instance, the reason why Linux took off and the GNU Hurd didn't can almost be attributed entirely to leadership differences.

Here the money bit comes in again. When people aren't getting paid, the barrier to exit is lower. You have to be respectful and kind and open and listen. It costs nothing to praise loudly but critizise softly. And be very wary of license changes.

I don't think most /.ers do anything at all. Personally, I contribute code. If I come across a bug in a program I use, or a feature I want. I often fix it. Then I submit a patch for it.

Future contributions are usually determined by the reaction I get. Sometimes, you don't even get one. Some projects don't seem to want bugfixes or more developers. And these are the ones which are prone to forking.

Re:Progress

[Brandybuck]

Try using something like WindowMaker and some non-GTK non-QT apps.

I'll have to disagree with you here, because you're comparing apples to oranges. Those KDE and GNOME apps seem slightly slower and less responsive because they've been jammed packed full of functionality. The KDE desktop does about a hundred times as much as the bare WindowMaker window manager. A KDE application is going to pull in ten times the functionality from the KDE librarieas than a bare xlib application will get from X.

The very same thing holds true under Windows, but few people are honest enough to admit it. Windows 95 was snappy and responsive under a 90MHz Pentium and 8MB RAM. But try putting Windows XP on the same system! The difference is that Windows 95 didn't have a tenth the functionality of XP, and that's not even counting the eye candy. When I compare the workstation in front of me between Windows XP Pro and FreeBSD with KDE, I am finding that while XP apps may sometimes start up faster, they most certainly are NOT more responsive. Your situation may be different, but in my work environment that is what I am seeing.

Re:Progress

[nathanh]

XFree86 wastes quite a bit of time on expose events, rather than remembering the contents of windows.

That is true in practise, though XFree86 has offered backing store for windows for ages. It just seems nothing ever used backing store, thus the expose events. The new Composite extension means that all windows effectively have backing store because all windows are drawn offscreen. With Composite running you can see a difference: exposed windows redraw significantly faster. That's because the exposed area is redrawn from the offscreen window instead of the exposed area being recalculated by the client. This also avoids a round-trip between the server and client (it's all done server side). The failing is that there is still some tearing because the redraw isn't sync'd to the vertical retrace. I think Keith and Jim have plans to resolve that problem as well.

Finally, even when client and server are on the same machine, display data typically needs to be copied, making performance about half as good as it could be.

That bit is not true. The X11 protocol stream is copied, not the display data. The performance hit was measured in the mid-90s at between 3-5% for operations like line draws, rectangle fills, screen blits, etc. The costs are dependent on the CPU speeds; as CPUs have gotten faster the costs of the X11 socket have decreased. I wouldn't be surprised if the actual costs of the socket are now sub-1%.

Of course, there are exceptional cases where significant display data is sent over the pipe. For example, Pixmaps and 3D texture data. There are extensions to handle those cases with zero copies.

Re:Progress

[bit01]

how many more forks will we have?

As many as people want.

Not as many as M$ wants to maximise a monopoly revenue stream.

What more do you need to know?

---

It's wrong that an intellectual property creator should not be rewarded for their work.
It's equally wrong that an IP creator should be rewarded too many times for the one piece of work, for exactly the same reasons.
Reform IP law and stop the M$/RIAA abuse.

Re:Progress

[dossen]

Have you remembered to point your secondary mouse (defined as SendCoreEvents) at /dev/input/mice or the equivalent on your machine. I have personally had plugging/unplugging of the external mouse (USB) working on my laptop for several years.

Re:Progress

[LazloTheDog]

Let me be the first to mark this historic occasion - a poster on Slashdot admitted he/she was wrong!

JM

Re:Progress

[glitchvern]

You can force it to use backing store by starting the X server with the +bs -wm options. According to Alan Cox this is better and feels snappier on most setups but will kill a tiny machine.

Re:Progress

[Anonymous Coward]

DRI isn't, no. But DRI is just an implementation of direct rendering, which most certainly is needed. nVidia don't use DRI, but they do provide their own equivalent.

Re:Progress

[Anonymous Coward]

I have this exact setup and don't have to restart X. The second USB defintion using core events should not refer to any paticular mouse device, but rather /dev/input/mice, which will exist at the startup time of the server and receive the events from mice as they are hotplugged and allow you to use them on the fly.

Re:Progress

[Anonymous Coward]

X.org 6.8 will include all the flashy cool new stuff people have been talking about for years, like translucent goodness

I'm glad the dam has finally broken and we can finally get the real productivity and usability improvements we've all been waiting -- ooooh, shiny!

Re:Progress

[Tokerat]

It's also worth repeating that the reason that GPLed code cannot be linked with code under the new XFree86 license is not in the XFree86 license, but in the fact that the GPL specifically disallows this.

After reading over the GPL, I still don't understand this issue. Surely linking does not create a derrivative work?

Re:Progress

[Bush Pig]

Just start from the outside, and work your way inwards ...

Re:Progress

[Bert64]

Well, here atleast i have gpm running in relay mode (otherwise the mouse fucks up when you switch between X and console) using /dev/psaux and type ps2, and i have the usb_hid driver which handles a usb mouse when i plug it in.. Right now i can use a ps2 mouse and a usb mouse at the same time, with both of them fighting to control the pointer - it looks weird but it works.

Re:Progress

[benjcurry]

It is amazing to see the progree that X.org has made, and it makes me feel a bit bad about all the wasted time waiting for XFree86...

How many more forks? We'll have more until one estabilshes itself as the standard... I think X.org is the one, though.

Re:Progress

[benjcurry]

I think the point stands, though: there aren't really (many) technical issues with X...just with XFree86 and their development process. If there were loads of issues technically, could X.org, in a matter of months, be on the doorstep of a release with true transparency, compositing and more, with 3d acceleration to come...I don't think so. It's good to see X being managed by forward-thinking individuals.

Re:Progress

[Performer Guy]

This I already know, I never said it was essential, however NVIDIA drivers have their own closed source dropin replacement for DRI that doesn't play nice with the DRI.

Re:Progress

[Performer Guy]

That's NOT the post I replied to. Just hit the parent link on my post. The post I replied to mentioned the Direct Rendering and said DirectX 3 was more elegant. I chose to mention the ABI as one means of illustrating my point about how easy the DRI is to use. So what the heck are you complaining about?

Re:Progress

[Anonymous Coward]

No modern windowing system allows the app to touch the hardware directly. Not Windows, not MacOS X, not BeOS.

No wonder my 2nd Reality demo by Future Crew doesn't work under Winbloze.

who forked from who..

[martin]

I thought Xfree86 was a fork of the original X11 development camp and that X.org is a refounding of the original X11 camp after lots of splits, esp with alot of Xfree86 dev guys getting annoyed and going 'back to their roots' as it where..

Could be wrong (and frequently am)..

Re:who forked from who..

[AKAImBatman]

As I understand it, you're right on the first part. XFree86 did fork X.org's work. The part that you're wrong on is that X.org didn't use XFree86's code. XFree86 was a fork specifically designed for the x86 platform. X.org didn't have that, and thus had to patch their codebase from XFree86's codebase.

Clear as mud?

Re:who forked from who..

[martin]

what I meant,

But from what I understand it alot of the Xfree86 guys are part of the X.Org group after a falling out and X.org going quiet for some time..

ie the code forked from X.org to Xfree86, but people *moved* from Xfree86 to X.Org

Re:who forked from who..

[AKAImBatman]

As I said, it's as clear as mud. You've basically got two groups who are really just part of the same process. Now they've decided to split into two pieces. It's all very confusing. =/

Re:who forked from who..

[bytesmythe]

Dev 1: And X.org!

X.org devs: Yeah! Oh yeah! Splitters! Splitters!

Dev 2: What?

Dev 1: X.org! Splitters!

Dev 2: We're X.org!

Dev 1: I thought we were XFree86!

Dev 2: XFree86... huh!

Dev 3: Whatever happened to XFre86?

Dev 1: He's over there... (points to lone man)

X.org devs: SPLITTER!!

Re:who forked from who..

[alex_tibbles]

just 'cos it has 86 in the name, doesn't mean it doesn't run on other arches.

Re:who forked from who..

[Anonymous Coward]

dude, I"m pretty sure you are completely wrong. The parent poster was 100% right. The first Xorg release a few months ago was *exactly* xfree's code at the last milestone release before the licence changes. Xorg is most definitely a fork of xfree.

Re:who forked from who..

[AKAImBatman]

*scratches head* That's what I said. The parent poster sounded like he was saying that X.org wasn't using XFree86's code. (Which apparently wasn't what he was trying to say.)

Let's Talk About X Baby

[grunt107]

The supposed 'modularization' that is to take place in future 'X' releases sounds promising - release enough to work (or 'major' fixes) and then extremely long development cycles can be diminished.

The one caveat is to not micro-modularize; do not release things for install/upgrade that cannot stand on their own (i.e. - limited functionality vs. not executable).

I would like to see 'X' go on a diet, though (if possible).

X in Windows?

[random_culchie]

What I'd really like to see is some support for X type connections in the next version of windows. I don't mean basing all of windows on X11 but perhaps allow remote windows sessions that are native. Not based on screen redraws like VNC.

Re:X in Windows?

[Zorilla]

It could happen. With each new version of Windows, we see it start to behave more like Unix, to exclude its naughty behaviors like running as admin at all times, but that can only be attributed to old, bad development on the behalf of the third party. At this point, we have permissions at the file system level, home directories (with many preferences stored within them), and the command prompt, which serves a purpose as an administration tool for scripting and such, rather than its previous use, which was to maintain DOS compatibility.

There are other examples, I'm sure. Post them if you got 'em.

Re:X in Windows?

[Anonymous Coward]

Cygwin have started using x.org and their X windows functionality on windows is really pretty good - you can run X apps (either from a remote machine, or the local ones provided by cygwin) in the windows desktop, and they appear in the right place in the task bar as well. I'd recommend anyone to check it out.

Re:X in Windows?

[ch3]

Like X.org in Cygwin?
If you use the latest release of cygwin, it comes with an X.org server that has a rootless WM (like the one in Apple Xfree server). With it, you can run X11 applications next to your Win32 windows as if they were native (same for remote windows).
I use it all the time for remote admin and this is great.

Re:X in Windows?

[dave420]

Seriously not trolling here, but most of that Windows stuff has been out for ages. Filesystem permissions? NTFS introduced those. Home directories? 95. Command prompt? You're serious? ;)

Of course there are going to be similarities - they're general-purpose pieces of software written to use the same hardware. That's why they're converging the way they are (linuxes having start buttons, graphical installs, GUI-everything, etc.)

It's not a "which one's better" argument, just an example of convergent evolution compter-style.

Re:X in Windows?

[Sheetrock]

Cygwin can do this.

It's a bit buggy, but it's 'native' (through cygwin.dll) and gives you X windows that render in the local Windows environment. Better to find native Windows applications, but it'll work in a pinch.

Re:X in Windows?

[Ancil]

Already happening. Windows' new rendering engine, Avalon, is completely vector-based. Here's what one of its designers had to say:

Avalon will support remoting at a higher level than DirectX. When remoting we will not rasterize on the server machine but instead we will send higher level graphics instructions to the client machine and then call DirectX on the client machine.

This will enable us to send less data over the network as well as reducing the server load because all graphics operations will run on the client machine. We also will get higher performance and fidelity rendering and animations because we will not need to round trip data across the network for these operations since they will be retained on the client machine.

This isn't surprising -- once you have a completely declarative presentation system, remoting becomes a lot easier.

Re:X in Windows?

[Zorilla]

Right, they have all been around for a while, but now all those features are becoming more important in the way Windows behaves. (Home directories and profiles on Windows 95 was a joke, but much has changed since then) Hopefully we'll be seeing more stress on only having admin access to important system files. This should be a part of a more general evolution of good ideas in computing.

And I do realize the command prompt has been around quite a while. I meant that its current purpose is to provide a way to do tasks at the command level (such as propagating tasks to the client machine from the domain controller upon logon), rather than a means of backwards compatibility.

Re:X in Windows?

[kahei]

This has been available for a long time, and (the client part) is available by default. mstsc.exe.

Re:X in Windows?

[dave420]

I understand what you're saying, but I still disagree with the command prompt issue. It was always for administration tasks (as batch files (or .cmd in NT) were always the admin's friend, especially when used with the built-in telnet server), and always for backwards-compatability. Obviously, backwards-compatability is a bit silly in a dos box these days, so that use has deteriorated slowly. I don't think it's gone through many changes at all. I know I've always used it for the same old things.

Lots of things about windows95 were a joke, but it was great testing before windows 2000, where the winning features were picked, and the crappiest ones were dumped. I guess any OS has to go through that to really find out what its users want or not. Linux has tried that, too ;)

Re:X in Windows?

[drinkypoo]

Slightly more information: Install cygwin, I usually install the default install, minus emacs, plus the whole X11 directory (then turning off some emacs stuff in there), and explicitly turning on a few things if they aren't already like ssh, ncftp, and whatever else. That part is up to you.

Once it's installed, I make a shortcut for the X server that runs "C:\cygwin\usr\X11R6\bin\run.exe XWin -multiwindow -clipboard -unixkill -nowinkill" and bingo, I have X. There's also another flag you'll need if you want your X desktop to span multiple monitors, but I only have one hooked up right now so it's a non-issue.

You can also turn off -multiwindow and run -rootless (I think) and then use a window mangler to manage your assorted X clients. If for some reason you want X clients and Windows apps to look even less alike, this is how you accomplish that.

Re:X in Windows?

[perlchild]

I am not going from first-party information here, but I'm sure I read for the feature list for terminal servers under windows, that they allowed connections to X. Good luck getting a product further up the "windows tree" down to the masses.

Re:X in Windows?

[Anonymous Coward]

*cough* Are you sure you understood the question ?

Re:X in Windows?

[zorander]

You're kidding Right?

Filesystem permissions on NTFS are a joke. In theory, they'd be great, but In reality, XP opens up everything wide for you upon install. They could take the OS X approach and prompt you for a password upon installing things and have a more secure desktop with no user inconvenience, but as it stands, Spyware X can be installed by a user and affect other users because generally, a user has way too much write permission the way MS has set up XP.

Home directories are worthless unless they A. work right and B. Come with well-implemented and executed file permissions. Again, this is an area in which XP is capable, but the issue of most users in an XP system having far too much power weakens it.

The Dos/Windows command prompt was, is, and always will be a joke. It feels like it was written by someone who never had to be productive at a command prompt. It hasn't had command completion until recently, it doesn't have 10% of the utilities (packaged with windows) that one needs to be productive, and it doesn't have enough device/file mappings to be able to truly take advantage of it. Just cause you can do ls doesn't mean you can do du -md 1 | sort -n or dd if=cf_img of=/dev/sdb bs=2k. As an additional kludge, the cmd.exe window is awful for running an editor in and doesn't resize well.

Just because Microsoft provides some of these features doesn't mean it does it well. Truthfully, I think they should give up on the command prompt. they'll never get it right, and remote administration can be done in a web browser anyhow (webmin, anyone?) It's not like they're going to build software remotely as source code is not a popular distrubution method on windows.

Basically, saying that Windows has a command prompt is like saying that linux has direct rendering support. They're both true, but in such useless ways.

Brian

Re:X in Windows?

[Anonymous Coward]

The DOS backcompat features in NT were always seperate from the native command prompt. They use the same terminal window, that's it.

The main criticism of NT's batch capability is that its mainly useful for simple logon scripts and the like. The real administration interfaces are largly behind objects that are better accessed with VBScript or something.

Re:X in Windows?

[dave420]

I know about the cmd/command.com difference, I was just making a point ;)

Yes, there are very powerful object-based tools in windows. Most recent MS server-based apps have command-line tools for configuration. I added an LCS 2005 server to our domain the other week all from the command prompt. It updated the schemas, propagated them, and configured the user accounts, all from the command line. Hella powerful stuff. Of course, being Microsoft, there is a pretty-pretty GUI alternative, too.

Re:X in Windows?

[uberchicken]

> ... email me conorb at gmail

Hey, that gmail spam filter not working for ya?

Re:X in Windows?

[timmi]

Windows 95 and 98 did NOT have home directories.

there was one and only one "My Documents" folder.

Windows NT on the other hand had c:\documents and settings\~username\my documents

Re:X in Windows?

[Evil Grinn]

The Dos/Windows command prompt was, is, and always will be a joke. It feels like it was written by someone who never had to be productive at a command prompt. It hasn't had command completion until recently

Guess that depends on your definition of recent. Completion has been there since Windows NT.

Re:X in Windows?

[Ancil]

Indeed I did. Most of the other people responding to the question did not, however.

The poster wasn't talking about "X Windows under Windows", that's been done by Cygwin and many others. Let's read the original comment again, shall we?

What I'd really like to see is some support for X type connections in the next version of windows... Not based on screen redraws like VNC

He's talking here about X-like connections, NOT X Windows. In other words, he'd like to see remote terminals based on drawing primitives. Current Remote Windows are based on screen redraws, as the poster mentioned. This is rather cumbersome, and completely impractical on less-than-stellar connections.

Re:X in Windows?

[jusdisgi]

I don't think [the command line has] gone through many changes at all.

Well, no, not yet; it's still cmd.exe, which only had minor enhancements to command. But the parent was talking about where Windows is headed, which makes the command line a particularly fitting example, because long-in-the-tooth will have MSH, which is vastly different and actually much more of a "real" shell.

Re:X in Windows?

