X11 Window System Turns 25 Years Old

An anonymous reader writes "The widely used X11 Window System has turned 25 years old today. Version 11 of the X Window System is likely to remain in use for many years to come for backwards compatibility with the many legacy applications, BSD/Solaris systems, and Enterprise Linux distributions. Meanwhile, Wayland is still working to unseat the X Server for the common Linux desktop."

Comments

X12?

[Greg Merchan]

Potential protocol changes were noted in the documents of version 11, such as the ICCCM which notes that the FocusIn event should have a timestamp and a previous-focus field. Has anyone out there considered just taking X11, making changes known to be needed, and dropping the protocol support for what's rarely or not needed anymore?

Re:X12?

[Greg Merchan]

Yeah, yeah, RTFA. Sorry for existing before coffee.

Re:X12?

[Paul Jakma]

People have considered that, and they're working on Wayland as a result.

Re:X12?

[Anonymous Coward]

and unfortunately, they're also dropping things that ARE needed.

Re:X12?

[Randle_Revar]

Too bad Wayland is terrible.

Re:X12?

[Paul Jakma]

Remote display protocols, like X11, VNC, etc., will still be able to render to Wayland displays.

Re:X12?

[jedidiah]

...and will be as useless as using VNC with a Mac.

Before you try to clone something, you should actually use it yourself first. Same goes for goading other people to clone something.

Re:X12?

[markdavis]

+1

There is nothing "wrong" with X11 that upgrading the protocol won't help. Wayland is an abomination that I hope fails because it is trying to solve problems by creating a whole bunch of new ones.

Re:X12?

[Paul Jakma]

The people working on Wayland have used X11. Indeed, in many cases they are *also* X.org developers. Hell, one of the people working on Wayland is Keith Packard[1], who's been working on X.org since longer than I've been using Unix. Indeed, he's been working on X11 since before many of us had even used a computer, indeed for anyone younger than 24 make that "since before you were born". Hence, to say the people who are working on Wayland do not understand X is just a ridiculous argument, and does not suggest the person making that argument has much credibility on the subject.

I'll be honest, I was a little sceptical when I read about some of the design decisions in Wayland. In particular, the decision to move some of the window management to the application (in general, that means the toolkit, like Qt, GTK+, etc) makes me wince a bit, because it will lead to the hung-window-syndrome we know and love from MS Windows. However, the people involved in Wayland know far far far more about the subject than I do (I have no experience of designing or implementing windowing systems), and I'm sure they know a lot more about balancing the various trade-offs for and against all these decisions than most of us.

As for the remote displays. I was initially concerned about that capability too. However, if you look into it, well there's nothing that stops X11 being used with Wayland - indeed X server to render to Wayland already exists. Generally, there's nothing to prevent whatever style of remote display protocol being implemented for Wayland, be that in applications directly or (more sensibly) in the toolkits.

1. Keith Packard on Wayland and X: https://lwn.net/Articles/491509/

Re:X12?

[Anonymous Coward]

indeed X server to render to Wayland already exists

In the same sense that an X server to render to Windows desktops exists, yes, but that does NOT make it a replacement for the functionality in X11 today.

Re:X12?

[Anonymous Coward]

GNOME is far more experienced in developing desktop environments than I, but I'm not going to trust them to not make near-unusable, change-for-its-own-sake design decisions. I'm not taking either side, just pointing out an appeal to authority.

Re:X12?

[0123456]

Remote display protocols, like X11, VNC, etc., will still be able to render to Wayland displays.

And Wayland will not be able to display efficiently to any other machine. It will require some shitty pixel-scraping technology like VNC.

As the world becomes more networked, the Wayland fanboys are trying to copy Windows by throwing away X11's separation of display and execution.

Re:X12?

[dslbrian]

I'll be honest, I was a little sceptical when I read about some of the design decisions in Wayland. In particular, the decision to move some of the window management to the application (in general, that means the toolkit, like Qt, GTK+, etc) makes me wince a bit, because it will lead to the hung-window-syndrome we know and love from MS Windows.

It causes more than that. This is a good read on the problems caused by CSD.

Re:X12?

[Paul Jakma]

Yes, it is an appeal to authority. However, if you're interested, you can easily find talks by those authorities explaining those trade-offs. E.g. the link I gave contains a long report on a talk by Keith Packard.

Re:X12?

[Paul Jakma]

And Wayland will not be able to display efficiently to any other machine. It will require some shitty pixel-scraping technology like VNC.