[Chester K]

What I'd really like to see is some support for X type connections in the next version of windows. I don't mean basing all of windows on X11 but perhaps allow remote windows sessions that are native. Not based on screen redraws like VNC.

RDP is much closer in implementation and functionality to X than it is to VNC; in that it doesn't send updates to the screen as bitmaps -- it sends font information, strings, window information, and bitmap information for actual UI bitmap objects (i.e., not everything). In fact, it's so tied in with the Windows UI at the lowest levels that for Microsoft to switch to X11 would probably be a step backwards.

Re:X in Windows?

[Eunuchswear]

there was one and only one "My Documents" folder.

Sorry, enable profiles on your Win9[58] system and you get a My Documents per profile (aka user).

C:\WINDOWS\PROFILES\%user%\My Documents or something like that.

No permissions of course.

Re:X in Windows?

[orasio]

That kind of completion sucks.
The NT console is painful.
Plus, it lacks many of the features bash has accustomed us to, for example parameter completion, that is another productivity leap.
When I used NT at work, with just a few months of bash console experience, I ended up installing cygwin, just to have a usable console, where I could launch my batch jobs, and write scriptlets.

Re:X in Windows?

[Anonymous Coward]

You actually use ncftp instead of lftp? For shame.

Trust me, if you ever used lftp seriously you would never want to use anything else. It's so shell-like and automatically handles lots of things for you (mirror).

Re:X in Windows?

[reallocate]

>> The Dos/Windows command prompt was, is, and always will be a joke. It feels like it was written by someone who never had to be productive at a command prompt.

To be fair, DOS was originally created for hardware that barely had enough memory to support a single app, much less a versatile shell and an app. A tiny shell was a necessity. When more memory became the norm, no commerical incentive existed to improve the shell. Vendors like MKS proved it was possible to create a Unix-like shell for DOS, but their target was a tiny niche of the DOS market.

Unix, on the other hand, was created by programmers for use by programmers. Little, if any thought, was given to the needs and interests of non-engineers. That has rather a lot to do with the fact that most people would still not be using Unix even if Microsoft had never existed.

Re:X in Windows?

[afd8856]

I'm sure there will be less then 100 people using the client-server feature of the Longhorn. Considering the resources it needs, the thin-client would need to be a really fat-client, I don't think this is feasible at all.

Re:X in Windows?

[geordie_loz]

Too true, but the most laughable bit about the security of 95/98 was the ability to "Cancel" in the login prompt for users and have direct access to the PC. Sometime things like networking wasn't accessable, but seriously, what does that matter when you could create an account and then login as that user for net access.

Got to give it to microsoft, their GUI's provide access to even the unusual activities like "hacking into a machine" ;)

Re:X in Windows?

[bored]

Windows has a built in "Remote screen protocol" its called RDP. Its whats used by terminal services/rdesktop. It also happens to be orders of magnitude faster than VNC or X. Its also one of the _MANY_ 3rd party components shipped with windows. The original version is written and maintained by Citrix.

Re:X in Windows?

[phoenix_rizzen]

Just to clarify one thing: Windows NT 4.0 has had command, directory, and filename completion in cmd.exe. That was released back in 95 (or thereabouts). Hardly "until recently" as you put it.

The only problem with cmd.exe's completion is that it is set to ESC or some strange character like that, by default. To change it to TAB (or whatever other key you prefer) meant editing the registry or installing TweakUI.

And there have been replacement command interpreters (like JP Software's 4DOS) with these abilities (and more) since the days of Windows 3.1.

Re:X in Windows?

[denlin]

oooooh, that was one of my favorites. & that's exactly how it was. cancel, no networking, add a new user, log off & back in, presto. back in the day, my boss thought i was a genious *picks nose*. well, i left the nose picking as an activity left to my office.

don't mod me as redundant, it's should be more like (mod +1 reminiscent).

Re:X in Windows?

[Rich0]

Now if it could only handle sftp (ftp over ssh) it would be GREAT!

I routinely use mirror to download mandrake patches for a box at home, but then have to sftp the entire directory since it doesn't support mirror, and the box I'm sending it to is too-locked-down for anything else...

Re:X in Windows?

[Anonymous Coward]

As an additional kludge, the cmd.exe window is awful for running an editor in and doesn't resize well.

I run Vim in a cmd.exe window every day. I don't really see what's so awful about it. It has some nice things like colors, 15 pt. andale mono fonts, shell command access with full piping. As for the resizing issue, type alt+space, s to resize. Or you could use the mode con command.

Re:X in Windows?

[Anonymous Coward]

> The only problem with cmd.exe's completion is that it is set to ESC or some strange character like that, by default. To change it to TAB (or whatever other key you prefer) meant editing the registry or installing TweakUI.

That completion registry key used to be set to 0 by default; not terribly useful for 99% of users who don't know about that regkey or how to use regedit. Requiring regedit to enable a feature qualifies as "not until recently" if you mean "a practical, usable feature for non-geeks".

Re:X in Windows?

[bit01]

It also happens to be orders of magnitude faster than VNC or X.

Nonsense. It may be faster than VNC depending on the hardware but it is a fraction of the speed of remote X. Citrix remote desktop is an abortion that makes the GUI practically unusable because of slowness, screen artifacts and lost mouse clicks.

---

It's wrong that an intellectual property creator should not be rewarded for their work.
It's equally wrong that an IP creator should be rewarded too many times for the one piece of work, for exactly the same reasons.
Reform IP law and stop the M$/RIAA abuse.

Re:X in Windows?

[BlueStraggler]

To be fair, DOS was originally created for hardware that barely had enough memory...

To be fairer, Unix was orginally created on a PDP-7, which likely had less memory than the original 16K IBM PC.

Re:X in Windows?

[bit01]

A tiny shell was a necessity.

Nonsense. Other OS' at the time, including RT-11, RSX-11 and others were running in similar memory with greater functionality.

Face it, PCDOS/MSDOS was nothing more than a rebadged hack gotten out of the door quickly for marketing reasons. It worked, but users suffered for a decade as a result and M$ still has a who-cares-about-the-users-as-long-as-we-can-sell-i t culture.

Little, if any thought, was given to the needs and interests of non-engineers.

Since MSDOS/PCDOS functionality was a strict subset of CPM/RT-11 etc. functionality, but incompatible for no good reason, this argument is meaningless. It was a quick hack, nothing more, and M$ managed to gradually leverage that hack into the vendor lockin they have today.

That has rather a lot to do with the fact that most people would still not be using Unix even if Microsoft had never existed.

Nice ad hominem.

---

It's wrong that an intellectual property creator should not be rewarded for their work.
It's equally wrong that an IP creator should be rewarded too many times for the one piece of work, for exactly the same reasons.
Reform IP law and stop the M$/RIAA abuse.

Re:X in Windows?

[reallocate]

It is irrelevant that DOS was a rebadge of a quick hack. It is also irrelevant what Microsoft's attitude toward users was or is. Typical computer users have never wanted to work at the shell prompt. If they did, GUI's would still be in the labs. Command.com might have become the best shell in history, but that would have made little impression on sales.

As for Unix, it has been around longer than Windows and everyone who tried to market it as a desktop for normal users has failed. As Linux is finding out, the best way to convince normal people to use Unx is to hide it. (I say that as someone who's happily spent 80 percent of his computer life in Unix and Linux,)

Re:X in Windows?

[dbIII]

What I'd really like to see is some support for X type connections in the next version of windows

Hummingbird Exceed, Xfree86 for windows ... there are probably hundreds of X implementations on MS Windows. If you are running MS Windows you should be used to getting third party software.

As for exporting windows applications to other machines via X, that worked with a third party application on NT3.51 - but MS made sure that it would never run again after that by closing their API.

Re:X in Windows?

[Anonymous Coward]

but then have to sftp the entire directory since it doesn't support mirror,

Eh? rsync over ssh?

By the way, lftp does support sftp. "man lftp" some day.

Re:X in Windows?

[Anonymous Coward]

lftp supports sftp. it is great

Nice Screeny's

[daxomatic]

Yummie soon available near you http://freedesktop.org/XOrg/X11R68ScreenShots

Re:Nice Screeny's

[Tyreth]

These look particularly good:
http://www.gnome-look.org/content/preview.p hp?prev iew=1&id=14958&file1=14958-1.jpg&file2=&file3=&nam e=E-Gnome+Dropshadows&PHPSESSID=6bc847f92378423559 c4dcafca392d23
http://www.lynucs.org/index.php?sc reen_type=1&scre en_id=1759409500411796a9ba106&m=screen

Anyone know what the MacOSX like theme(s) they are using are? The drop shadows give a fantastic touch.

Re:Nice Screeny's

[Anonymous Coward]

The drop shadows are from the compositing manager run with the xserver (I've had them for months with the fd.o xserver). It looks like they are using enlightenment with the winter theme.

Re:Nice Screeny's

[Anonymous Coward]

Yeah, and you spelling "Idiot" as "Idoit" is a mark of sheer genius.

Re:Nice Screeny's

[Anonytroll]

Yes, I know it. First of all, on those screens, they are running Enlightenment with the Winter theme. The GTK2-theme is called "Milk" and indeed a port from an OS-X theme.
That thing at the bottom is called "Engage" and part of the E17 applications that are available (even though E17 as a whole is not).

Re:Nice Screeny's

[Anonymous Coward]

thx 4 troll thx 4 lots of things thx

KDE and Knome infect X ?

[vi (editor)]

I fear that in the long term windows manager features will included into the X server. The cooperation between X.ORG and the KDE / Gnome teams doesn't bode well.
Such an integration would destroy the versatility and uniqueness of the X protocoll. Indeed X would degenerate to a remote enabled clone of the Windows desktop after some time.
Yes, want Linux/BSD on the desktop but not this way.
This is like getting an elephant into your car by cutting him into pieces.

Re:KDE and Knome infect X ?

[gr8_phk]

Agreed. Under the list of supported apps and stuff they list a library of clipart. WTF does that have to do with X?

X needs to remain low level to stay relevant.

Re:KDE and Knome infect X ?

[JohnFluxx]

So.. you don't have any evidence for this happening, or any technical facts at all, but you have a fear for it happening?

Re:KDE and Knome infect X ?

[evvk]

I must second you here. I fear that, say, ten years from now there will be no easy way to switch to window managers like Ion, ratpoison, larswm, the newer clones of these, and whatever new innovations might happen during that times. WIMP policies will have been so deeply integrated into the basic windowing system. X (which is just a graphical input/output protocol!) and the ICCCM are excellent in that they don't dictate policy too much and thus allow for this kind of experiments and research without the system having to be rewritten from the ground up. Research into new interaction techniques must not be forgotten and WIMP considered the final evolutionary step of GUIs. (Infact, it was just the first step!)

Re:KDE and Knome infect X ?

[Anonymous Coward]

I actually disagree. Go figure huh?

I think that X should develop standard application interfaces that all desktop communicate with. No I am not talking about xterm pos x,y but events interfaces - what should occur on mouse click, on drag etc. The low layer needs to interface with the os as X currently does but it needs an abstract layer that can change over time without affecting the backend and visa-versa. I have not program much at all in X so I may be off but I know that any all desktops handle events differently. If the X would setup API to describe events then the desktop can make the nice pretty interfaces that I really don't care about but other people really like. What I am looking for is a standardized interface.

Re:KDE and Knome infect X ?

[mark_lybarger]

x talks to your video hardware, hopefully through kernel drivers.

kde and knome are layers on top of x. i'm sure you realize that.

x tells the video card to draw stuff on the screen.

kde and knome tell x what that stuff should look like. kde and knome wrap windows with decorations. (title bars, etc).

if a user wants to be able to resize their desktop on the fly (go from 1024x768 to 800x600 resolution perhaps), the functionality has to be available in X, then also able to be controled by the window manager.

likewise, if you want window drop shadows, X has to be able to support that and you can just enable it in your window manager of choice.

this is not at all about slicing an elephant up to stuff into an automobile (where did that come from?) it's about providing kool features that everyone can enjoy and not have to duplicate all over the place. it's about letting the xerver talk to the harware capabilities that are in most semi-modern hardware these days.

finally, stop fearing the advancements of the X server. they're long over due and there's a lot of folks out here that welcome the advancements with open arms!

Re:KDE and Knome infect X ?

[Anonymous Coward]

Insightful?????

Do you have any example of what you are talking about or were you just trying to sound like you knew what you were talking about?

And don't confuse the xorg X11 implementation with freedesktop.org in general.

Re:KDE and Knome infect X ?

[drinkypoo]

A library of clip art is not harmful as long as it is in a separate file that you don't need. As it stands X.org is still using the packaging model of XFree, it's in several archives and you don't need them all to build X from what I can tell (although gentoo does it for me so I'm not too worried about it.)

Re:KDE and Knome infect X ?

[Pecisk]

Hmmmm, IF you are talking about freedesktop.org, that it is NOT Xorg. Freedesktop.org is collabration place for various projects to interact and make integration easer. I DON'T think anyone will mess with X protocol, so your worries are little bit over the top, IMHO.

Re:KDE and Knome infect X ?

[canavan]

I fear that in the long term windows manager features will included into the X server.

That's something I'd actually like to happen - the way it was in the NEWS system (just better). In my opinion, it would actually be beneficial to replace the window manager with a script running inside the server, or even to allow application to upload scripts to the server that handle their Menus so that there are no unnecessary delays going back and forth between the X-Server and a remote client for stuff as trivial as moving a window or opening a menu.

Re:KDE and Knome infect X ?

[Telex4]

You're confusing X.org and freedesktop.org

X.org is a windowing system.

freedesktop.org is an umbrella project bringing together desktop-related Free Software projects.

So X.org is but one of fd.o's many projects, including that clipart library. That certainly isn't going into X.org ;-)

The grandparent of this post is also either making things up or very misinformed. X.org have no plans to integrate things like window managers into X. X.org will remain an implementation of the X protocol, and will in fact become increasingly flexible for GUI developers, whether they want a simple window manager or a full desktop environment.

Check you facts before you agree with a Slashdot post ;-)

Re:KDE and Knome infect X ?

[Anonymous Coward]

Well, I would not fear that as much as that the window managers themselves become dependent on the explicit integration with upcoming X extensions; as long as they are, after all, extensions, one can just disable them.

Re:KDE and Knome infect X ?

[Brandybuck]

Don't listen to the propaganda coming from Freedesktop.org then. he politics promulgated by the fd.o leadership are regrettable, but fortunately no one much listens to Daniel and Havoc except Daniel and Havoc. The software that fd.o hosts is completely independent of fd.o. In this regard it's not much different from freshmeat. The rest of fd.o still serves as a useful clearinghouse for desktop interoperability standards. While Havoc has a lot of clout within Redhat and GNOME, he has no authority within any other destkop's developer community.

Compositing

[maharg]

I know I'm going to get flamed by all the 80x24 textmoders out there, but compositing is cool

Re:Compositing

[gatzke]

It is eye candy, and it is cool, but I still have trouble using it for real work.

Maybe for some applications (drawing?) it might be ok, but reading web pages and writing code, transparency makes my eyes hurt...

I remember the first time I saw a neat looking transparent eterm years back. It was very hard to use for any length of time at all, but it is neat to show people.

Back to my glowing monochromatic lime green vt100 serial terminal for another round of nethack. "Welcome Luddite the evoker!"

Re:Compositing

[Smallpond]

Misadvised developments like this are the reason that you need 3GHz processors to run applications that should run on a 486. Please name an application in which compositing gives a better user interface than tabs or just overlapping windows. Compositing makes it difficult to select elements or identify the source for screen objects. The supposed advantage (you can put more stuff on the screen at the same time) seems more like a disadvantage in most cases.

Re:Compositing

[GoRK]

Compositing is not all about transparency you know. Being able to do things like offload drawing onto the powerful graphics cpu's present in most new computers these days is the most obvious beneft. There is also a nice benefit to the multi-desktop pagers and desktop preview widgets out there, since it's a lot less work to capture and resize image windows. For users without a lot of screen real-estate, switching techniques such as Apple's Expose can in many cases give a user a real benefit. Most of these things would be slow or impossible to do without good compositing. Every time I drag my kterm across a thunderbird window, I feel the pain of X (Nvidia's drivers do not help much either)

Re:Compositing

[plimsoll]

I run 40x25, you insensi-
tive clod!

Re:Compositing

[Anonymous Coward]

Misadvised developments like this are the reason that you need 3GHz processors to run applications that should run on a 486. Please name an application in which compositing gives a better user interface than tabs or just overlapping windows. Compositing makes it difficult to select elements or identify the source for screen objects. The supposed advantage (you can put more stuff on the screen at the same time) seems more like a disadvantage in most cases.

Hey, asshole. I'd like to see you run a browser that has tabs on a 486.

In fact I've run recent versions of X on overclocked 486 machines (one of those weird pentium accelerator doo-dads) and it ran fine, but slow.

Look obviously you don't know what compositing is.

The transpariances are just for shits and giggles. To say "see it works" sort of things. Eye candy.

But it can be applied to lots of different technologies and technics to improve the user interface.

Another thing that it does, right now, is remove any of the tearing and akwardness that X has. No more "trails" as you drag windows from over one window to another.

This isn't screwing it up or dragging the performance down, far from it, if your computer doesn't handle it it is purely optional. In fact you have to turn it on to use it like this:

section "Extensions"
Option "Composite" "Enable"
endsection

Actually what it is doing is going to enable developers to use the hardware more effectively. To use the proccessing power to create a more pleasent, easier to use, and more unified place. It's about enjoyment of the computer.

This composite feature enables developers to get creative about things. Lots of new ideas are going to suck, but eventually some very good ones will help out.

Hell. Your probably one of those guys that said "Who the fucks needs Opera and it's tabbed browsing, I have IE 3.0 and Windows 95 and that runs just fine on my 486!"

Re:Compositing

[gatzke]

True, transparency, dragging windows cleanly, and pager preview are nice. They are still eye-candy, but it does help with the wow-factor.

We have huge GPU's, so why not use them. Good point.

It is amazing the strides linux has made over the last few years. As you knock out more and more of the limitations, people can't say "I can't swich because linux doesn't do anti-aliased fonts or transparent windows." They may bitch on usability, but that is another problem...

Re:Compositing

[BenjyD]

Things like translucent error dialogs (see Jef Raskin's book), Expose, any oddly shaped app that you want to see the background through the gaps in (like the Dock).

It isn't just the ability to draw translucent windows. By rendering to an offscreen buffer and then compositing on the screen, you can do all sorts of transformations and effects that make the desktop easier to use. Apple's Expose effect is probably the best example.

Re:Compositing

[Anonymous Coward]

Misadvised developments like this are the reason that you need 3GHz processors to run applications that should run on a 486.

Rubbish. If your CPU/GPU isn't fast enough to handle it, don't switch it on in the first place.

Please name an application in which compositing gives a better user interface than tabs or just overlapping windows.

The composite extension isn't just used for overlapping translucent images. Things like drop shadows may look pretty, but they give a better impression of depth, highlighting the edge of windows and thus increasing usability.

Re:Compositing

[jsebrech]

Laugh all you want, but at one point I would occasionally check my email by dialing with my modem-equipped palmpilot into the internet, running a telnet session to my mailserver, and running pine in a shell session.

Amazingly, it worked just fine, and was perfectly usable.

On the other hand, VNC from that palmpilot, though I tried it a few times, was just not usable. So I will admit that every once in a while those who claim the shell has better usability do have a point.

Re:Compositing

[Patoski]

Please name an application in which compositing gives a better user interface than tabs or just overlapping windows. Compositing makes it difficult to select elements or identify the source for screen objects. The supposed advantage (you can put more stuff on the screen at the same time) seems more like a disadvantage in most cases.

Any graphics editor which utilizes layers, print layout editors (layer blending), sound / movie editors (merging tracks), music sequencing (merging tracks) and to some extent word processors / HTML editors (think "invisible tags"). Being able to offload compositing onto the GPU in these would be a great help to these types of CPU hungry applications.

Re:Compositing

[CaptnMArk]

>Being able to do things like offload drawing onto the powerful graphics cpu's present in most new computers these days is the most obvious beneft.

This is true, but it will make X require a modern gfx card.

And the only really supported (with adequately fast open source drivers) one is Radeon 8500-9200.

Re:Compositing

[GoRK]

This is true, but it will make X require a modern gfx card.

How do you figure? There are software fallbacks on every windowing system out there that does hardware-accelerated compositing - even the X work. Before you hit reply there to say "But it's slower!" remember a couple things: Modern 3d graphics cards are getting slower and slower software framebuffers (ie copying raw pixel data to the screen) as they put more focus on the hardware acceleration stuff (ie copying around texture data via DMA or somesuch) -- you will take a speed hit if you DONT implement more and more hardware assisted techniques. Next, you really have to define what you mean by a "modern graphics card" because there are almost no devices out there that require X to simply draw to a raw framebuffer except for maybe some embedded platforms. Even framebuffer-only devices implement minimal hardware acceleration such as page-flipping. We can hope that any software-only fallback is going to be as efficient as possible, so it's naive to assume that writing software routines to emulate the hardware compositing will be any slower than the software routines that are 'faking it' now via various techniques. In all likelyhood, it would actually be faster, or if not faster, it would require less of a memory footprint.

Re:Compositing

[StormReaver]

"Please name an application in which compositing gives a better user interface than tabs or just overlapping windows."

That's easy. I created a visual database schema designer which has a table search function (columns too, but no need to go into details) that brings up a non-modal window with all the possible matches in a tabular format. Clicking on one of the items in the list causes the corresponding table to center on screen and flash a few times. If the table centers behind the results window, then the table is hidden and the flashing can't be seen. A translucent results window would solve that problem better than anything else.

Re:Compositing

[plimsoll]

| Laugh all you want [...]

Who's laughing? The /etc/lilo.conf on a
particular host of mine passes the 40x25
kernel parameter "vga=0x0100".

`top` looks real funny at that reso-
lution, and `ls -l` is useless, but
`mplayer -vo aa [...]` at 40x25 has a
coarseness that is almost heraldic in
its low-tech beauty.

Sometimes I have to go back to 80, but
not if I can help it. It makes me agora-
phobic.

Re:Compositing

[ninjadroid]

I would sure like to have some real transparency for the fanslider widget me and a friend created. Screenshot here.

This is a variable-precision widget, geared toward use in audio applications. Click the slider, then drag out to create a "fan" indicating your increased precision. However, without transparency, I can only do this with clip masks (i.e. specify the window's shape). With a vertical fanslider, this is no problem when dragging to the right; I only have to clip the window once, then I can adjust it's width to create a properly sized fan. But going the other way, things start to suck. The implementation reasons are a bit boring, but basically, I can't just "set it and forget it." The result is that the fan is choppy as shit going to the left.

Real Transparency would make this all go away. If we stipulate that this is a useful and productivity enhancing widget, then there is in fact at least one scenario where compositing is a definite functional plus.

My $0.02.

Re:Compositing

[Bert64]

Well VNC is slow as hell.. who would want to use VNC anyway? X11 supports remote displays NATIVELY and is much faster.. Ofcourse it's still slower than a text based console, but what would you expect?

What about Y?

[Outsider_99]

With all this talk of X, ive remembered Y-Windows http://www.y-windows.org/ Does anybody know whats happened to Y? According to the road map, version 0.3 should have beed out 4 months ago.

Re:What about Y?

[jandrese]

Yeah, people realized that they liked all of their old applications and the project never really took off.

Re:What about Y?

[saxa]

So what's on with Y ?? Txs

Re:What about Y?

[Anonymous Coward]

Exactly.

X Windows has backward compatability. Even with the new extensions and stuff old apps will still work fine.

Even if they move to pure OpenGL and you have minimalist hardware it would still be possible to have only 2d drivers.

Just chalk it all up to the wisdom of the original designers of the X windows protocol. The multilayered technics realy pay off. (like in TCP/IP) A lot of the bad things they did, they knew were bad. They just figured they fix it when they got around to it, but they never did.

Like the old blit-blat fonts for instance.

Now you can still run apps with old monocrap sized fonts on a modern X Windows, but I don't have to put up with that crap with my browser. If I want a nice new true type font, I just copy it's .ttf file into my ~/.font directory and I can use it in all it's glory.

Very neat stuff.

Now if X was a monolythic monster, or backward compatability had to be broken to improve it, we Unix-like OS users will have nothing to look forward to except for a complete rewrite. With out X's flexibility there would be no point in keeping it. But right now it is very usefull and hopefully will be for a long long time.

I mean we won't want to end up like Windows, right?

The one gigantic monolythic complex peice of software that is WinXP is teh suck, and they are completely redo'ing it from scratch for Longhorn.

Hell they will still have to keep Win32 stuff in Longhorn ON TOP OF all the extra crap they are piling on, JUST BECAUSE ALL THEIR APPS RUN IT.

Think of it like classic mode for OS X and OS 9. Except now imagine every single commercial app in existance now has to work in classic mode.

Hell MS can't even get a SERVICE PACK out on time. And when it did come out it was a complete peice of crap in terms of security. 2 serious flaws have already been found, and that's not counting the 2 little security bugs that were found within a day.

They DON'T EVEN HAVE A 64bit PORT OF WINXP YET!!!

And x86-64 is like 90% compatable with x86!!

And yet they want to make a new OS completely from scratch?

They must be completely nuts. Loony.

(Of course if Longhorn is out in 2006 I'll taking out the other side of my anonymous face. But I don't have to worry about that.)

Thank goodness we don't have to face crap like that, thanks to our trusty old X Windows.

Open code means a open future.

Re:What about Y?

[Anonymous Coward]

version 0.3 should have beed out 4 months ago.

Yeah, they should have beed out, but then they got hooked on this ilovebees bullshit, and now it will be another six months before they're all beed out again.

NVidia driver issue?

[Lispy]

He avoids answering the question about XFree86 Driver compatibility, especially with regard to NVidias binary drivers. This is a big issue for a lot of users and I hope that NVidias cards will be capable of using the new extensions.

As far as the article goes all I can read is that they work with both major graphiccards vendors but only ATI delivered so far. Or did I miss something?

Re:NVidia driver issue?

[Anonymous Coward]

Uhh, compatibility with nvidia and ati binaries is one of their release requirements, f00

Re:NVidia driver issue?

[Anonymous Coward]

I don't know if you missed something in the interview but the nvida drivers definitely work with the new xorg and the extension. (I'm using them right now).

Also if you take a look at the xorg mailinglist you'll find that the guys at nvidia are working happily with the xorg devs.

Re:NVidia driver issue?

[McLoud]

To what I've read in their mailling list, they can use the binary nvidia driver right now. There's some issues, but not directly related to the nvidia driver.

Re:NVidia driver issue?

[Lispy]

Ok, thanks for all your replies. That makes me rest assured. I was pretty sure they wouldnt release it without NVidia, but the interview sounded suspicious so I thought Id ask... ;-)

Re:NVidia driver issue?

[Flammon]

I've got an ATI Radeon 9000 PRO and using I'm using the ATI binary drivers. It crashes my system all the time when I start delving into GL apps.

Check the ATI Petition for Adequate Drivers in Linux to see how happy the current ATI users are with ATI. Last time I checked, there were 9919 Total Signatures which have been added in the last week.

Re:NVidia driver issue?

[Anonymous Coward]

Same here (with the mobile version of the 9000), trying to start any GL aplicattion hard-locks the system (not even sysrq works).

As per usual

[frankthechicken]

From the interview:-

For the less code-inclined, there's always lots of documentation to be written! Manpages need to be written, documentation needs to be released Xorg 6.7. converted from random archaic formats to DocBook, et al. This is one area that really badly needs some love from those with the requisite skills.

I realy wish that this was a higher priority among developers, as it would greatly help both new users, and future developers.

Don't bother with the next cool widget until the docs are up and understandable.

Re:As per usual

[EsbenMoseHansen]

In my experience, well-written code beats documentation every time. Documentation is always out of date, or just slightly wrong.

On the other hand, user manuals and such are very important. Linux generally shines in this area (compared to windows, at least), but still, the more the better. Writing such is, however, not a skill every developer has. Try tla help if you doubt me :) And besides, lots and lots of users could conceivable write these, and thus pay back some of what they have gained for free :-D

Re:As per usual

[Anonymous Coward]

I realy wish that this was a higher priority among developers, as it would greatly help both new users, and future developers.

I really wish those users who keep asking questions and keep asking to create manuals took a little bit of their time and wrote some of the stuff they were thought by the developers. Is it so wrong that a developer develops and asks you guys to write down those things he explains to you?

Don't bother with the next cool widget until the docs are up and understandable.

Ohh man, don't bother ranting on slashdot but help with what you learned from the developers!

Re:As per usual

[Anonymous Coward]

The type of documentation you're talking about is a waste of effort. By definition, a program's code is a perfect description of the program as it exists. How can it be anything different when the program is actually generated (compiled) from the "documentation" that describes it?

There's no need for superfluous documentation as all the information found in a man page (and more) can be found in the source code. "Need help? RTFSC!"

Unfortunate...

[Sheetrock]

I was somewhat hopeful that the impending death of X-Windows would lead to the development of a windowing system designed specifically to take advantage of the more advanced features of NetBSD.

Indeed, most of the X-Windows targets would benefit from a native implementation of a windowing system -- native implementations could run something like 23-27% more efficiently because of the layers of abstraction that are currently necessary. Most people don't need internationalization and most could use a simpler interface to get the printer working.

I think there's an underlying fear to reimplement that comes from (and I hate to say it) a certain sense of elitism in juggling three or four fontservers or digging through a million XFConfig-4 lines to get TV-Out working. But perhaps the focus needs to be put back on the basics?

Re:Unfortunate...

[Anonytroll]

Most people don't need internationalization

I beg to differ. There's a world outside of where you live. In that world, internationalization is an issue. Or would you like to work with a system that displays everything in (for random example) French, because internationalization was "not an issue" for the developers?

Re:Unfortunate...

[Anonymous Coward]

I was thinking of English. Most computer literate users speak it well enough, and if not by all means pick up the internationalization pack.

Re:Unfortunate...

[gl4ss]

**I was thinking of English. Most computer literate users speak it well enough, and if not by all means pick up the internationalization pack**

but that is not what internationalisation is all about, I for example use my computers in english, yet I write and read Finnish on them every day.
äöäöäöäöäöäöäöääöäÅÅÅÅÅÅÅÅ&#19 7;

Re:Unfortunate...

[cortana]

But without internationalisation, software developed in, say, French, would present a French user interface to you!

i18n means that gettext, or whatever, simply pulls out the en_US strings and the user is none the wiser.

Re:Unfortunate...

[dago]

"Most people don't need internationalization "

Are you really sure about that ?

Re:Unfortunate...

[zerblat]

Maybe, maybe not. As computer usage becomes more common, the number of users who don't know English well enough increases. Besides, even though non-English users might not have any problem with English user interfaces, it sucks pretty bad if you can't e.g. write/display documents in your own language.

I'm sure you would love writing emails to your friends in English, but using the Cyrillic alphabet...

Re:Unfortunate...

[cortana]

Not to mention the fact that there is more to the process than translating an app's user interface.

* the 127 symbols of ASCII are, surprisingly enough, not sufficient to display non-english languages (and even with ASCII I can't type my country's courrency symbol (£))
* text is not always written left-to-right; many languages also have extremley complicated rules for compositing glyphs; for example, I believe Arabic characters have all sorts of weird rules about whether (and where) a horizontal line is drawn.

Run kcontrol --reverse some time for a trippy but pointed example of what this is like, BTW :D

Re:Unfortunate...

[kahei]

Most people don't need internationalization

I despair. I do not pass GO, I do not collect $200, I just despair.

Re:Unfortunate...

[Anonymous Coward]

GREAT troll.

no seriousely, this is the best one i have seen on slashdot.

and NO ONE managed to catch the funny part.

so i will beat the joke to death by explaining it:

He said the abstraction layers are a problem. WHAT IS NETBSD based on. a layer that is independantly implemented per architecture. an abstraction layer for the different types of chips it will run on.

beautiful

(either that or the guy is just dumb, i hope he was doing it intentionally cause it was far funnier)

Re:Unfortunate...

[Anonymous Coward]

I think with a statement like internationalization is not needed, it almost has to be a troll. But, the responses s/he illicited with said troll are great. Posts like this reaffirm the notion that trolling can in fact be an art form.

Re:Unfortunate...

[Anonymous Coward]

uhm...
USA less than 300 Million people.
Europe (except England as they speak english), more than 400 Million

India more than 800 Million
Kina more than 2 billion

I think you need to think again about your "most people". Or do you mean, "most people in USA"?

(all numbers are from my memory and might be somewhat wrong)

Re:Unfortunate...

[Anonymous Coward]

Or do you mean, "most people in USA"?

You mean there is a world outside the states???

Re:Unfortunate...

[Felonious Ham]

How about this: _I_ don't need internationalization. I think maximum flexibility and accessibility are fantastic, necessary. However, if it were possible to get a faster version of whatever that didn't support wacky characters, or crazy formatting, I would be all over it. I'm never going to use the Fahrvegnugen translation, so why is it necessary that my desktop instance support it? Which brings to mind another nit I have with Gnome: the combo on the file browser that has character encoding options. Why would I, arrogant English speaker, ever need to change the encoding of file names in the file chooser? To put it another way, I've never, in my 20 years of using a computer, used this "feature". It doesn't eat up gobs of memory, or chew on the cpu, but it's wrong to put it on the GUI!

That's my 2p.

Re:Unfortunate...

[Anonymous Coward]

Completely offtopic:
Hey.. wtf? Umlauted characters work on Slashdot again, now? 'Bout time!

Hyvää!

Re:Unfortunate...

[O2dude]

actually...

Europe 400 Million people
India 800 Million people
Kina 2 billion people

USA 1 Million people...
and 299 Million neanderthals with credit cards.

Thats X.ORG

[MikeDX]

rather than X.COM which is of course paypal..

unified desktop

[tobi-wan-kenobi]

this looks like it can really challenge m$'s monopoly on desktop computers one day. looking at their project-roadmap, it sure looks like that's what they are going for (mono, wine, open office, samba, ...). i really hope that they succeed in creating a tightly-knid net between linux desktop developers. in any case this is a valuable resource for getting information about various graphical environments (though a bit confusing for newbies, i wagger). hopefully they rise high above their humble goal to be a "collaboration zone".

Re:unified desktop

[dave420]

They have to make it more coherent first. It's come a hell of a long way, but it's not there yet. OSX, windows, etc. all have tight GUIs, where graphical elements are actually designed by designers. I know it sounds trolly, but please listen. It's tiny things like that which show up immediately to a user from another operating system. Sure, there's no "right" or "wrong" when it comes to such things, but they're off-putting if you're not used to them. I've got nothing against linux, but when I see a linux desktop, even if it's got the latest Aqua-esque theme, there are some graphical elements (column headings, window buttons/borders) that are maybe one or two pixels off, which stands out a mile.

If they could be fixed, giving a more clean-looking GUI, linux would make much more headway into the desktop market.