No, that isn't true. Nothing prevents Wayland supporting toolkits from using whatever kind of display protocol they want, from low-level pixel grabs of windows, to using more abstract drawing commands that match the higher-level structure/behaviour of the application. Indeed, a toolkit could easily use X11 as that protocol, if it wished - and that's already possible with XWayland, I understand.

Re:X12?

[Anonymous Coward]

Right, so instead of doing it cleanly in a single place, we're supposed to do it in N different places in likely incompatible ways.

Good plan, there.

Re:X12?

[Greg Merchan]

It's not that ridiculous. There's more than one aspect of X to understand. I know next to nothing about drivers or anything else internal to X, but I have, I believe, a decent grasp of the things above those layers which are documented. Unless the documentation is lying, there are things on X that have, especially for the sake of modern UIs, been done incorrectly for years. Focus handling is the main thing I keep harping on. Just about everything should be using the globally active focus model by default, but most things don't. Motif and other Intrinsics-based toolkits allow it. Gtk+ does not. I don't know if Qt allows it, but it didn't do it by default last I checked. Too many people "just know" that's not the way you do things, so they keep doing them the way they've been doing them and then can't get things like drag-and-drop to work properly without a bunch of ugly hacks. There's too much knowledge of what is, getting in the way of learning what could be. My fear for an X12 would be that, like Wayland (or maybe just Weston, can't tell yet) developers almost did, they would lock-out getting things done the right way.

The X selection mechanism, for another example, is something that could be used to greater effect. But most people think of it as just the PRIMARY selection, the clipboard, and part of drag-and-drop, if that much. Fewer people recognize it's part of window management and the message tray protocol (or whatever it was called). It has more potential, but again you get people who know very well what X is, but don't recognize what it could be.

Re:X12?

[hairyfeet]

It all comes down to the market, and Ubuntu is for HOME users, remember the "Linux for humans" advertising? So how many home users are remoting into their desktops using X Server? My guess is you could count that number on 2 hands and still have fingers left over, remote control of desktops is just something that most home users don't even think about, much less do.

And the simple fact is for non enterprise uses X is fricking terrible, just piss awful. I could post Thom's epic rant from OS News how simply switching from a video to his chat causes the whole damned desktop to crash, or how Mozilla won't support hardware acceleration on Linux because of X being buggy and crash prone, by why bother? This isn't exactly a secret and daring to go against group think just gets modded down anyway.

If you want more mainstream users you are gonna have to give them a nicer experience and X is about as nice as a kick to the groin. Sure it works great in the enterprise but 1.-The enterprise isn't running full GUIs, watching videos, popping up chat windows and all the other things that home users do and 2.-Because enterprise pays people to work on the problems they encounter enterprise bugs are usually fixed pretty quickly. Again this isn't the case for home users.

So Canonical sees a problem and is trying to route around the damage.,..isn't that how things are supposed to work in FOSS land? you can always fork away if you don't agree with a direction?

Re:X12?

[Anonymous Coward]

People have considered that, and they're working on Wayland as a result.

and unfortunately, they're also dropping things that ARE needed.

including Unix:

"Wayland is a display server protocol and a library for Linux implementing that protocol."

So, no wayland on *BSD, HP-UX, AIX, True64, ...? Or is that just a Wikipedia brain fart?

Re:X12?

[Anonymous Coward]

How does CSD affect the ability to have sandboxed GUI apps?

Re:X12?

[Paul Jakma]

We already have N incompatible ways today - X11, VNC, RDP, NX, etc. All Wayland intends to do is to move *1* of them from the display server out to the toolkits. That's not going to have much visible effect to the user, except perhaps the smoother rendering of local apps, because the local path can have a simpler architecture now

Re:X12?

[laffer1]

My initial problem with wayland was the network support. I've realized I can live with using VNC or some other monster instead. The real issue with wayland is that it's Linux only. Remember when Linux was a clone of UNIX? Now it's trying to drag us all every which way and reinvent everything. This is a real pain for BSD developers and probably lousy for Oracle, IBM and HP too. The latter can "afford" it... but still.

Re:X12?

[Anonymous Coward]

"Those who don't understand Unix are condemned to reinvent it, poorly."

They usually go through a phase where they get their shitholes torn open by megapricks like Bill Gates and Steve Jobs before they discover the truth. Some don't survive.

MS and Apple users are all cunts. I wish they would all stop being cunts.

Re:X12?

[Anonymous Coward]

Wikipedia. I've corrected shit like that a few times but eventually I gave up (and my IP was blocked).

Re:X12?

[jbolden]

Most likely BSD will get a port of Wayland long before Gnome and KDE migrate to Wayland only.

If you mean Linux is now setting standards for the Unix GUIs it has been for over a decade.

Re:X12?

[jbolden]

There are 3 situations.

A) Local desktop
B) High speed low latency Network.
C) Low speed or high latency Network.

X11 is optimized for case (B). Given we can't do much about the speed of light, relative to the size of the earth (C) type problems are permanently unsolvable for X11. Moreover, given that networks are focusing on reducing jitter while increasing bandwidth to implement QoS, we might see latency increases over the next generation or at least not large decreases.

X11 is not a solution to networked desktops.

Re:X12?

[jbolden]

And Wayland will not be able to display efficiently to any other machine. It will require some shitty pixel-scraping technology like VNC.

Actually Wayland is going to implement something like RDB but that will be at the GUI level (i.e.KDE or Gnome).

Re:X12?

[jbolden]

Really? How is X11 going to solve the problem with high latency connections? What upgrade can work with X11's model?

Re:X12?

[markdavis]

Here is your answer: http://www.nomachine.com/ http://en.wikipedia.org/wiki/NoMachine http://freenx.berlios.de/

And it is still X11, and it really works. So yes, it can be done.

Re:X12?

[Anonymous Coward]

Linux is terrible. There, now we're both in agreement. /sarc

Yep, engineers with PHDs see a problem that your feeble little mind can't comprehend. Actually, anyone who understand how computers work can see Wayland is superior in almost every way, except some minor corner cases.

Re:X12?

[dbIII]

Don't blame X for faddish behaviour breaking the gimp tool kit for the twentieth time since it was taken out of the hands of the gimp developers by a bunch of people playing politics.

Re:X12?

[Greg Merchan]

I didn't blame X for anything. I actually like Gtk+ and have longed wished this problem would be solved. (You see my name. You can dig up a bug report.) But I haven't been able to convince the right people to see focus handling (and other things) the way I see them. Unfortunately, "the right people" means not only Gtk+ developers, but also Qt developers and at least a few window manager developers. I'm working on what I hope is a more convincing argument, but this is really not my day job.

I also have this nagging feeling that compositing window managers may have changed the scene so much that my understanding no longer applies. But if that were so, I would trade a compositing window manager for a properly working non-compositing window manager in a cycle.

Oh, but gimp itself is probably part of the problem. The gimp way of handling windows was snafu last I checked.

Re:X12?

[Anonymous Coward]

Standards are something others can follow, not reinventing everything every two months to "preserve incompatibility"

Re:X12?

[Anonymous Coward]

Linux needs to stop copying Windows and to a lesser extent OS X.
If you want binary logs, incompatible file formats, non-remotable apps, arbitrary undocumented design, and unexplained crashes due to frankensteinism, just use Windows and leave UNIX alone.

Thanks

Re:X12?

[Anonymous Coward]

Just because it doesn't meet your usecase, it's not a solution. Got it.
Do you work for Canonical? Or GNOME? Maybe redhat?

Re:X12?

[Anonymous Coward]

To be accurate, RDP is essentially a shitty pixel scraping technology (It displays everything, not just a particular app like X).

Re:X12?

[Paul Jakma]

The toolkits applications use (e.g. GTK+, Qt, Wx, etc) will continue to support rendering to X11. There is a rootless X server for Wayland, which will continue to exist.

In short, X11 will continue to work just fine with Wayland. Just think of Wayland as a better architecture for the X server.

Re:X12?

[serviscope_minor]

The X selection mechanism, for another example, is something that could be used to greater effect. But most people think of it as just the PRIMARY selection, the clipboard, and part of drag-and-drop, if that much. Fewer people recognize it's part of window management and the message tray protocol (or whatever it was called). It has more potential, but again you get people who know very well what X is, but don't recognize what it could be.

Could you elaborate? I thought I knew the selectoion mechanism and Xdnd pretty well...

Re:X12?

[serviscope_minor]

I used to respect Keith Packard. Much of what I've read fro mhim about Wayland versus X11 has been very FUD filled. It seems like he's simply bored of X11 hacking to me.

Re:X12?

[serviscope_minor]

All Wayland intends to do is to move *1* of them from the display server out to the toolkits.

No, it moves N of them (one for each toolkit) and only for the toolkits which support it. The result is a massive mess.

Re:X12?

[Paul Jakma]

GNOME and KDE are not going to migrate to Wayland only. They're going to continue to use toolkit libraries, and the toolkit libraries will handle the details of rendering (or indeed, other libraries underneath the toolkit library will).