Mod this as you will. I can smell the flames already.

Re:unified desktop

[cortana]

No, you are correct.

Most users new to Linux will be using Gnome and KDE. Most of the programs that they will run into (those which don't do everything themselves, poorly, such as games) are written for GTK+ and Qt. A user's first impression of either desktop environment would be improved tremendously if the default themes for each environemnt didn't look like complete ass.

Fortunatly, the default theme for Gnome 2.8 will be Indubstrial, which is based off the very smooth Industrial theme.

Likewise, the next version of KDE (perhaps 3.3 already?) uses Plastik instead of the godawful Keramic theme.

Now all we need is for freedesktop.org to finish their cunning gtk engine that uses qt to draw everything--thereby unifying the look (if not the functionalty, behaviour, feel) of both desktops.

Although personally I want the opposite effect: to make the few qt/KDE apps I run look like they use Industrial, not switch my entire desktop over to Plastik. :)

Re:unified desktop

[dave420]

I'm not too au fait with linux desktops (I only ssh into my linux boxes at work, and I don't use it on the desktop partially due to the reasons I've described already). I looked at a couple of screenshots from this new release, and I saw some of the ol' familiar one-pixel-off features I've been talking about. You can see in that screenshot, the right-hand window border by the scrollbar on Thunderbird is not straight, and the text on the column headings isn't vertically aligned properly. Are those due to the theme, or the technology behind it?

If linux had a tight GUI, I'd give it more thought. I'm really not trying to be elitist or trollerific here, but it's so distracting when these things crop up. (I'm also a web developer, so when things aren't pixel-perfect, my skin crawls)

Re:unified desktop

[Anonymous Coward]

Likewise, the next version of KDE (perhaps 3.3 already?) uses Plastik instead of the godawful Keramic theme.

I'm running KDE 3.3. Sadly, it defaults to Keramic. Switched it to Plastik, ran the qt theme engine for my gtk apps so they use Plastik, and use the Plastikfox theme in firefox ... my desktop looks more consistent than the average windows desktop now. Feel is slightly different, but most end users heads don't explode even when they run something as alien as Lotus Notes, let alone Windows Media Player, so I don't think consistency is *that* important.

Re:unified desktop

[Anonymous Coward]

I'm also a web developer, so when things aren't pixel-perfect, my skin crawls

In that case, I hope you never look at your websites in more than one browser - because I have yet to find a single site which is pixel-for-pixel identical in both FireFox and IE.

A web developer who demands things be pixel-perfect is about as sensible as a taxi driver who only accepts passengers when their fare is going to come out as an exact number of dollars.

Re:unified desktop

[dave420]

It's EASY to get things pixel-perfect on the web. Of course they're not going to be exactly the same on each system, but you can get cut-up graphics to adhere to each other easily across any platform you can think of (which is the entire idea of a good HTML coder and HTML code). I've been doing it for YEARS. sheesh :)

Re:unified desktop

[cortana]

This is because Thunderbird draws its own column headings; if you zoom in on the offending area, it looks much more obvious.

I think we'll be stuck with Mozilla projects doing their own graphical things for quite some time. :)

I don't actually know what theme that desktop is, I would guess Windowmaker or some derivative, spiced up to use the new features of this Xorg release.

If you are looking for a pleasant, well-integrated environment, you should really try Gnome 2.6, with the Industrial theme. The default GTK+ theme isn't too hot; fortunatly Indubstrial (based of Industrial) will be the default theme of Gnome 2.8.

Re:unified desktop

[Technonotice_Dom]

Here's my take on it:

I use Linux on my systems (with Gnome as a DE) and I know what you mean some times. Part of the problem is that people write programs for different purposes - e.g. some people will write a program using Qt and the KDE libraries causing the programs to look one way, while somebody else will use Gtk or another toolkit.

It obviously isn't just down to the toolkit, but also depending on who the application is targetted at, most developers (generalising I know) don't have the time (or don't want to, or don't have lots of experience) to make their application pixel perfect.

Gnome has some usability guidelines and I think anybody would testify to the fact that Gnome itself and applications based around the HIGs have a very consistent feel. Likewise KDE has some HIGs (currently redrafting I think) but it doesn't have anywhere near the emphasis on the programs in the KDE collection IMO.

As well as defining the HIGs, part of the problem is to educate interface programmers and try to ask them to follow the guidelines - and more importantly, for people who have experience in usability (and that includes all users) to comment, suggest changes etc.

An interesting example was on the KDE Usability mailing list the other day - Celeste Paul posted a usability report on KHangman. The coder behind it appeared shortly after and immediately began following the conclusions of the report. I'm sure almost every programmer will be happy to accept constructive criticism for their work.

If you think a menu item or something isn't right, file a bug report against it - and try to include a suggestion (even if it isn't a complete solution) for how it could be improved. It only takes a few minutes. </rant>

Re:unified desktop

[dave420]

You see, that's my beef. On Macs and Windows, the apps generally don't draw their own graphics. Usually, the only time they'll do that is if the graphical devices they want (specialised window shape, specialised style) aren't available in the OS. If an app wants a column header, it'll draw the standard OS one. I think that's the attitude linux needs to take, otherwise it'll always look a little "dodgy" from an aesthetical point of view.

Re:unified desktop

[pyros]

Thunderbird doesn't use the system themes, nor does Firefox or the Mozilla suite, they aren't GTK apps. Evolution is the mail client that should have an integrated look/feel. Epiphany is the browser. Galeon is another browser that should have the consistent system look/feel, but it isn't bundled like epiphany.

Re:unified desktop

[be-fan]

If Windows has such a "tight" and "well-designed" GUI, then please, tell me:

1) Why does Luna look like a pre-schooler threw up after eating several crayons?
2) Why do MS Office, MS Visio, and MS Visual Studio all look different (hint: they use different toolkits!)
3) Why does every other Windows apps (Winamp, Windows Media Player, Ephpod, etc, etc) use their own weird-looking skin?
4) Why do the buttons on every single installer (Wise, InstallShield, MSI) all look different?

Re:unified desktop

[be-fan]

That's where you are wrong. Windows apps *do* draw their own graphics. That's why Office doesn't have the same scrollbars as Luna apps, and why Visio XP has those god-awful Keramik-style blue-gradient toolbars.

Re:unified desktop

[cortana]

There are sooooo many apps on Windows that draw their own controls, often because the programmers just have no clue. :)

In this case it's Thunderbird's fault--GTK provies a perfectly nice listbox control, but I'm sure they have a good reason to use their own for the time being. Mozilla and co used to be much worse, Try comparing an early build of Firebird to the latest Firefox and see how many controls have been changed to by drawn with GTK.

The biggest problem in this area is Openoffice. It does everything itself--probably because Gtk+ and Qt weren't mature enough to consider using when the project was started, and in an uncommon stroke of common sense somebody at Sun didn't make them use CDE (yuck).

I read some time ago that OpenOffice 2 will have the user interface layer completly separated from the rest of the code, so that it would be able to use both GTK+ and Qt; but unfortunatly I can't find any more information about this process, leading me to believe that it has been abandoned until a later verson. :(

Re:unified desktop

[LordK2002]

I agree entirely. The most annoying thing in my day-to-day Linux use is that the column headings in GTK list boxes are a few pixels too tall, which looks clunky and wastes space.

K

Re:unified desktop

[dave420]

If you read my post, that's exactly what I'm saying. Bear in mind that Microsoft has a huge art and design department, whereas Mozilla doesn't. I'm talking about design integrity, not politics. The MS apps that use their own graphics also give those graphics back to the OS. You can write your own software to use their toolbars/whatever, which is cool.

Re:unified desktop

[dave420]

I use windows very frequently, and yes, lots of people do draw their own controls. They do it because windows doesn't offer the graphical device they want. What I see on Linux is software developers using random libraries to draw their graphics that exist in other libraries, instead of drawing from a common pool (no pun intended). Every time I've seen a windows app drawing its own controls, I get the same feeling I get when I see your average linux desktop. Disjointed graphics, floating ill-aligned controls.

I guess if linux had a universal, well-defined, mature and skinning-friendly graphical toolkit lying around, most of these "what the hell is that?" problems would disappear.

It's ironic that MS also uses custom graphics in their office suites, usually because they want a more stylised look to their apps. Their usually work out OK as MS has a pretty decent design department. When some of the smaller linux developers try to do theirs, it doesn't always come out that good.

am I running round in circles here, or is it just me? :-P

Re:unified desktop

[be-fan]

No you can't. You have no idea what you're talking about. Mozilla using it's own toolkit, and thus looking different from the rest of GNOME (which uses GTK+) is *exactly* the same thing as Office using it's own toolkit, and thus looking different from the rest of the Luna desktop. In both cases, attempts are made to make things look similar, but there are slight and even egregious differences (Office doesn't use the Luna scrollbars! Arrows look slightly different. Widgets are slightly flatter).

Re:unified desktop

[LizardKing]

Thunderbird doesn't use the system themes, nor does Firefox or the Mozilla suite, they aren't GTK apps.

That's strange. I wonder why Firefox and Thunderbird depend on GTK+ on my laptop. Must be a packaging error. It's also strange that when I installed the GTKstep theme Firefox and Thunderbird started using NeXTstep like widgets as well as my GTK+ applications.

Odd that. Unless you're talking out of your arse.

Re:unified desktop

[chez69]

have You used a recent build of firefox? it looks just like any of my other gnome apps and even uses the new gnome file selector

Re:unified desktop

[Anonymous Coward]

am I running round in circles here, or is it just me? :-P

Yes, and yes.

Re:unified desktop

[pyros]

I was always under the impression that they were Chrome apps. It could just be Mozilla. Anyhow, I'm mostly referring to how Firefoz/Thunderbird/Mozilla have their theme engines, and their own themes. Red Hat has always done some extra packaging so that the default theme picks up GTK colors and icons. But if you use any other theme.

Re:unified desktop

[JimMcCusker]

I'm currently using the gtk-qt engine, with Plastik. Everything looks rather nice (I kinda like Plastik, myself), and Firefox, Thunderbird, and Gimp all look great with their new clothing.

Re:unified desktop

[spitzak]

You are wrong. Many, many Windows applications use their own toolkits. My own work (fltk) which was designed for Linux, is obviously very popular for Windows.

The main reasons are: to get a better toolkit API, to reuse code they have to write for the app anyway, to get portability to non-Windows, to get exact control over the appearance, to avoid screwups if Microsoft changes the size or appearance of their widgets, and many other reasons.

One thing you can say about Windows was that befor the "Luna" XP theme came up, it was pretty easy to get exact pixel-accurate copies of their controls and other toolkits copied them quite closely. This may lead you to think everything is using the same toolkit when in fact it is not. In fact most of the consistency on Linux is due to toolkits being made to match Windows98 so the Windows port matched.

I agree with a lot of what the various posters are saying here, just disagree with the "unify the toolkit" solution. If X had a "unified toolkit" it would have been Athena and we would still be using the middle mouse button to move the sliders, and it would look like crap, and "configuration" would consist of being able to swap black and white. X has proven that an almost 20-year old interface, because it was designed as a low-level, is able to implement things that were not even dreamed of then. We need to keep these ideas.

It would be nice if an api was provided with something like "draw a raised button with this UTF-8 string as a label" and that "themes" could modify this. More importantly than "consistency", this would avoid people writing their own with subtle bugs where the positining is off, like you complained about.

Re:unified desktop

[spitzak]

The "non-straight scrollbar" appears to be caused by box edges being drawn above it. The scrollbar has a very thin box edge. Above it is a column header for the browser and it has a very thick edge (the fact that it draws an edge here at all appears to be a mistake?). The top area seems to have a bit of a fuzzy edge. Your eye tends to see the window edge as going down the center of these dark edges, rather than the right outside, and sees a crooked line.

In a lot of ways these bugs are actually caused by the use of toolkits. If Thunderbird drew everything in a single big block of code, it would probably look a lot more consistent.

I also agree that the text in the column titles is 1 pixel too low. But if you look at the selected line it appears 1 pixel too low there as well. Thus it is likely that the fix is to change whatever code they are using to center text vertically. I have run into this a lot, it requires careful tweaking. Do not use n/2, use n>>1 so that it rounds in the same direction even if n is negative! Then TEST at both even and odd font sizes and box sizes!

The first problem (misaligned boxes) I have certainly seen plenty of times on Windows. I agree that the miscentering of fonts is less common on Windows (it actually provides a GDI32 call to center text in a box).

Re:unified desktop

[Remillard]

1) Why does Luna look like a pre-schooler threw up after eating several crayons?

2) Why do MS Office, MS Visio, and MS Visual Studio all look different (hint: they use different toolkits!)

I don't know about Visual Studio, but Visio was another one of those software projects that Microsoft acquired when they realized that it could be useful to have a pseudo-drafting type of drawing program. It was only in Office 2000 that Visio showed up as an "integrated" application. It is still first generation and only shows many differences in the splash screen and such.

3) Why does every other Windows apps (Winamp, Windows Media Player, Ephpod, etc, etc) use their own weird-looking skin?

I think you answered this one under #1. ;-)

Re:unified desktop

[BenjiPenguin]

I think a lot of it is psychological, as in it's what people have been using for awhile, and to them it looks solid, even though to a newcomer it's probably pretty ugly.

Re:unified desktop

[upsidedown_duck]

Probably the only conclusion that can be made about the desktop debate is that nearly everyone on all sides of the argument is living in denial. For every inconsistency in a UNIX/X Windows desktop there is one in MS Windows, for every compatibility problem in a UNIX/X Windows based desktop there is one in MS Windows, for every administrative problem in UNIX/X Windows there is a problem in MS Windows. It's a global case of one's own farts not stinking, one's own kids not being jerks, etc. I think the desktop debate has decomposed down to the point of Ford vs. Chevy, complete with those stickers of Calvin peeing on stuff.

A much more important debate is open standards in MS Windows and MS Office. Let people live with their own problems in their desktop of choice, but at least let them communicate.

Re:unified desktop

[Brandybuck]

The Inconstancy of the Windows Desktop

One of the more insistant and vocal themes heard in the desktop debate is that that Unix desktop needs to be like Windows. It is said that multiple widget toolkits, inconsistant dialogs, and other evidences of a decentralized development model must be removed before the masses will accept a Unix destkop. This cry for uniformity can be especially shrill, almost as if the very survival of a certain free operating system depended upon it. But is the underlying premise true? Is Windows really a consistant and uniform desktop?

The answer is resoundingly negative.

While conducting a quick survey of configuration dialogs under Windows, in an attempt to understand what a newbie user of my software would be familiar with, I discovered that there was no standard procedure for these dialogs. Even configuration dialogs from the same manufacturer varied wildly. By all Slashdot accounts, Windows users must certainly be mentally damaged from their constant exposure to such inconsistant interfaces.

Where is the configuration dialog located for a Windows application? Using the Windows system I use every day at work, I discovered that even this simple item was highly variable. Microsoft Word had two configuration dialogs, "Tools->Customize" and "Tools->Options", while Microsoft Outlook added an additional "Tools->Services". Microsoft WordPad had only one under a completely different menu "View->Options". Moving on to non-Microsoft products, I see that Adobe Reader and Quicktime Player have "Edit->Preferences". But lest you think those are consistant, Adobe Reader has a single dialog, while Quicktime Player has a submenu of three dialogs. Firefox and Roxio Creator Classic follow the WordPad model of placement.

What about the dialog contents themselves? Microsoft Word has modal tabbed dialogs, while Microsoft Outlook has a modeless tabbed dialog without a help button. Adobe Reader and Firefox have modal dialogs using a listbox instead of tabs to separate the pages. Quicktime Player is similar, but uses a combobox instead of a listbox. Some of these dialogs had help buttons while the rest lacked them.

Okay, what about the look and feel? Certainly the Windows platform has a consistant widget set? Sadly, no. Adobe Reader has an almost-but-not-quite Win2K look, that matches neither the Windows Classic nor Luna themes that comes with Windows XP. Roxio Creator Classic has a "brushed plastic" look with odd splitter controls. Quicktime player has, of course, a look and feel straight out of another operating system! Comparing native Microsoft applications only improves matters slightly. Microsoft Word has a completely different toolbar style than Microsoft WordPad! I could continue on to some truly egregious examples of inconsistancy, but I'll leave that as an exercise to the reader.

I think by now that I have thoroughly debunked the notion that the Windows desktop is uniform and consistant. The question remains though, is the Unix desktop better? The answer is similarly, "no". But since Windows isn't consistant, the urgency of the question is clearly lessoned. Newbies aren't going to be rendered insane by seeing Evolution running alongside Konqueror. They aren't going to go running back to Windows when their distro forgot to include Plastik icons with Mozilla.

Re:unified desktop

[LizardKing]

It's not a RedHat thing, as my laptop is running NetBSD. The Mozilla family uses XUL, which is a way of specifying an interface in XML - broadly similar to libglade. The interface is then rendered using the underlying toolkit, which is GTK+ on Unix, although there is or was a rare Xlib interface at some point.

Re:unified desktop

[pyros]

ok. but for a long time running the mozilla packages from redhat used system colors and fonts and stuff only in the default theme, and the mozilla.org package did not in any.

Re:unified desktop

[Anonymous Coward]

Fortunatly, the default theme for Gnome 2.8 will be Indubstrial, which is based off the very smooth Industrial theme.

Actually, I think they decided in the end to use a variant of the Smooth engine, rather than Indubstrial. I might be wrong though... I'm going by stuff I saw on mailing lists.