E.g. KDE does not contain an X11 protocol implementation, it isn't even written to directly use Xlib or XCB. It's written on top of Qt. And Qt already supports rendering to Wayland via lighthoutse. Ditto for GNOME and GTK+/GDK.

These libraries already support multiple rendering backends, from rendering to dumb framebuffers (for use on embedded devices) to rendering to X11 across a network. These libraries are simply going to acquire another such rendering backend, for Wayland. The existing rendering backends, like X11, are not going to suddenly disappear. Not any time soon. Not any time where X11 sees any kind of significant use or vendor support.

Re:X12?

[Anzhr]

Thanks. I had forgotten about Packard's involvement and appreciate the link.

Re:X12?

[Anonymous Coward]

This is supposed to be progress?

Today, with X11, I have a single built-in system that provides remote display. It just works. It works the same for every single program, regardless of the toolkit it's based on. It works for GTK programs written yesterday and Athena programs written 20 years ago -- and yes, I still rely on some of those, for niche tasks.

Is it perfect? Perhaps not. But it's simple (for me, the end user), and it's consistent.

If Wayland is going to replace that with "fuck you, end user, you better pray every single application you use decides to implement its own slightly different solution or something", then I'm struggling to see how it is supposed to benefit me, the end user.

If I wanted OS X, I'd buy a Mac.

Re:X12?

[Rich0]

If an app is untrusted typically something will be done to the window directions to indicate this. That way when some java applet pops up a dialog box that looks like the OS standard dialog box it has an ugly border that says "DANGER" all over it.

However, if the client controls the window decorations, then the OS has no control over how its windows appear, and the client can spoof a "trusted" dialog box.

Re:X12?

[Rich0]

We already have N incompatible ways today - X11, VNC, RDP, NX, etc. All Wayland intends to do is to move *1* of them from the display server out to the toolkits. That's not going to have much visible effect to the user, except perhaps the smoother rendering of local apps, because the local path can have a simpler architecture now

Yes, but right now EVERYTHING supports X11, and almost nothing supports any of the others except for bridging software that basically converts from X11 to those protocols. So, the *1* thing that you propose to move is the *1* thing that has universal support. Toolkits aren't going to implement direct support for any of the others (or if they do they'll pick 1-2 and you'll end up with apps that can run on RDP and apps that run on NX and so on). So, if you get rid of X11 then you're stuck with VNC screen-scraping, since that is the least common denominator, and it is pretty lousy.

Re:X12?

[unixisc]

Looks like that's what Wayland sets aside to the periphery - remote connecting to desktops. In fact, Wayland gets rid of it, and allows an X11 subsystem to use its services, so that if one has to, one can use that X11 to remote connect to their desktop. And if one doesn't use it, Wayland should work great.

Re:X12?

[Anonymous Coward]

What is pixel scrapping?

Re:X12?

[unixisc]

I can't see anyone developing Wayland for Tru64 - the latter is dead. HP-UX is an Itanic only Unix - why would they be interested in Wayland - even HP dropped Itanic workstations long ago. As far as the BSDs go, none of them currently have any plans to support Wayland. So for now, it is Linux only. Later, once Qt and GTK+ support Wayland, it would make sense for PC-BSD (but not others) to support Wayland.

Re:X12?

[Agronomist Cowherd]

Not true. RDP has support for hosting single applications from multiple servers on your screen. No one uses it that way except for enterprise stuff like Citrix, but it's part of the protocol.

And today, X remoting is essentially "shitty pixel scraping", because all the frameworks don't use X server rendering, they draw to local buffers and hand that to X (that's how you get pretty font support, for example). So X is schlepping pixmaps around, and if you're going to do that, why not just get rid of the rest of the protocol cruft: hence Wayland.

Re:X12?

[Paul Jakma]

No one, AFAIK, is getting rid of the existing X11 support in Xaw, Tk, Motif, GDK, or any of the other X11-supporting toolkits applications are written in as part of adding Wayland support (if Wayland support is added at all, e.g. it's unlikely for Xaw and Motif ;) ). Wayland will just be another rendering target backend in these toolkits.

No one is getting rid of X11... Calm down, relax.

Re:X12?

[jbolden]

Thanks for the tip. I like the theory and I'd agree that in theory such an approach might work to get (c) to an acceptable speed. But the actual experimental results are:

a) NX is a lot faster than X
b) NS is about on par with VNC

And VNC is still rather slow.

So at least so far it doesn't appear to solve the problem of high latency or low bandwidth.

Re:X12?

[markdavis]

Well, the technology difference between X/NX/NC and VNC are huge. One is compressed widgets, the other is bitmap copying. I would think that X/NX/NC would all be much faster than VNC, but I have not compared them all to each other.

Are you saying you tested NX and don't think it solves high latency issues? It should.... I know several people who think it is fantastic. I suppose it also depends on just how high the latency/low the bandwidth is.

Re:X12?

[jbolden]

I suspect that QT or KDE will move quickly to Wayland.

You can do things when the video system and the application buffer share RAM (i.e. very high bandwidth, 0 or near 0 latency connection) that you can't when they don't. KDE and Gnome are going to want to do those things. It is entirely possible that QT will offer those features and than offer a way to simulate a unified buffer with 2 buffers communicating but in practice such a system could be very flaky.

Ultimately the Unix GUIs will have to choose what scenario to optimize for. With X11 being so dominant they had no choice but to go with X11's choice: LAN style bandwidth and latency. What happens on a LAN is much much slower than what happens inside your computer's RAM.

Re:X12?

[jbolden]

I'm just reading the test results: http://vis.lbl.gov/Events/SC08/RemoteX/index.html

NX does seem to offer some improvement. That there is room for more, for example HP's SSH is much faster than Sun's (the open source one commonly used), which could help a lot as well. NX definitely seems like an interesting approach for getting 2-10x as much effective bandwidth. It also claims to reduce some of the latency problems. It doesn't appear to be faster than VNC yet, but it would be nice if NX created a fast network transparent system.

I mentioned 3 cases:

a) Running an application on a machine sharing ram with the video card.
b) Running applications on a machine close enough to the video card that the latency between them is lowish and the bandwidth is plentiful and performance is irrelevant.
c) Running applications on a machine where either the latency is high or the bandwidth is limited

If because of NX (c) became acceptable then we are really down to a much deeper choice as to whether building for (a) as the primary use case more sense than building for (b) & (c). The big issue with X11 right now is that it doesn't handle (c) well. NX seems to already make (c) better but not good enough.

Re:X12?

[Paul Jakma]

I suspect that QT or KDE will move quickly to Wayland.

Qt *already* supports a number of rendering backends, including X11, including *Wayland*. So when you say "move", with the connotation that X11 support must be dropped in order to add Wayland support, then existing reality has already disproven your argument.

Re:X12?

[grumbel]

And Wayland will not be able to display efficiently to any other machine. It will require some shitty pixel-scraping technology like VNC.

That "shitty pixel-scraping technology" is how all apps are written today. Nobody draws his vectors with X11 any more, they all use Cairo, Qt or whatever to get pretty anti-aliased lines and subpixel rendered text not supported by X11. It might not be the nicest approach in the world to do it that way, but that doesn't change the fact that it's the way things are today. X11 has been obsolete for a while and is essentially just dragged around so you can run some old Motif applications.

Also the oh so awesome network transparency of X11 is actually quite shit, as it doesn't allow basic things such as moving applications between displays or screen sharing and all that other cool stuff that is completely trivial with VNC.

Re:X12?

[grumbel]

That doesn't sound like a new probelm, as X11 has allowed borderless windows for ages.

Re:X12?

[allo]

move will mean something like new effects, which work only on wayland or something like this.

Re:X12?

[jbolden]

I understand that, I said that. Right now QT offers support for Wayland the way it does for any foreign interface. That's very different than QT/KDE making use of Wayland only or Wayland primary features that either don't work or don't work well under X11. We'll know whether that happens around 2017 or so.

Re:X12?

[jedidiah]

In other words you are advocating the MS-DOS mentality.

It doesn't have to be robust because "it's only for home use".

Unfortunately that rationale doesn't work here because all of the best video drivers are not going to be supporting Wayland. Beyond the whole "we actually do real work with Linux" sort of people, you also have a whole bunch of other people that actually like to get the most out of their graphics cards.

Wayland will do exactly squat to address that as the companies that make the best hardware and provide the best drivers have no interest in Wayland.

Right now Wayland is a lot of hype and nonsense. Nothing has actually materialized. It has proved nothing. It's really only rallying point for mindless X haters repeating other people's propaganda.

Wayland even fails for the "but it's only for home use" argument. It actually fails harder in that regard.

Replicating the usefulness of my Nvida blob? Good luck.

Re:X12?

[Greg Merchan]

Window managers are supposed to claim a selection for each screen they manage named WM_Sn, where n is the number of the screen. All this is really used for right now is to let one window manager take over from another one. The selection is one of a special kind of selection known as manager selections. Basically it makes sure that some shared resource is only controlled by one program at a time.