Re:unified desktop

[lobotomy]

I have to speak up here. There is one area where Windows is consistent and Linux is a mess: keyboard shortcuts. In every one of those examples you cite, I could sit down and navigate through all of those without a mouse with no problem. I would know how they work. They would be consistent.

Ever try Linux apps without a mouse? Next to impossible. Whatever happened to alt+F4 to kill a window? alt+F3 to minimize -- or was that send-to-back, my memory fades as less and less use the old ways. Somewhere around the time Enlightenment became the default window manager in Red Hat, those old window control shortcuts went away. Why? I speculate that some brain trust thought it was too much like Windows and thus needed to be removed. In actuality, it was part of IBM's CUA (Common User Access). Notice the "C"? alt+F4, et al. worked in Windows 3.1, OS/2, Motif, IRIX, SunOS, etc., etc.

What do we have now? A huge mess. I don't even know what the keyboard shortcut is to close a window in my current system (Fedora Core 2). And when we get into the applications, it gets even worse. Part of the problem is figuring out just whose responsibility it is: the toolkit? X? the Window manager? the application? All of the above? None of the above?

Why do I want consistent keyboard shortcuts? Because I want to be productive and not have to rely on a point-and-drool interface. If I know what the shortcuts are, I can type much much faster than I can use a mouse. It seems that my only hope now is the accessibility people: they also want consistency.

Now, before some smartass replies that I could just customize my shortcuts, let me shoot that down. How would that be "Common"? I could set up my account to be just the way I like it. Then what happens on the next system? And the next? And the next? Am I supposed to spend half of my life customizing brain-damaged user interfaces? No thanks. I work on Windows, Linux, IRIX, and Solaris. Three of those do a pretty good job at consistency -- one of them is a failure (I leave it as an exercise to the reader as to which one it is).

Re:unified desktop

[omicronish]

I've thought about this before, and then concluded that while things may look different, they at least behave the same generally. You can't guarantee that similar widgets of various *nix toolkits will work the same, especially menus, which all seem to behave in different ways.

The fact that Visual Studio menus have icons on them and a different color scheme from most applications is no consequence to me. At least I can consistenly press Alt, F, down, O to get the Open dialog box.

And about why some Windows apps use their own skin, blame it on the developer (yes, even WMP) :P Windows has no say in the development of the Winamp UI.

Re:unified desktop

[Brandybuck]

First of all, my survey was on configuration dialogs. I didn't just pick this area out of my hat, but it just happened to be a major inconstancy I tripped over the other day.

You are right however that keyboard shortcuts are more consistant under Windows than elsewhere. This is despite the fact that the Windows programmer STILL has to connect most of them up by hand. A few like Alt-F4 might be hardwired by the system to specific events, but most require cooperation by the developer. Ditto for standard accelerators.

But direct your anger at the correct party. Microsoft's standards are for Microsoft. Solaris doesn't use it, despite your assertion that it does. What Solaris does instead is to follow the Open Group's keyboard standard. For those keystrokes that the OG doesn't ignore, they pretty much follow the Microsoft way (Alt-F4 is used for "close" and not "close window"). The problem is that the Open Group standard is very hard to find and very expensive when you do. So Free Software developers simply don't bother with it. That's where the inconsistancy is coming from, some developers will use Microsoft's standards while others will studiously avoid them.

Re:unified desktop

[Sinner]

1) Why does Luna look like a pre-schooler threw up after eating several crayons?

I might as well ask, "Why does 'classic' look 10 years out-of-date?"

I like Luna. It makes Windows un-ugly enough to use. This is no small achievement. It's at least a generation behind the latest KDE and Gnome designs, but then, this is proprietary software. Lower your expectations.

2) Why do MS Office, MS Visio, and MS Visual Studio all look different

This basically comes down to the same thing as "Why aren't Microsoft developers worried that everyone outside Microsoft hates them?" "Because they're too busy worrying that everyone inside Microsoft hates them."

Why does every other Windows apps use their own weird-looking skin?

Because if they used the 'classic' look, their applications would look 10 years out-of-date. And not everyone has Luna.

4) Why do the buttons on every single installer all look different?

I don't even know why Windows applications use those installers. Are Windows users still impressed with them? Even open source applications have them! The Open Office installer is pretty cool the first time, but by the third or fourth time I'm wishing someone would port apt-get to Windows.

I like those Microsoft installers that give you a big square "Click here to install" button when you run them. Do they think I'm running the installer for some other reason?

But what's even better than that is when you use "Add/Remove programs" to uninstall some software and it runs an uninstaller program! There seems to be two schools of dependancy management in the Windows world; a) ask the user, and b) blindly delete/overwrite everything.

Sorry, got a bit carried away there.

Re:unified desktop

[LizardKing]

RedHat may well have provided tweaked packages with their distro which included the work of one of their hackers, Chris Blizzard. He has done a lot of work on Mozilla, and I recall seeing a post from him in which he said he was working on making Moz honour themes. He also did much of the work of porting Moz to GTK+ 2.x.

Re:unified desktop

[Anonymous Coward]

When I right click on a hard drive parition and choose "Format", I want it to just format it right away. Do they think I clicked "format" for any some other reason?

When I accidentally double-click "Uninstall Windows.exe", I want to uninstall windows immediately, do they think I ran it for some other reason? I want my installers to make me coffee!

x.org rules!

[DMJC-L]

the new xserver kicks ass.. I've got it running on my desktop, compositing is a great effect, and with proper integtration with programs, promises to change the way i use my pc for the better... btw windows don't stutter when I move them! drop shadows are sexy too.. hopefully we'll get PLG features in a compositor in the next few months.

Re:x.org rules!

[JohnFluxx]

PLG features?

Re:x.org rules!

[Anonymous Coward]

PLG?

Re:x.org rules!

[Anonymous Coward]

PLG features?

Puffy Leprous Gonads

X.org is getting up to speed qute nice

[Anonymous Coward]

One of the things that has always bothered me about XFree86 in the past 6 years I have used linux is XFree86's kind of lag in new releases... development seems to move at a snail's pace, and let's be frank, it's almost the same as it was back in the good ol' unix days.

I for one enjoy X.org and a windowing system that can hopefully be kept up to date and have more active development.

But my question is... how many more forks will we have?

James Carr

Re:Does it run linux?

[DarkOx]

wrong again, I think most would say there is a major trend in IT right now to move the applications back to the back room. Noticed all the web based applications lately. I would call that a thin client app. Lets face it a web browser makes a terrible thin client, but its the quickest way to convert all those idle computers on peoples desks back into terminals. I think the future is think but not fat client architecture. It will all for end user systems to have much longer life spans and make handhelds ever more practicle.

Prepare to be blown away

[Anonymous Coward]

I'm currently using a beta release of the new Xorg and whoa is it nice. Finally true transparency, nice real dropshadows, etc. are possible.

There are probably more exiting features than the inclusion of Composite in the next releas (XDamage seems to be a great step forward for X over the network for example and XCB looks interesting too, RTFI) but hey, I'm just a sucker for eyecandy. ;-D

All in all I do get the impression that we all should thank Mr. Dawes for behaving in a way that lead to a fork of XFree. Xorg and freedesktop.org put the development of X back on track and it is only just beginning.

Finally, thanks to all the folks at freedesktop.org for doing such a great job and putting the fun back in my computer.

Re:Prepare to be blown away

[Queuetue]

Where do you actually see the new eyecandy? Dropshadows, transparency, etc ... Don't new window managers and applications have to be built to take advantage of them? (I've only read part of the article, so sorry if I missed this answer there.)

Re:Prepare to be blown away

[Anonymous Coward]

Yes and no. ;-D
You need a composition manager to see the eye candy. Currently there is only xcompmgr, the example composition manager from freedesktop.org. And they also provide transset a program to make windows transparent once xcompmgr is running.

Now all these things are of course not integrated into the WMs and DEs, so currently if for example you run xcompmgr -c (this option gives you the nice shadows) with gnome or kde everything will have a shadow and there is no way to exclude certain things from it.

So in this respect, yes the window managers and apps will have to be rebuilt to fully take advantage of the technology. But afaik they are allready working on a composition manager for metacity (the gnome wm) and there has also been some work done for kde.

So the technology for the eye candy is finally here, you can already have a very nice impression of what this technology is capable of but there is of course still work to be done.

Re:Does it run linux?

[kidgenius]

The other nice thing about using a web browser as a terminal is that HTML is very, very easy to write. I'd put it on par with creating a Windows form in VB. The other plus side to an HTML interface is that you now have a "program" that is mutli-platform. Mac, Windows, Linux can all view the page and get the program to do whatever it is supposed to. It's almost like Java, but I would think it is a lot lighter.

Re:Does it run linux?

[dave420]

I don't think the web is a terrible thin client - it's incredibly cross-platform (heck, even my phone has an HTML browser on it), has a wide variety of input methods and control, and can support client-side processing of small-ish chunks of data. I use it for web-based apps all the time, and I've not found something it can't do yet :)

Time for X11R7 or even X12

[Viol8]

Its time a load of heads sat down and decided on the features that are required in the next MAJOR release of the X windows system/protocol. None of this piecemeal "we'll add it in as an extension" rubbish thats been happening for the last 10 years as this is becoming unmanageable; "My server has the dbe extension but not open-gl, your server has shapes but not etc etc etc." Just put ALL modern graphics requirements in the base protocol and write new extensions for Xlib and work from there.

Re:Time for X11R7 or even X12

[Anonymous Coward]

Please try to at least get a little bit of information before starting a rant.

The folks at freedesktop.org are in the middle of a whole redesign of X. For example, this is probably going to be the last monolithic release of X, they are working on a replacement for xlib (xcb), there is a lot of discussion about putting certain function that are now provided by X into the kernel (RTFI for more information).

Does that count as a major redesign?

Re:Time for X11R7 or even X12

[cortana]

Then in ten years time, you will end up with exactly the same situation we are in today; obsolete crap in the base protocol, all effective new development in extensions. Except that you will have utterly broken backwards compatibility in the process. :)

Time and time again, X11 has showed us that it is better to provide mechanism, not dictate policy--even unto the protocol itself.

The Extensions mechanism provides the X11 protocol with extrodinary forwards compatibility.

You can take a modern X11 Window Server from 2004, connect to it a crufty old X client from some godawful old piece of embedded hardware from twenty years ago, and have it work perfectly. At the same time, your modern server can perform nifty tasks that the protocol's designers never dreamed would be necessary, such as, well, everything Keith Packard and co are doing today. :)

Re:Time for X11R7 or even X12

[Anonymous Coward]

You are 100% correct that the plan is to replace xlib, which would essentially create an "X12".

However, this information has not been publicized to a great degree, probably because everyone is consumed by the XFree fork and the shortterm changes to the codebase.

Re:Time for X11R7 or even X12

[Anonymous Coward]

reminds me of the dilbert cartoon where PHBs recommend doing everything they are currently not.

"it should be monolithic, not modularized"
then you would complain:
"i dont use any of that stuff, make it modularized, what moron said it should be monolithic"

Re:Time for X11R7 or even X12

[Anonymous Coward]

Submit a patch..

Re:Time for X11R7 or even X12

[Viol8]

"The Extensions mechanism provides the X11 protocol with extrodinary forwards compatibility"

Yes it does , but at some point its time to say that "this functionality would be better served being a core part of the system". Eg transparency.

Re:Time for X11R7 or even X12

[Viol8]

"Does that count as a major redesign?"

No. I'm not interested Yet Another Open Source Groups pet project , I want ALL parties who use and/or develop X to sit down and agree on a COMMON standard that comes shipped AS THE DEFAULT on all unix systems, not an optional alternative you download of some website no one has heard of.

Re:Time for X11R7 or even X12

[Anonymous Coward]

This is from Keith Packard and X.org, so it is coming from the "core" and not some student's pet project.

BTW, who's really left in the X11 World? You've got Linux/BSD distros, some Sun workstation stuff, and Hummingbird/XCeed on Windows. There's not exactly a lot of parties left.

Re:Time for X11R7 or even X12

[cortana]

But it wouldn't. Plenty of hardware can't do transparancy, and is used on systems that aren't powerful enough to do the job in software. Besides, it would break the protocol.

People said that fonts would be better served by making font rendering a core part of the system. What do we have today to show for it? A crufty, obsolete, nonextensible set of functions for drawing glyphs on the server side, that no new development uses because Xft/pango/fontconfig work together to do a much better job on the client side.

No one foresaw anti-aliased text, Unicode, truetype fonts, glyphs drawn with an alpha channel, etc. Fortunatly the mechanism that X provides allows a client to use these features without requiring every X server it comes into contact with to be upgraded to X12 or whatever.

Re:Time for X11R7 or even X12

[Viol8]

Sun

HP

SGI

IBM

You know , small little companies like that who all write their own X servers for their versions of UNIX.

Re:Time for X11R7 or even X12

[Anonymous Coward]

*sigh*

And they are all members of x.org that brought you this server you are right now bitching about.

Re:Time for X11R7 or even X12

[be-fan]

All four of those companies are members of X.org.

Re:Time for X11R7 or even X12

[FooBarWidget]

And how many modern X servers don't support modern popular extensions? Let's face it: XFree86 and X.org are the most popular X servers. Their codebases are compatible so the NVidia driver works on X.org too. Both of them support modern extensions like Xrender. Commercial X servers that don't support modern extensions will lose customers.

In order words: the free market will force other (commercial) X server vendors to support new, popular extensions.

Re:Time for X11R7 or even X12

[EllF]

It's amazing how readily some people will bitch. For we -have- a standard, one that works precisely because it is so extensible -- you can use ancient clients without any problem because the core mechanism is (relatively) simple, and your modern system can take advantage of things like Xrender support via extensions.

Of course, you could tell that this was a case of ignorant ranting when the original poster referred to X.org, which is helmed by Keith Packard, as a pet project.

Re:Time for X11R7 or even X12

[groomed]

Time and time again, X11 has showed us that it is better to provide mechanism, not dictate policy--even unto the protocol itself.

Um, X is a textbook example of that philosophy gone horribly wrong.

To its credit, the X consortium tried to rectify their mistake thru the ICCC, but again they fell into the trap of creating something with so much misguided "flexibility" that it's almost impossible to find any apps which actually implement the ICCC in full, let alone cooperate in any meaningful way.

The ICCC has been such a joke, that ten years on something as elementary as copy-paste is still a hit-and-miss affair. And what about projects like the CDE, or Enlightenment? Everybody's who been serious about using X to craft a desktop has felt the need to introduce policy above and beyond what X has to offer. That's not a sign of X's flexibility or time-tested design: it's a sign that X sucks.

Let's hope freedesktop.org manages to beat some shape into the mess that is X, but they'll only succeed if they're willing to provide policy rather than mechanism. Bad policy that's followed by many is much more useful than good policy that's followed by few, or no policy at all.

Re:Time for X11R7 or even X12

[Minna Kirai]

No one foresaw anti-aliased text

Funny, but anti-aliased text was available on consumer PCs two years before the X project even started...

Re:Time for X11R7 or even X12

[dbIII]

And what about projects like the CDE, or Enlightenment?

Both were a success and both taught us a lot. Every recent window manager has aspects of Enlightenment in it. CDE a failure? Ever thought of how KDE got it's name? Gnome of course was originally a politically motivated copy of KDE, and hence CDE twice removed, but of course it has gone well beyond that and is useful.

Of course all of these things are applications that live on top of X, have absolutely nothing to do with the X consortium or the ICCC, so are irrelevant to your argument. Things could be better, but please try to be in the geographical vicinity of the discussion.

Re:Time for X11R7 or even X12

[groomed]

Both were a success and both taught us a lot.

They taught nothing that MacOS didn't already teach us.

Ever thought of how KDE got it's name?

KDE got it's name because CDE was a failure. Otherwise, there wouldn't have been a need for KDE.

Of course all of these things are applications that live on top of X, have absolutely nothing to do with the X consortium or the ICCC,

They're desparate attempts to bring some much-needed structure and policy to the X platform. In that sense they're very closely related to the ICCC and the X consortium.

Re:Time for X11R7 or even X12

[dbIII]

Re: CDE, enlightenment
They taught nothing that MacOS didn't already teach us.

There is a very long list of differences between the three.

KDE got it's name because CDE was a failure

I have a couple of machines running that "failure" not far from me now. The major problem with CDE was an attempt to define what desktop everyone would use, and not everyone is happy with a definition they have no input into - and it has taught us that one size does not fit all. Gnome has stepped onto this path a few times.

Enlightenment took the opposite extreme. You want something as different as radial menus? The functionality to do it was built into Enlightenment for a while. Things like the difficulty of animating icons without nasty hacks (animate the entire root window but the only changes are in the icons) led Raster and others to develop some libraries to sit between X and applications to solve some of the problems they saw when developing enlightenment. Once again, all this happens in a layer above X.

They're desparate attempts to bring some much-needed structure and policy to the X platform

No - once again they are applications that sit on top of X, not part of X or X extensions. A statement like that is like saying a visual basic program redefines the entire win32 kernel.

Remember X itself handles fairly simple stuff - to put things in perspective the twm window manager is a level above it.

Re:Time for X11R7 or even X12

[MegaFur]

but again they fell into the trap of creating something with so much misguided "flexibility" that it's almost impossible to find any apps which actually implement the ICCC in full, let alone cooperate in any meaningful way.

But isn't this also true of TCP/IP and all the related protocols that run on top of it? I mean, I've `chatted' with several different SMTP servers via telnet and ftp servers via a good ol' CLI ftp client and, ya know, it's the darndest thing--different implementations of the same service don't all do things the same way. And some are more broken than others...