The freedesktop.org System Tray Protocol uses a manager selection so applications can find the "system tray" to put icons in it. I don't know if that spec is still used.

The principle idea that hasn't seen much use is side effect targets. Normally selections are used for moving data around, but the ICCCM specifies more than that. Requesting a ConvertSelection on a side effect target is basically a remote procedure call.

Suppose you have something that allows operations on groups of windows, like a task bar. One thing I do on Window that I can't recall doing on X is asking for a group of windows to be tiled or cascaded. The EWMH spec, so far as I can recall, doesn't provide a way to do this or even an obvious path for extension to do so. The request to cascade a list of windows would have to somehow include all of those windows, and can't be broken up into a series of ClientMessages without using something to specify that a sequence of ClientMessages has begun and ended. A side effect target is ideal for this. The task bar calls ConvertSelection on WM_Sn with a target of CASCADE and a property which contains the list of windows to be cascaded. The window manager then cascades the windows listed and lets the task bar client know if the operation succeeded or failed.

Re:X12?

[Anonymous Coward]

Wayland is our generations GNU Hurd chances are that it will never actually reach "production".

The OpenSource community needs those lightning rod projects to keep the "Not Invented Here" crowd occupied enough that they don't pollute existing products too much.
The only real problem here is that the tech press tendency to go ohh shiny new awesome leads them to build up way to much hype and declare older more functional products for dead prematurely.

Re:X12?

[highphilosopher]

Can you say "indeed" less often in your posts? It makes you sound either Ye Olde English, or like we aren't believing you as you typed it.

Re:X12?

[serviscope_minor]

Wow thanks!

I thought I knew it pretty well (I have implemented Copy/Paste/Xdnd in Xlib before), but I never even thought of all the possibilities.

This is what makes me sad about Wayland. X11 is far from perfect, but it seems that people are so desparate for shiny that they're prepared to discard all the truly inspired parts of X11 too.

Re:X12?

[Rich0]

Yup, and 5 years ago nobody was proposing requiring a specific sysvinit implementation in order to run Gnome.

I'm sure X11 will be left perfectly intact while Wayland is still gaining acceptance. The question is what happens when X11 only has 2% market share? Keep in mind that the Slashdot crowed is basically that 2%...

Re:X12?

[Rich0]

Yup. And things inside sandboxes probably won't be allowed to make direct Wayland calls anyway, so I tend to agree that this particular issue seems to be a non-issue.

Re:X12?

[Paul Jakma]

Well, it's a very different set of people working on Wayland, from the one doing the kitchen-sink init. The ones working on Wayland seem to have thought about the compatibility issues.

Re:X12?

[diego.viola]

You mean Qt.

QT is QuickTime.

What is Wayland?

[Anonymous Coward]

Posting AC cuz I'm not a karma whore

https://wiki.ubuntu.com/Wayland

Re:What is Wayland?

[rubycodez]

Canonical being behind it may cause lag in adoption, since people are fleeing Canonical's UI ideas like creatures from a forest fire, even if they stay within the Ubuntu family it's Kubuntu or Xubuntu or Lubuntu....Unity and GNOME3 are inferior ivory tower designed, user-need ignoring crap

Re:What is Wayland?

[Paul Jakma]

Wayland isn't a Canonical thing. There's a bunch of people from various companies working on it, including Intel and RedHat.

Re:What is Wayland?

[Anonymous Coward]

How come Slashdot moderators always give points for these shills pointing out that Gnome 3 is ignoring users? They are NOT. They are just ignoring elitistic engineerish power users that Slashdot is full of, and catering to ordinary users. Mods, you are full of shit really.

Re:What is Wayland?

[westlake]

Canonical being behind it may cause lag in adoption, since people are fleeing Canonical's UI ideas like creatures from a forest fire

I am tempted to say that the geek takes flight whenever a UI is designed for other users --- a far greater user base.

Re:What is Wayland?

[Anonymous Coward]

GNOME3 may suck, but its innerbits are lovely under Cinnamon... try Mint yet? And do you remember the old assumption that things like UI would always be done badly by big companies? And how that's totally still true? (MS: two steps forward, one back. Apple: recommended for ages 6-12).

The linux community does how it always has, let the big groups do the heavy lifting and the small groups put the buttons in the right place. The big mistake was shifting major distros onto the mainstream.

Re:What is Wayland?