And yet, somehow the Internet just keeps on tickin'.

Actually, if you take a step back, you can see this same sort of semi-broken-ness in various Western political systems on both the local, national, and international levels. I agree it would be better if the system were all perfect and stuff. But I don't think it will ever be. And most people seem content to accept it in its semi-broken state as long as it works good enough for what they require*, and as long as it remains flexible.

Naturaly, there are always those that aren't happy with the current state of affairs and gripe about it*. But most of those gripers don't actually do anything to try to fix it.

Do you?

* (I gotta agree with you on the copy-paste issue though. Uncertain copy-paste is bad news. Maybe we can form a Concerned Users For Better Copying And Pasting initiative or something. :-) )

Compositing is just the tip of the iceberg.

[Anonymous Coward]

And pretty slow it seems.

Right now I am running fedora core 2 and am using the latest release from X.org's CVS.

It seems stable and all that, but it's slow.

GLXGears I am scoring 285 fps with xcompmgr off
and 60-70 with it on. (that turns on the composite features).

Although it does have my dri drivers turned off in both cases (using intel i830-type video driver). I am recompiling as I type right now to enable the new i915 driver for it to see if that makes a difference.

But other people have reported it to be slow. Probably would be nice on my other computer using the Nvidia FX 5900 XT, but I don't want to mess up my desktop with a CVS-based X server.

All in all it's pretty stable and shows the progress that XFree86 was holding back on, unfortunately. Yea for X.org

Oh and also for that guy that says he was nervious about X.org and Freedesktop.org and KDE/Gnome "working to close together". He is a idiot. This isnt' X Windows, this is just the X SERVER. It's one part.

What I'd worry about more is X.org and Linux getting to cozy and unintentially making it more difficult to run on other Unix-like OSes.

X.org has a open invitation for all Unix developers and it would be great if they would get more of their input. (Especially the BSD's)

The future looks good. X.org would like to strip away the dual nature of X's drivers (Mesa/Dri OpenGL drivers + XFree86-type 2D drivers) and get the X server running on pure OpenGL!

That means instead of having to write 2 versions of drivers for video cards, now they only have to worry about the OpenGL version. This means it's easier to get good drivers for Linux and other Unix-like OSes that use X.org servers, and quicker too.

Also the Cairo project is going to be integrated bringing in Vector-based Windows and graphics libraries into X windows and allowing them to also be OpenGL accelerated.

The MS Longhorn waiters, eat your heart out. This is going to be some cool stuff we will have in the next couple years.

Of course OS X is openGL, too, but the cool thing about X windows is the flexibility. All these changes will keep complete backwards compatability with older programs (X clients actually in X terminology), while removing bloat for features that nobody uses/completely obsolete and streamlining developement thru modularlization and extensions.

Stuff like Damage is reducing the X networking load considurably too, making wide spread use of X terminals in businesses and schools more and more fesable.

And all sorts of other improvements are coming.

Changing over to X.org seems to have been a fortuninate move.

Re:Compositing is just the tip of the iceberg.

[Anonymous Coward]

If you have been following the mailinglist you should know that there most definitely are people working on X who use other (than linux) operating systems.

Re:Compositing is just the tip of the iceberg.

[Anonymous Coward]

Although it does have my dri drivers turned off in both cases

well duh. No hardware acceleration == slow. With hw acceleration, it's near no-op, free operation.

Re:Compositing is just the tip of the iceberg.

[caseih]

Actually many normal operations appear much faster with composite on than off. Dragging windows, for example. The other extensions plus the off-screen rendering make X appear to be a lot smoother and faster. I remember running an early beta of xserver (kdrive) with the extensions using only the VESA driver. Without the composite manager on, the system was slow some things were just painful. Turn on the manager and normal operations appeared to be almost an order of magnitude faster. When the synchronization stuff is added into GTK and QT, resizing windows will also appear to be much much faster and smoother.

Wtf has the printer got to do with X?

[Viol8]

Please explain how the printer relates to the X windows system. I'm all ears...

Re:Wtf has the printer got to do with X?

[TheRaven64]

In a well designed windowing system (such as Display Postscript, Quartz and even GDI although it falls down in a number of other areas), the drawing commands sent to the windowing system are exactly the same as the ones sent to the printer. This makes it very easy to create true WYSIWYG applications (you don't need to write an X11 rendering path and a PostScript rendering path for the same data, and hope you've done it correctly). The Xprint extension provides this functionality to X11.

Re:Wtf has the printer got to do with X?

[grrussel]

Printing to the screen versus printing to paper?

Why distinguish? An application should be able to use the same commands to draw on screen as to a printer, which is just a different display device.

Re:Wtf has the printer got to do with X?

[Viol8]

Its not a different display device , its a batch hard copy device. If you don't believe print something out then try to get the printer to update it.

Re:Wtf has the printer got to do with X?

[Viol8]

Rubbish. Printers are print once then forget, whereas screens can consist of lots of constantly moving graphics. They're entirely different enviroments and the only place where your argument would be valid is in publishing situations where the screen has to match the printer output exactly and moving graphics are irrelevant.

Re:Wtf has the printer got to do with X?

[Anonymous Coward]

Argued like a true dingbat: Because paper is a static diplay medium therefore '0 0 \moveTo 10 10 \lineTo' is beter than 'XDrawLine(0, 0, 10, 10);' Therefore we should force people to learn both even though one is a subset of the other.

Re:Wtf has the printer got to do with X?

[johannesg]

So you need an entirely different interface for printers because on the screen you may have animation? That's a pretty amazing leap of logic.

Answer these two questions, if you will:

* What do you suggest we should be using on the printer? * Why does it have to be different from what we use for the screen?

For the record, I'm a software developer myself, and I'm extremely happy that I do not need to write and debug my code twice.

Re:Wtf has the printer got to do with X?

[johannesg]

So I guess you also believe we should really be using an entirely different set of commands and functions for accessing harddisks and CD's? Since they are also read/write and write-once, respectively.

Personally I'm happy with using cp for copying files. I wouldn't like to need cp_hd2hd, cp_hd2cd, cp_cd2hd, and cp_cd2cd. I'm also happy with using ls for seeing directories, instead of having to choose between ls_hd and ls_cd. I could go on for a while, but I think you get my drift...

And the same thing is true for rendering: whether it is to a dynamic device or a write-once device _really shouldn't matter at all_.

Re:Wtf has the printer got to do with X?

[Viol8]

Thats fine for simple line drawing. What however if you want double buffering such as X windows supports , or even stuff like XORing while drawing the line? Or would you suggest that everytime someone moves a window a few centimeters the ENTIRE screen definition in postscript gets resent to the graphics drawing device? Or what about stuff like OpenGL? YOu want to implement texture mapping, shadowing, lighting etc in postscript? Could you pick an uglier language to do it in?

Re:Wtf has the printer got to do with X?

[Viol8]

Theres no "may" have animation about it. When was the last time your saw a computer screen that never changed?

"Why does it have to be different from what we use for the screen?"

Double buffering, real time shading, lighting & caustics, textures, Z buffering, bump mapping, motion blur, flare, scanline rendering etc etc etc

You ever done any graphics beyond cut-n-pasting a picture into word?

Re:Wtf has the printer got to do with X?

[Anonymous Coward]

You've got it backwards dingbat. You don't use PostScript to draw to the screen. You use X drawing routines to draw to the printer. Since paper has a subset of the functionality of a screen, all of the X commands that don't make sense on paper could just be ignored when drawing to the printer device. The application writer will have to know which X commands don't make sense when drawing to the printer, but it's a heck if a lot easier than learning a whole new language just to get something to print.

Re:Wtf has the printer got to do with X?

[Viol8]

The X protocol is very much 2 way. It relies on you creating windows and graphics contexts and using those handles to communicate with the server when you request an action. HOw to you square that with a printer which as far as graphics goes is very much a passive device and just spits out what it gets fed? Would you suggest implementing some sort of server on it and you then have a long winded communication with it to draw something rather than just chucking a large postscript file at it? Sorry , seems crazy to me.

Re:Wtf has the printer got to do with X?

[johannesg]

Thanks, yes I have done a lot of graphics beyond cut'n'pasting into Word. I write commercial data acquisition, analysis, and presentation software for the space industry (mostly in C++), and I've written quite a few interesting presentations that run on both Windows and UNIX, and can display stuff on the screen _and_ on the printer. And I'm telling you, I want the same primitives for the screen and for the printer. Because in the end they are both just rectangles of pixels, and I don't want a different interface just because that rectangle is interpreted differently. I have better things to do with my time than write and debug the same rendering code twice just because some two-bit punk thought it would be nice to have utterly different interfaces.

I'm still unclear about why you think it would be a good idea to have those different interfaces, BTW. Just naming a few capabilities that are unique to the screen does not invalidate my position. To put this in OO terms (I'm going out on a limb here assuming you are a programmer): screen rendering and printer rendering both inherit from generic rendering. They should not be unrelated classes.

BTW, is there any good reason why an image on the printer should _not_ support real-time shading (is it actually required to be slow?)? Is there a good reason for printers _not_ to display 3D images that make use of lighting, textures, and the other stuff? I mean, if I had a problem like that I'd sure appreciate the ability to just render OpenGL to the printer, and having the OS accelerate the entire thing through the GPU. That would be much better than having to rewrite my entire fucking 3D render engine in fucking postscript, if you'll excuse my language.

How about your qualifications? Or are you just a trollish fanboy who happens to think that whatever X does, must therefore be the One True Path?

Re:Wtf has the printer got to do with X?

[Viol8]

"I'm still unclear about why you think it would be a good idea to have those different interfaces, BTW"

Because a screen requires constant information flowing to it which requires a 2 way on-going protocol. Printers are fire and forget, they have no need to be 2 way. And why would you want to build loads of graphics functions into a printer when you can just send it a simple bitmap that you rendered on the computer?

"some two-bit punk thought it would be nice to have utterly different interfaces."

I think those "punks" were a bit more clued up than you.

"How about your qualifications? "

I've written 3 freeware games and a number of realtime simulations in 2 & 3D. To be childish for a second I think that beats some pissy "presentation software". Whats that ,
fucking bargraphs and the like? BFD.

Re:Wtf has the printer got to do with X?

[Anonymous Coward]

The X protocol is very much 2 way. It relies on you creating windows and graphics contexts and using those handles to communicate with the server when you request an action. HOw to you square that with a printer which as far as graphics goes is very much a passive device and just spits out what it gets fed?
What the hell do you think a graphics card does? Doesn't it also spit out to the screen whatever you feed it. An a printer is not a 1-way device. Papers get jammed, toner cartriges run out, etc. so it needs to comunicate errors back.

Would you suggest implementing some sort of server on it and you then have a long winded communication with it to draw something rather than just chucking a large postscript file at it?
You don't think PostScript is long winded communication? Have you ever looked at PostScript code? Did you know that you can write ray-tracers in PostScript? Implementing an X server in a printer shouldn't be any more difficult than implementing a Turing-complete interpreter for the PostScript language with a stack, variables, functions, et al. And even if the printer only speaks PostScript, the X server running on the computer could implement a X -> PostScript translation layer. So evey X app could use it instead of everyone re-implementing the wheel to generate the required PostScript to send to the printer.

Sorry , seems crazy to me.
Simple and practical ideas always perplex dingbat minds.

Re:Wtf has the printer got to do with X?

[johannesg]

I don't want to build all those graphics functions into the printer, I want the OS to handle it. Did you miss the remark about the GPU speeding up my rendering? I wasn't talking about putting a GPU in the printer.

So we agree that you create a bitmap first and then either display it on the screen or on the printer. ***WHY*** does there need to be a difference in how you create the bitmap?

Here's a suggestion: Windows has a unified rendering model. MacOS, as far as I'm aware, has it too. AmigaOS had it. In fact the only OS I could think of that doesn't have, and that by strange coincidence has piss-poor printer support, is UNIX. Current thinking, as exemplified by the Cairo project, suggests that unifying rendering is a very good idea. So all the evidence points to a lack of proper design by the X people, and no matter how much you rant about two-way protocols (whatever that is supposed to mean? Do you actually read back the pixels after writing them?) is not going to change that.

Re:Wtf has the printer got to do with X?

[4of12]

Amen.

A resolution independent system would really help out on this problem.

NeWS, Display Postscript didn't catch on.

Maybe SVG will.

But I've wondered whether the underlying ddx for X shouldn't move towards a generic OpenGL driver...

Re:Wtf has the printer got to do with X?

[Anonymous Coward]

Well, it just happens that you use different commands to write to a HD than to write to a CD. Same thing applies here.

Besides, how will you handle page breaks on a printer which doesn't even exist on a screen ?

Re:Wtf has the printer got to do with X?

[dbIII]

Why distinguish? An application should be able to use the same commands to draw on screen as to a printer

Does your double sided monitor do it's page flip on the long edge or the short edge? How about a 600dpi monitor with 42" on the short edge and the other distance governed only by the length of the roll?

There are some reasons to treat these things differently - and it is usually done at the application level.

Re:Wtf has the printer got to do with X?

[Viol8]

"so it needs to comunicate errors back."

Which bit of "as far as graphics goes" didn't you understand? I assume english is your first language?

"Did you know that you can write ray-tracers in PostScript? "

No shit! And you ever chucked postscript code at a dot matrix and seen what comes out? Printers are not just fancy laser jobs. I'm talking about ALL printers. Or are you saying that all logging printers must now have some complex CPU instead of the simple 8 bit job and TTL they get away with at the moment and everyone has to pay more just to make your life a bit easier?

"Simple and practical ideas always perplex dingbat minds."

As you demonstrate so well.

Re:Wtf has the printer got to do with X?

[Viol8]

One reply: Dox matrix. The printer system has to deal with ALL printers and dox matrix and other simple "dumb" printers are used in LOTS of places still that have to do logging or simple continuous data dumping. I see no reason for these to be made complex just to keep people like you happy at an API level.

Re:Wtf has the printer got to do with X?

[johannesg]

Two words: "character terminal". The output system of the computer has to deal with ALL types of terminals, whether X or simply dumb VT100's. These are still used in lots of places, so... Wait. You don't use X to talk to a VT100, there is a dedicated terminal interface for that.

We could do the same thing for printers then: have one interface which is primarily text-oriented, and laid out much like a stream (this can be used to drive text terminals and matrix printers), and one that is primarily graphical (which can be used for X terminals and modern printers).

So your dot matrix can still happily print using nothing more than fopen, fprintf, and fclose (or whatever language you prefer). And at the other end of the spectrum, complex simulations (bar graphs? Don't make me laugh) can still be visualized using complex render commands.

Composite extension

[xenostar]

And of course, the great news in the new release is the addition of xserver's composite extensions to X.Org.

Thankyou sir

[Nailer]

I was somewhat hopeful that the impending death of X-Windows would lead to the development of a windowing system designed specifically to take advantage of the more advanced features of NetBSD. ...for cheering me up on this otherwise glum evening.

Re:Does it run linux?

[Anonymous Coward]

*posting AC because ruthless mods will say this is flamebait*

Probably because it IS flamebait!

Whose task is copy&paste

[dtietze]

This triggers something that's been bugging me for ages. Whose responsibility is a decent cross-application copy&paste framework, which works as one would expect?
As someone who often puts together presentations, marketing slides, flyers for printing, etc., this is my single greatest annoyance about Linux at the desktop (and we're using Linux on all our desktops; heck, we're even a SUSE technology partner). Copying text between my Java IDE and OpenOffice gives me only about half a page of text - the rest is simply lost. How on earth can I simply copy from GIMP into an OpenOffice presentation like I can copy/paste from PaintShop pro to PowerPoint? The last time I tried, I couldn't even copy/paste consistently between various KDE apps.
As much as I hate to say it (and I really hate to say it), this is *the* one thing that Windows does right. More or less seamless application integration which works the way I need it to work.

Dan.

Re:Whose task is copy&paste

[BenjyD]

Copy and paste works fine for me (in KDE3.3, openoffice 1.1.2). I can copy part of a drawing in Kolourpaint and paste it into KWord or Openoffice and it appears as an image in the document. It doesn't seem to take notice of circular selections, but that's hardly surprising.

It seems that the GIMP just doesn't support this yet.

Re:Whose task is copy&paste

[KamuSan]

And that's exactly why I bought a Mac.

Linux to tinker, Mac to use.

BTW. Windows doesn't really copy/paste well though. Formatted copy in WOrd gives me a headache and Excel doesn't keep to Microsoft's own UI guidelines.

Re:Whose task is copy&paste

[dtietze]

> BTW. Windows doesn't really copy/paste well though. Formatted copy in WOrd gives me a headache and Excel doesn't keep to Microsoft's own UI guidelines.

Oh yes, I know what you mean. Copying between different Word documents is a pain (who told Word that I want to copy the style information as well as the content? Thank g*d for the "paste special" command). And copying from IE to Word copies the HTML styles and lots of other formatting? What's that all about?

But STILL..... most Win apps can be expected to play well with each other through the system clipboard.

Dan.

Re:Whose task is copy&paste

[radish]

Recent versions of the Office apps fix the formatting problem. When you paste, a little menu appears letting you select "Keep Formatting, Reformat, Text Only" etc. Works well, and means you don't have to keep remembering to use Paste Special.

Re:Whose task is copy&paste

[imroy]

It's simple. You save to a file in app A and then open it in app B. Honestly, where's the attraction in having your data floating about in a clipboard like some etherial juggling act? Without extra tools you can only hold one thing in the clipboard. If you have to transfer many items you end up copying and pasting things one at a time, like a two-person boat transferring people across a river. But you can have as many files as you like. And they're only limited by your disk space, not your RAM+swap.