[0123456]

Hint: who do you think made up most of Gnome's user base?

Re:What is Wayland?

[Anonymous Coward]

"People" or just a very vocal minority on Slashdot?

Re:What is Wayland?

[Anonymous Coward]

Gnome 3 and all the Gnome 3 lover fanbois eat shit.

Re:What is Wayland?

[Anonymous Coward]

Another hint: who pays these "shills"???? Govt? Microsoft? Who? MATE Project?

Let me know! 'Cause I too hate Gnome 3 and a few extra bucks wouldn't make no harm for me.

Re:What is Wayland?

[Anonymous Coward]

And do you remember the old assumption that things like UI would always be done badly by big companies?

Nope, never heard of it. It's also untrue. DECWindows on VMS was georgeous. X on SunOS/Solaris was gorgeous. Even AIX w rs6000 X was pretty.

The real sad part of all of of this is even people who've used X for a couple or three decades can still be surprised by how it works. Eg., ssh -X to another box, then run "xscreensaver-demo &". It pops up on the box you ssh'd from. xscreensaver helpfully warns you that your screensaver daemon is running remotely, and if you don't share a filesystem with that box, you might wish to sort that out. Er, no I didn't want that actually.

Huh. Thanks.

Re:What is Wayland?

[unixisc]

Canonical is not the only ones behind it. Qt is looking at supporting it from Qt5 onwards, and that would enable KDE5 and the next Razor-qt to be Wayland compliant. Similarly, GNOME too is looking @ supporting Wayland. That's the big 3. On the OS side, Wayland is currently Linux only, and BSD currently has no plans to adapt it.

Re:What is Wayland?

[rubycodez]

yes, cinnamon is good, but it is fork of mutter, the gnome3 desktop manager. it fixes many things the GNOME3 developers wouldn't address

Re:What is Wayland?

[rubycodez]

no, the desktop managers on the commercial Unix were mostly inferior to GNOME2 in features and responsiveness. CDE, OpenLook, ick. The only decent one was SGI IRIS in IRIX, very feature heavy but with crisp response, not with the mushy lag of its competitors from HP, IBM, DEC

Re:What is Wayland?

[rubycodez]

sure it is, Canonical is the heavyweight linux distro behind it. fedora being a tiny shadow of its former self after RedHat pissed off people like me with their attitude and treatment of the community

Re:What is Wayland?

[rubycodez]

I know a few hundred linux users, but not a one is using Unity.

Re:What is Wayland?

[rubycodez]

Canonical is the heavyweight distro behind it, but they are losing mindshare in the UI space

Re:What is Wayland?

[unixisc]

Once Qt and GTK support Wayland, KDE, GNOME and other Qt/GTK DEs, such as Razor-qt, LXDE, et al will follow. And once they are on Wayland, distro support too will fall in place. Besides, a lot of Ubuntu based distros are out there, and even if they don't adapt Unity, they may still adapt & support Wayland, simply b'cos of the improvements it'd bring to KDE, GNOME and others.

xeyes ...

[PPH]

... needs bifocals. But otherwise, it all runs fine.

I can remember back in the 'old days' running X over a 28K dialup. But now, with 100 Mbit and up LANs and decent broadband, I can run most apps without being able to differentiate between local and remote clients.

It still just works.

Re:xeyes ...

[Animats]

I can remember back in the 'old days' running X over a 28K dialup.

Right, when the only tools used were XTerm and XClock. XClock wasn't very useful, but at least it was "graphical".

BSD/Solaris systems

[fm6]

Solaris is based on System V, The rebranding from SunOS to Solaris (back in 1991!) coincides with the move away from a BSD-based OS.

Re:BSD/Solaris systems

[Anonymous Coward]

Solaris is based on System V R4, which was itself an attempt to integrate System V and BSD. (With mixed results from my POV.) So not just a move from BSD -> System V.

Happy Birthday

[ADRA]

I only met you 12 years ago, but I've been obsessed with you ever since. Its been so long, that I just had to say something now.. Please please don't jump the shark like so many others these days... I just love you the way you are!

Re:Happy Birthday

[Anonymous Coward]

Yeah, it's been around for a quarter century, maybe people will stop calling it X-Windows now, finally!

Re:Happy Birthday

[Sique]

You think, XXV Window may be more appropriate?

Re:Happy Birthday

[unixisc]

Shouldn't it be X-XXV Windowing System?

Better than usual from Phoronix

[mickwd]

Surprisingly level-header article, given the source (Phoronix).

I really do hope Wayland sorts out a good scheme for remote access. At the moment it seems to be just ignored.