This Linux geek is quite happy with the cut buffer, complete with copy-on-select, thank you very much.

Re:Whose task is copy&paste

[jgardn]

Copy & Paste is not part of X. I wouldn't want it to be. Why should X know how to handle an HTML document and whether to paste just the text or the text and the images or the html code? Why should X be able to recognize URLs and email addresses that you copy, and suggest that you store them in your bookmarks or address book? Would you want copy and paste only in X applications? Why not from the text console as well?

Copy & Paste is not something X should even know about. Instead, it will be some other system service or a service initiated just for the user who is logged in.

To get Copy & Paste to work all we have to do is agree to a standard IPC method and a standard interface. Once we decide on that, it's a matter of getting all the apps up to the standard. So far, only Gnome and KDE have agreed to anything, and I think they both want a more general solution.

Re:Whose task is copy&paste

[Blakey Rat]

Why is that "hardly surprising?" I think your expectations are too low here. The only thing you need to do to copy a circular selection is to pad the rest of the rectangle with transparancy, why doesn't the paint program do it?

Re:Whose task is copy&paste

[BenjyD]

Because kolourpaint is a pretty new application, so features like that aren't implemented yet. Are you offering to send them a patch if it's so easy?

Re:Whose task is copy&paste

[Simon]

You save to a file in app A and then open it in app B. Honestly, where's the attraction in having your data floating about in a clipboard like some etherial juggling act?

The attraction is that a clipboard is about a BILLION times faster and easier to use than juggling files, if all you want to do is move something from app A to app B.

I can't believe this kind of stuff needs to be explained to people/geeks. WHAT PLANET ARE YOU PEOPLE FROM?!?

--
Simon

Re:Whose task is copy&paste

[Anonymous Coward]

Are you offering to send them a patch if it's so easy?

Are you? If not, please do send in patches instead of bitching around on Slashdot. This is Free Software, it does not work unless people are prepared to write the code!

Re:Whose task is copy&paste

[Control Group]

That's the kind of insular attitude that causes so many problems for Linux. Face it, whether you like c&p or not, many, many users like it quite a bit. If Linux is to be only a niche OS for the computer-savvy, then no universal c&p is just fine. If Linux is to become a commonplace OS on the average desktop, then it has to cater to what people actually want, not what superusers tell people they should want.

Besides which, the advantage to c&p is simplicity. Copying an image from a web page and pasting in paintbrush is more convenient than saving as something and then opening (rt click, copy, win+r, 'paint', rt click, paste, 'ok' on the resize to fit dialog; rt click, save as, click 'desktop', type name, click save, win+r, 'paint', file, open, click 'desktop', open). Ditto moving data from Access to Excel. Rather than save as a CSV file and open it in excel, ctrl+a, ctrl+c, win+r, 'excel', ctrl+v.

There are certainly times when saving and opening is more appropriate, but c&p is quick and easy. It's a simple tool, but for simple uses it's ideal (this is the same reason I don't buy into add-on apps that expand the size of the clipboard. The whole point for me is that it's fast).

I haven't used Linux in almost two years, but one of the two reasons I stopped was quite literally the difficulty in copying and pasting (though the ability to copy & paste at the console is spectacular, and MS has no excuse for not supporting it). The other was that I couldn't manage to get my wireless NIC working, but that's a whole different story.

Re:Whose task is copy&paste

[FooBarWidget]

Ehm... clipboardhandling already is part of X. It's called the "selection mechanism".

In fact, there is a standard. People used to misunderstand it but now it's documented clearly. http://www.freedesktop.org/Standards/clipboards-sp ec

The X selection mechanism supports contents negotiation, which means that you can copy rich text and images from Mozilla and paste it in OpenOffice with the same markup. In fact, it already can. I tried it - it works. Even images are preserved. Try copying some tables in Gnumeric and paste them in Mozilla Composer. It works - the tables are preserved!
The X selection mechanism is actually technically almost the same as MS Windows's.
Unfortunately not all apps use content negotiation, I don't know why. Gimp is the worst example - it doesn't use the X selection mechanism *at all*, not even on Windows! It uses it's own internal clipboard. So you can copy & paste inside Gimp but not between multiple Gimp processes.

Geez why do I have to post this this over and over? Isn't it about time everybody on Slashdot knows about this?

Re:Whose task is copy&paste

[spitzak]

You are right on track about the correct implementation, but it should not be visible to the user.

Any program that can work with graphics already has code to open a graphic file, interpret the contents, and insert it. It is stupid to have to have totally seperate code to interpret a memory buffer in a totally different format provided by an unrelated "copy and paste interface".

X has a chance to skip all the crap because the current copy and paste does not work anyway. Data copied should either be "UTF-8" or "URL", which are the only two things I have seen work in X programs anyway. If the user selects an image or anything that is not obviously a text string, then it should write that data to a temp file, and the copy should be the URL of that text file. And then when they paste it, the recieving program inserts that file, and gets to reuse all the interpreter they need anyway to insert data from files.

They also need to realize that the "middle mouse click" is not cut & paste, but in fact a much advanced version of drag & drop which allows the user to rearrange windows and launch & kill applications between when they "drag" and when they "drop". Ideally there should be NO difference between dragging a selection and middle-clicking in the receiving application.

Re:Whose task is copy&paste

[wormbin]

The above url was mistyped.

Freedesktop ClipBoard Specification

Thanks for the info!

Re:Whose task is copy&paste

[Blakey Rat]

No. I use a Macintosh and all the paint programs I use already do this, and have since 1984.

Re:Whose task is copy&paste

[BenjyD]

I already have several patches in KDE (KImageEffect module). I also run my own free software project which is currently at around 14,000 lines of code, 99% of which is my own work, with around 2,500 downloads/month.

Re:Whose task is copy&paste

[BenjyD]

No, new as in hasn't been in development very long. It's nothing to do with date.

Re:Whose task is copy&paste

[Minna Kirai]

The attraction is that a clipboard is about a BILLION times faster and easier to use than juggling files, if all you want to do is move something from app A to app B.

Furthermore, since X is a network-transparent protocol, there is no guarantee that two windows on the same desktop will even have access to a common file system! They might be running from servers in completely different cities, with no communication besides X.

I can't believe this kind of stuff needs to be explained to people/geeks

Hopefully, he was trying to be funny.

Re:Whose task is copy&paste

[Minna Kirai]

It is stupid to have to have totally seperate code to interpret a memory buffer in a totally different format provided by an unrelated "copy and paste interface".

Code reuse is good, but it could be accomplished in a better way than what you suggest.

If the user selects an image or anything that is not obviously a text string, then it should write that data to a temp file, and the copy should be the URL of that text file. And then when they paste it, the recieving program inserts that file,

And kill network transparency while you're at it?

Today I can ssh to two remote computers, run X apps on each one, and C&P data between them. Your proposal will break that ability, either completely, or only partially (in ways that are inconsistent to the end user).

A better approach:

1. On a Copy action, the application informs X that data has been copied, and makes a local copy in a RAM buffer (a step that can be optimized out if the document itself is unchanging).
2. On a Paste action, X goes back to the application where the Copy was invoked and tells it to write the data to a file, but to a special pseudo-file pointer that goes not to an actual filesystem, but rather to a buffer handled by the X server.
3. Then the X server hands that same pseudo-filepointer to the Pasting app, which then reads it like a file in the normal way

That requires one extra layer of indirection in the file i/o code, but allows code reuse while maintaining network transparency. (Naturally, in the common case that both applications are on PC (a) and the X server is on PC (b), a form of short-circuiting should take place to avoid excessive network traffic)

Re:Whose task is copy&paste

[spitzak]

Network transparency is why I said "URL", not "filename".

Obviously the idea requires some extra stuff, like getting url read/write out of Qt/Gnome and into the basic Linux system where it belongs (I should be able to "cat /http/slashdot.org" and get the main page, without "cat" being rewritten). Also it would help if there were clean methods of creating temporary files that just stay in memory.

Re:Whose task is copy&paste

[Brandybuck]

Even your 1984 Mac paint programs had to actually code this functionality into the software. That didn't happen instantly and by magic. KolourPaint is brand new, as in "it's still in the middle of being potty trained" new. Give it a tiny bit of time and that crucial bit of functionality will be added.

In the meantime I expect that KolourPaint can do thinks your 1984 Mac programs could only dream about. Like color, transparency, etc...

Re:Whose task is copy&paste

[Minna Kirai]

Network transparency is why I said "URL", not "filename".

What kind of URL were you proposing, then? I assumed "file://" at first. But if you claim it'll be network transparent, then "http://" comes closer- but still isn't good enough. For starters, you'd need to install webserver capability in each X applcation with a Copy command...

And even so, one can't be certain that two X applications on different computers will be able to reach each other by hostname, or even ip address.

Re:Whose task is copy&paste

[captaineo]

If you mount tmpfs (or is it shmfs?) on /tmp, you get persistent (until reboot) file storage that is backed by the system page file.

Re:Whose task is copy&paste

[spitzak]

I would guess that if the programs can draw to the X server on some machine, they can access files on it. So some URL that gets to the current X server would work.

If the X server is going to be a data storage facilty, I would like to be able to access it using normal Unix file I/O calls. Maybe this is part of the X server, or maybe it is expected to be another service running on the same machine.

Re:Whose task is copy&paste

[dbIII]

Whose responsibility is a decent cross-application copy&paste framework, which works as one would expect?

The application developers should probably read some docs on the underlying system before trying to implement a MS Windows style registry and MS Office style OLE.

Re:Whose task is copy&paste

[KamuSan]

Good!

Shame that I'm stuck with an old version of WOrd at work :-(

Re:Whose task is copy&paste

[yRabbit]

Ah, GIMP on Windows has "Copy to Clipboard" and "Paste from Clipboard", you can actually make it use the Windows clipboard if you want it to. GIMP 1.2.5, for me, used to clear it after being pasted (for some reason?) but GIMP 2 didn't the last I knew.

Re:Whose task is copy&paste

[Blakey Rat]

So they did the complex stuff before working on the very basics of the program (like copy and paste)? And people wonder why open source apps suck so much.

Re:Whose task is copy&paste

[FooBarWidget]

I'm saying it doesn't use the native clipboard by default.

Re:Whose task is copy&paste

[Brandybuck]

Well of course! If you are not a professional software developer this might not make sense, but it's a very common thing to do. You do the hard stuff first because it's hard and will take the longest. The easy stuff you can fill in along the way. That's probably why there are so many unfinished projects on Sourceforge, because a lot of people decide to work on the easy stuff first...

suggestion

[suezz]

my biggest complaint is the configuration of X. xf86config should just be plain outlawed. I am an experienced unix admin and love linux but the only real complaint I have with is the configuration of X. I can get it working with no problem with xf86config or x86setup - but I really like what fedora has done - it is a non issue and you don't even have to mess with it at install time - this is the way it should be. I have installed fedora on at least 20 to 30 computers and they all went without a hitch and I didn't have to have the monitor sync rates. thanks fedora and keep up the good work!!

Re:suggestion

[Anonymous Coward]

Fedora wrote a GUI fron-end to the configuration file.

the config file is still the same.

Your suggestion then comes down to: it would be great to have a simpler, reliable X configuration tool with the X server.

However, the distributions seem to have this covered already.

Is there any need to spend time on it?

Re:suggestion

[Scott Francis[Mecham]

Your suggestion then comes down to: it would be great to have a simpler, reliable X configuration tool with the X server.

However, the distributions seem to have this covered already.

Is there any need to spend time on it?

When even Knoppix can't get a DVI-fed LCD monitor working above 1024x768; and the XF86config it produces bears no resemblance to a GTF-produced modeline for the exact same resolution and frequency? I'd say there might be some reason there.

Re:suggestion

[mikefe]

How about merging what the distros have done into the base distribution?

Re:suggestion

[Anonymous Coward]

removing the modelines knoppix adds and letting X figure things out for itself works much better.

Re:Does it run linux?

[hey]

Terrible for some things. Not terrible for others. Look at all the websites that are applications. Eg travel sites. Select your city, date make a query.

Transparency

[Anonymous Coward]

The transparency (aka the composition extension) while will make it in, it will be unstable and will be turned off by default

Re:Does it run linux?

[freqres]

Try a web-based schematic capture or CAD/CAM then. I would be interested in your results.

If anyone knows the power of X . . .

[Anonymous Coward]

If anyone knows the power of X it must be Huey Lewis and the News.

Re:PRECISELY MOD UP+

[lokedhs]

"Internationalisation" is spelled exactly like so... In all english-speaking countries except the US.

Re:PRECISELY MOD UP+

[Anonymous Coward]

Uh.. that's not a spelling mistake, it's an internationalisation issue. "Internationalisation" is the English spelling. "Internationalization" is the American English spelling.

Re:Does it run linux?

[Anonymous Coward]

no you are posting AC because you are a pussy.

if you could back up anything you just said, maybe you would have been modded up.

but you are retarded pussy, that is why you got modded flamebait.

or you are just a loser, who happens to be an AC pussy.

if you want to post something controversial, back it up witha real account.

(why yes i know i am AC, but i am not posting anything that ANYONE will disagree with)

Re:Does it run linux?

[dave420]

I've not come across them, but most people wouldn't want a thin client running those anyway, as they require a lot of graphical processing, which isn't what thin clients excel at. Most areas where people would want a thin client can be suitably served (no pun intended) by a web-based system :)

Re:PRECISELY MOD UP+

[Ed_Moyse]

That is great! Internationalisation confusion in a thread about internationalisation. ;-)

Re:PRECISELY MOD UP+

[tehcyder]

The spelling with "s" is correct in the UK at least.

Aspect oriented X server

[Anonymous Coward]

The X developers should rewrite the server from scratch using the Aspect Oriented methodology and for example the AspectJ programming language. Many of the X extensions really touch all parts of the server which is exactly the kind of problem aspect oriented programming was designed to solve.

Using AspectJ, an extension such as the Damage extension could be written in a weekend.

Also rewriting the server in AspectJ would allow the developers to leverage the full power of the Java language. With Java reflection the core dispatch code in the server could be replaced by just a few lines of code. The RENDER extension could be completely removed from the server and replaced by using the delegate design pattern to forward X requests to Java2D

The Fresco project had huge potential, but never managed to escape the legacy language C++. It seems everybody working on window system is stuck in the software engineering practices of the seventies.

Re:Aspect oriented X server

[Desval]

Funny, I did not think that C++ existed in the seventies.

C++ was originally C with Classes created at AT&T in the early eighties (83 I think but I could be wrong). ANSI standardization came much later.

Re:Does it run linux?

[Anonymous Coward]

Main problems with X11 as a "thin client" architecture.

1) Most X11 implementations are nowhere near "thin".
2) Windows X Servers traditionally have been hella expensive
3) You are suggesting that people use Unix-based RAD tools, which have historically been mega-inferior.
4) X11 has shitty network performance, when compared to RDP, ICA, DCOM, HTTP, etc.

Distributed X11 is only really useful for running heavyweight apps like OpenOffice or something. Other than that its effectively only gets used for local windowing. Nobody builds a thinclient app around it anymore.

I'm up to your challenge

[hummassa]

Please name an application in which compositing gives a better user interface ...

I worked in a GIS (geoprocessing) application to an electrical company. In the user's screen, a map showed up with all polls and wires that are in a location. If you clicked on a poll with, e.g., a transformer, a translucent (big) tooltip came up with all of the transformers specs, where the electricity was coming from, where it was going to, etc (like 20 lines of text). Without dismissing such tooltip, the user is capable of clicking in another poll in the map, and only the contents of the tooltip changed, (maybe it's position if it were possible to move "away" from the current part of the map. The user could even click thru the tooltip, in a poll that was showing below it! (there was a menu item/toolbar speed-button and a hot-key to close the tooltip, obviously)

This kind of interface is *very* practical and would be impossible without translucency. I implemented it in a no-nonsense 15 minutes under BorlandC++/w2k.

Re:I'm up to your challenge

[Mitchell Mebane]

I worked in a GIS (geoprocessing) application to an electrical company.

And you still don't know how to spell 'pole'. Sad.

:P

basically 90% of developers moved to X.org

[Anonymous Coward]

weeding out the remaining 10% bitching license part...
You can look at it like getting rid of the snails...

I'm using the new X.org

[thejuggler]

I just installed Slackware 10.0 and it came with the X.org system. I didn't even know about the change. I happily went into the config file and configed my video card, monitors, screen and all just like I used to with XFree86. After saving I started X like normal and all ran just fine.

I wasn't until I was reading later on that I realized there was a different X on my machine. Even then I was getting confused because much of Slackwares online docs have not been updated to refect this change.

I like X. X is good. Some X'es are better!

Re:I'm using the new X.org

[Eunuchswear]

I wasn't until I was reading later on that I realized there was a different X on my machine.

The reason you didn't notice a difference is that there isn't one.

The X.Org monolithic X is just the XFree86 one from a microsecond before the license change.

More or less nothing has changed in XFree86 since the license change.

Re:I'm using the new X.org

[Brandybuck]

More or less nothing has changed in XFree86 since the license change.

Except that X.org starts up about ten times slower than XFree86 in exactly the same environment. Everything else seems just the same, just the startup is dog slow. I have no idea why.

Re:Nice Screeny's - with HTML

[pyros]

screenshot 1.

screenshot 2

Re:Does it run linux?

[Anonymous Coward]

"if you want to post something controversial, back it up witha real account. (why yes i know i am AC, but i am not posting anything that ANYONE will disagree with)"

Well post with your "real" account then, you fucking twat! Jesus, what the hell is it with people and Slashdot karma. IT IS NOT WORTH ANYTHING!!!!

I'm posting A/C 'cos I simply don't see the point of karma. Fuck you all, deep in the ass with an upturned pineapple.

take a hint from Quartz Extreme

[caveat]

and offload the work to the GPU. sure, you still need a semi-decent graphics card, but nothing screaming fast (i have a Radeon 9000 Pro in my PowerMac, it works just fine). yes, it would be nice if every app ever developed would run smoothly on a 50mhz 486, but c'mon...an 800mhz PIII isn't THAT expensive. see the other posts as for why you'd want compositing in the first place.

terminal interface

[HBI]

Web browsers cannot do this, and using the browser for applications better served by a terminal interface is why we've taken a great leap backward in usability.

I would love to see CCM for X

[richie123]

the only major problem with x.org I see at the moment is they are adding mostly eye-candy extensions, but things like screen and printer matching are the practical features missing in Windows that could attract a lot of Desktop Publishing and graphics apps to Linux. I think the composite extension is cool, but I would love to see more usefull stuff added.

Re:I would love to see CCM for X

[chtephan]

That's not an X issue.

I think Owen Taylor is planning to implement this in Pango. It's all but trivial.

yet again, Quartz did it first

[caveat]

it's not EXACTLY what you suggest, but it does use PostScript rather extensively...i really appreciate the "Print to PDF" feature, it's proven itself to be very very useful.

Not developers, artists!

[Anonymous Coward]

Well, the problem isn't so much that free software developers don't care about documentation. It's that technical writers don't get involved so much in free software as developers do. There are similar problems with artwork and music for free software games etc.

If you want good art, documentation, music, etc., then start complaining to your friends who do art on computers just for kicks. They could be helping a worthwhile cause!

MOD UP, FUNNIEST REPLY EVER.

[aussersterne]

Now.

Re:Does it run linux?

[DashEvil]

(why yes i know i am AC, but i am not posting anything that ANYONE will disagree with)

I'm sure (s)he would disagree with that.

More likely windows comparisons and GPU usage

[Anonymous Coward]

I think people are more likely to compare the wealth of slick windows-based apps that wow people with unexpected 3D effects at the drop of a hat. Transparency is interesting, but essentially useless, as most people say, but making 3D easy for APPLICATION developers will really change some things. It may even lead to new, unexpected forms of user interface. THAT's what Free Software has to compete with -- a shifting kind of application; not a shifting graphics capability.

On another note, I'd like to see GPUs used through a lower-level library than X. There are plenty of intensive computing tasks that can use GPUs, so it'd be nice to have X ask for the GPU, from a system that shares the GPU resources properly, rather than just hijack it.

Re:More likely windows comparisons and GPU usage

[spitzak]

Your idea of a lower-level GPU seems to be where things are going. It looks a lot like the final result will be a single OpenGL driver that can draw all over all a screen. X would be a seperate program that talks to the driver. Everything you do that changes the screen display is turned into OpenGL.

Some things need to be added to OpenGL still. Fonts and non-square clipping regions and a method to "draw this image from user memory using the current xform" that does not require creating a texture object. The most complex thing needed is some interface so that a connection to the OpenGL is limited to a certain surface, so that an X program can draw raw OpenGL without any chance of it being able to draw outside it's windows and with the same speed as though it had access to the actual driver.

If this is done right we might even see non-X experiments that can occupy the same screen as X does, with windows (or whatever) overlapping correctly.

Re:PRECISELY MOD UP+

[Anonymous Coward]

Actually, spelling it "i18n" resolves this issue

dps

[Anonymous Coward]

So get GNU to reopen development on Display [Post|Ghost]Script or Display PDF again.

Hope that compatability is retained

[Baki]

It would be very sad if changes in the X protocol or Xlib would make "new" clients unavailable on other X-window platforms and/or would no longer be network transparent.

In this cases it would no longer be possible to remotely work on a UNIX/linux server with windows X-emulators (such as exceed), nor would the typical linux open source app be able to run on other UNIX variants. Which would be very bad for UNIX as a whole and thus also for Linux which is a part of that world.

Re:Hope that compatability is retained

[Trelane]

This isn't a problem. All new functionality is done by extension to the old X11. I actually keep arguing that we should standardize an X12 which removes things not used much anymore and which includes (sans extensions) the functionality which is wanted, and provide and X11 interface for compatibility (if you have to make an extension extension because there are so many extensions, you need to move to the next version!). However, they don't want that; they want compatibility (which I don't hold to be orthogonal if things are done right, but I'm not (yet) and X hacker, merely an armchair X pundit. ;)

Anyhow, rest assured that compatibility is top priority with these changes. You just won't be able to see the shinies. :)

Re:Hope that compatability is retained

[jasontwarnock]

That's not a problem, X can not not be network transparent, X communicates via networking protocol, and if they changed the protocol it would still be networkable, and the other clients would just have to adapt slightly.

Re:Hope that compatability is retained

[Anonymous Coward]

I don't think that X.org can make that kind of move. The X Window System is owned by the OpenGroup. They are the ones that set the standards. XFree86 was an implementation of those standards.

What I can't figure out now is what links there are between x.org and the opengroup. X.org used to be the opengroup's website devoted to the X Window System standards and example implementation. It doesn't appear to be anymore but I can't verify.

NR

Re:Hope that compatability is retained

[Anonymous Coward]

Umm. By "can not not", do you mean "must"? I couldn't not fail to be slightly unconfused by that.

Re:Hope that compatability is retained

[eckman]

we should standardize an X12

Don't you mean X100? Everyone knows 100 comes after 11!

Re:Hope that compatability is retained

[jbolden]

The open group owns the standards to things like Posix and the trademark for Unix. They don't own X. X isn't owned by anyone but X.org is the official standards organization so that is exactly where something like X12 should be set. Xfree86 is a paying member of X.org and implements a standard. Freedesktop.org is not a member of X.org.

Re:hah

[Anonymous Coward]

No you're wrong, it's allready there.

Just use TWM, it is about as visually appealing as Win2000 and just as Win2000 it doesn't have virtual desktops.

Front Page + usability

[Anonymous Coward]

Ars Technica is developed with MS FrontPage! argh
<meta name="GENERATOR" content="Microsoft FrontPage 5.0">

Does this explain the ubber-ugly unusable web site they have?
Every time I want to read an Ars Technica's article I copy it to a text editor and read it there.
Why Ars persist with no alternative style-sheets? DO YOU COPY ME ARS???

Err, no, NeXT did it first...

[Abcd1234]

... with DPS. Of course, the fact that the Quartz guys incorporated this technology isn't a surprise, given that much of OSX inherits from the NeXT legacy. Incidentally, similar work was also done by Sun in the form of the NeWS (which is considered by some to be a superior technology).

Re:Does it run linux?

[Desval]

I never thought of HTML as programming.

Unless you are doing just basic CGI GET/POST, the underlining logic of the forms (no matter how simple) still has to be implemented in some other fashion (JavaScript, VBScript, etc.). Consistent support for these across multiple platforms or even multiple browsers/servers is spotty at best (key word is consistent).

HTML makes a decent markup tool for cross platform documentation (its original intent), but a poor method of application development.

It has aways struck me as degrading application functionality back to the level of the IBM 3270 series terminals (which supported form based screens), but not doing as good of a job.

Sorry for the rant, but it drives me crazy when dealing with the limitations of web applications when asked to convert some of our native clients to it so that the sales department can use the latest marketing buzz words.

A Modest Proposal

[Tony]

See, I think we should move X entirely onto the GPU. Why screw around with X on the primary CPU? Make it all run off the video card. That'd make it *real* fast.

NOTE: This is satire. Not particularly great satire, but satire nonetheless.

Re:Does it run linux?

[pclminion]

Actually, the web is more of a zero-client, not just thin. The browser is a given, so it literally takes no additional software at all on the client side to implement quite a lot of functionality.

Face it, thin- and zero-client architectures are going to become more and more popular for everything, because it leaves much more control in the hands of the service provider, which in turn allows more flexibility in business models.

And I don't think the web is a bad medium for those architectures at all.

Resolution limit

[cmaxx]

Is anyone fixing the fundamental resolution limit in the protocol?

Re:Resolution limit

[spitzak]

Are you talking about the 16-bit numbers used in some of the interface?

For window dimensions this is not likely to ever be a problem.

However it is a big problem for graphics. In effect the only way to reliably draw an arbitrary graphic is to do intersection testing and clipping in memory against +/-32767 planes before sending the lines. This is very difficult and expensive (and silly since X has to then do further clipping aganst the actual window area anyway). So most programs don't do this, resulting in graphics screwups when you zoom in or scroll your graphic sufficiently. Win32 also has the exact same problem, incidentally.

The solution appears to be that the entire X drawing code is being replaced with Cairo or OpenGL, or something like it, which accepts floating-point coordinates. The old code will remain only for back-compatability and will probably never be fixed, but it does not really matter.

Re:Does it run linux?

[HawkingMattress]

You must be kidding ! Obviously you never had to write a real dynamic webapp or you'd realize it's something like 1000x harder than writing a vb app. Of course, static hmtl is easy, but that's not the problem at all...

Just an example i had to deal with today : had to make a combo box where each line is made of 3 aligned columns from a database. The only way to do that in html if you want the columns to be aligned is to pad each columns with nbsp; codes. (you're not going to make a table in a select statement of course..)
Result : For 7400 records in the database, the generated HTML code for the select control takes 2.7 Megs. Nice, still don't know how i'm going to handle that... On the other hand, the old access/vb app the client is using now just uses a vb combo control which automatically binds to the db, probably fetches and manipulate the data in binary form and loads in microseconds.
It probably took the developper about 5 minutes to make this, I struggled all day to find a way to make things better, and i'll probably have to make an applet tomorrow to handle that. Still no idea how many time it'll take...

In fact in a lot of case it would be easier to have to code a custom server and client than to write a webapp, because http was *not* designed to do that, it was designed to allow you to fetch files from a server and that's about it.

Erm...

[Anonymous Coward]

XFree 86?
X.Org 6.8?

Something about backwards compatibility?

Older, slower machines?

[Platinum Dragon]

Potentially dumb question:

What will the new features and code in X.org's next release mean for people still futzing around on sub-Pentium III machines? I'm running FC2 on a PII-266 with 224 MB of RAM, with an ancient ATI All-in-Wonder (not pro, not 128, not Radeon--just AIW!). It runs reasonably well with a fairly glitzy GNOME 2 theme, considering that I have a webserver and nameserver running as well. If and when the next release is included in a future FC milestone, should I expect things to run slower, faster, or about the same as now?

Poorly designed/ugly website

[Burz]

It figures.

Re:Does it run linux?

[KoolyM]

Give CSS a try.

The Misinformation Campaign Rolls Along...

[Brandybuck]

I see the misinformation campaign continues. Let me quote from the site itself: "None of this is "endorsed" by anyone or implied to be standard software, remember that freedesktop.org is a collaboration forum, so anyone is encouraged to host stuff here if it's on-topic".

In other words, this is merely a hosting site. Except for its smaller size and narrower focus, it's not any different from freshmeat or savannah. The software at fd.o is not official and not required for desktops to use. They are not standards which must be adopted for "freedesktop compliance". While there have been a few people clamouring to make X.org the "official" X11 of GNOME and KDE, frankly it's not going to happen.

To take a specific assertion, "While freedesktop.org is, in many ways, a fairly loosely organized community project, we're all really minions of Havoc": this is completely wrong. Havoc Pennington is an employee of Redhat and a GNOME developer. He has no authority over X.org, KDE, XFCE or any other project outside of Redhat and GNOME. He has extremely little input to any of the projects hosted at fd.o.

Freedesktop.org started out as a great idea, and for a while it was very useful. But then it got infected with politics, and quickly lost most of their relevancy. If not for their recent hosting of independent software projects, they would have faded into obsurity. Frankly, would anyone care about fd.o if X.org wasn't hosted there?

Re:The Misinformation Campaign Rolls Along...

[Anonymous Coward]

I would.

Cairo. DBUS. HAL. the .desktop standards.

All of these are part of fd.o

YOU are spreading the misinformation, troll.

Re:The Misinformation Campaign Rolls Along...

[Brandybuck]

Cairo. DBUS. HAL. the .desktop standards.

Excepting the .desktop format, none of them are fd.o standards, and even the .desktop format is still a "draft". Havoc's unilateral pronouncements that something is a "standard" means nothing more than he wishes it was. Most of the real de facto standards (like .desktop) were created by the "little guys" of KDE and GNOME working together without the benefit of Havoc's blessings.

In all Open Source projects, the people who get things done and the people who strut about crowing are two separate groups with very little intersection. Freedesktop.org is no different.

X.org in Mac OS X

[JoseFilipe]

Anyone knows if Apple will be switching to X.org? Is there any reason for doing so?

Re:X.org in Mac OS X

[charnov]

Not a chance in hell...they developed their own next gen graphics layer and window manager(Quartz, Quartz Extreme, and Aqua). Personally, I would kill for linux to switch to Apples way of thinking for a GUI, but it'll never happen. I got into an argument years and years ago with Torvald (along with tons of other people) about including GGI into the kernal and got my ass kicked. Never been with linux since.

Re:X.org in Mac OS X

[charnov]

Haven't been...wow...pardon my grammar. ;-)

Re:X.org in Mac OS X

[Anonymous Coward]

The OP isn't talking about replacing the OS X desktop with X.org (or at least, I hope not). But OS X does include an X server (based on Xfree?), and that's what the OP would have been referring to replacing.

Re:Does it run linux?

[Brandybuck]

Corporations are moving towards all web-apps. In my company I have seen one rolled out to replace a desktop app approximately once a month for the past year.

Unfortunately these webapps all require Windows Internet Explorer running on a Windows NT/XP desktop. It's not about thin clients, it's about being able to manage a multinational corporation from a single IT building in New Jersey.

QuickDraw

[ZigMonty]

Err... hasn't the classic mac os always been like this? Didn't you use QuickDraw both for drawing to the screen and to the printer?

Re:X in Windows? otaku

[Anonymous Coward]

child molestor boy raping slaughtering murderer! why do you keep following me online. everyone i talk to and everywhere i go, there you are, jusdisgi, you know i know who you are and i cant live with how you raped me and gave me HIV, please, stop stalking me you god forsaken raping molestor! let me die of AIDS in peace.

The motivation of open source developers

[Snaapy]

While I love open source, sometimes the fact that it is done for nothing is one of the things that ensures it is developed slowly. Unless you are a full time student, most people are working a day job to put food on the table. Without the cash motivation it is not always easy to spend the time and effort necessary to make a great project. I am not saying the money is what is important to them, though being comfortable, being able to buy a workstation and not living on the street is.

Or not. Some of open source folks are pretty idealistic and income isn't so important thing for them. (Naturally you must have some income to pay your living expenses.)

There is quite good survey trying to find out open source developer motivations. It's made by Boston Consulting Group (big and famous guys in management consulting).

A slide that presents the motivations of paid and non-paid open source developrs.

Obviously, the best case would be that your paid for what you want to do (develop open source).

LANG=pt_BR slashdot-spell

[hummassa]

yes I do: P-O-S-T-E. Poste. Pronounced "pos-ch" (open "o", like in POD).

Re:LANG=pt_BR slashdot-spell

[Smallpond]

muito engraçado (LOL)

Re:Does it run linux?

[freqres]

Why wouldn't those apps work with a thin client? I've use both schematic capture and CAD/CAM in the past with X terminals. Why wouldn't a thin client machine with a nice big screen and a good 3d card work that did something similar to X or MS Terminal Services. The reason a web browser is lousy for these things is because it is stateless, a very limited 'widget' set and the client side scripting is very limited. You can propose to use something like java which gives you a lot more flexibility, but then why not just skip the browser and use something like Java Web Start and not have all the browser interface cruft in the way.

Re:PRECISELY MOD UP+

[Anonymous Coward]

Hey, thanks to your post, I now know why i18n is called i18n. i, 18 letters in between, n.
No, I haven't searched about it before. It's just one of those little things I've wondered but hadn't cared enough to remember to look up. ;)

My own pointless DOS shell gripe

[MegaFur]

The thing that's always pissed me off about that ol' orginal DOS COMMAND.COM is the use of `/' as a switch character. Why, oh why did he/they do this? Why!? Just imagine how much more happy the Universe could be if they'd used `-' for switches and, thusly, later uesd `/' for directories. The world would be a better place.

Re:My own pointless DOS shell gripe

[Anonymous Coward]

At first I was thinking that the DOS ancestor CP/M was the reason for the '/' option character. But this article clears the matter up. It seems M$ used the '/' in some of its early programming tools.

Re:X in Windows? otaku

[jusdisgi]

Awesome!

It's nice to know that I'm so important around here. After all, it's not every /.er that has his own pet troll!

X itself has perfect cut-and-paste support

[Anonymous Coward]

• something as elementary as copy-paste is still a hit-and-miss affair. [...] it's a sign that X sucks

For the record, your criticism is invalid because there is actually excellent support for cut-copy-and-paste in the design of X. What you could have criticised is that there are, unfortunately, many applications, which are written for X but which are not part of X itself, that do not implement cut-copy-and-paste correctly. Criticising the design of X itself because the applications for X have inconsistent and/or broken cut-copy-and-paste facilities is illogical.

The same flawed criticism of the design of X is frequently raised on slashdot. Please read the following very informative thread ignoring all the comments by Minna Kirai.