I wish people who set out to *replace* an existing piece of software would endeavor to replace it in its entirety, not just the subset of features that they happen to be interested in.

Re:Better than usual from Phoronix

[Balinares]

Err... What's wrong with the X11 protocol as a remote access scheme?

Just because we're taking it out of the rendering loop in Wayland doesn't mean we can't still use it outside.

Re:Better than usual from Phoronix

[nurb432]

Wayland from what i can tell is short sighted and more of a throw back to the island mentality. People who think RDP scenarios can replace network transparency really need to get a clue.

Its the wrong way to head.

Re:Better than usual from Phoronix

[arielCo]

Here you go:

New XWayland Protocol Published

Re:Better than usual from Phoronix

[Randle_Revar]

Agreed. And we lose all the window managers, and if you want to write a new one, too bad! You have to write a compositor instead! And don't get me started about the decorations being controlled by the app instead of the WM. Grrr...

Re:Better than usual from Phoronix

[causality]

Agreed. And we lose all the window managers, and if you want to write a new one, too bad! You have to write a compositor instead! And don't get me started about the decorations being controlled by the app instead of the WM. Grrr...

Anyone who has ever used Lotus Notes on WinXP can understand why that's a terrible idea.

"Oh is it checking for new mail? Yeah I can tell because the window is blank, featureless, and won't respond to input for several seconds..." Letting the apps have this control is why so many Windows systems feel subjectively sluggish and less responsive. At least when you're used to X with a low-latency kernel.

I'm aware that it is possible to write Windows applications that don't suffer this problem, so it's not necessarily a failing in Windows per se, but it is a design decision that leaves open an undesirable possibility that doesn't need to be there.

Re:Better than usual from Phoronix

[fast turtle]

yea! something called Firefox has this problem. Maybe they'll finally get their act together and fix it.

Re:Better than usual from Phoronix

[markdavis]

+1

People who don't understand what you said have never used thin clients. We still use them at work and they are great for particular situations. For example- one host manages the login process and window manager and apps are launched on various different hosts and displayed and the user doesn't know they aren't all local on their machine.

X11 might be old, but the architecture is still completely valid. Instead, people are trying to away and destroying forward compatibility. All they chant is how it will be possible to run Wayland apps in an Xserver running under Wayland. But that does NOTHING for apps that are designed for and compiled for Wayland. THEY will have zero network transparency and will not run on remote Xservers.

What should have been the goal is upgrading X11 to perhaps something like X12.

Re:Better than usual from Phoronix

[fikx]

So, How do I run a Wayland app remotely using X11?

Re:Better than usual from Phoronix

[Anonymous Coward]

Yeah, islands tethered to a continent of cloud-based, data-mining, app servers. It's like putting the throttle, brakes, and steering on your car under the control of the people who sell billboard advertising space. Oh, wait, I'm giving It ideas, but It's probably already working on that, given the autonomous vehicle initiative.

At least our monitoring devices will have nice pretty displays that take complete control of your system and decide, well, you didn't really need network transparency after all, did you? I sure hope Wayland isn't just jacking up a vintage Daytona Spyder to run a log splitter, but it looks that way. that's not all that's getting tossed out, and it looks like somebody wants to really dumb down and disempower the desktop, to, oh, Windows level. Or the other commercial alternatives. Follow the money as usual.

Salty the Peanut

Re:Better than usual from Phoronix

[Paul Jakma]

All they chant is how it will be possible to run Wayland apps in an Xserver running under Wayland. But that does NOTHING for apps that are designed for and compiled for Wayland. THEY will have zero network transparency and will not run on remote Xservers

Where are these apps? Why will their authors have chosen to write their applications directly on top of a low-level rendering library? It's pretty unlikely. Rather, they'll be writing using a toolkit, like GTK+/GDK or Qt, etc. And these toolkits *already* support multiple rendering backends, from bare-framebuffers, to X11 (via Xlib or XCB or whatever).

Re:Better than usual from Phoronix

[Anonymous Coward]

unfortunately, i think it's going to end up like the unity debacle. Everyone will hate it, but a few major distros will push it as the standard thing, and it'll do great damage to the public image of linux, further marginalizing linux on the desktop.

The unity folks STILL haven't admitted that it was a mistake, no matter that the hate is all but universal, so i'm not optimistic than wayland will fare any better. They seem just as intent on ignoring feedback as the unity folks are.

Re:Better than usual from Phoronix

[CajunArson]

X11 as a remote access scheme is actually craptacular in many ways, with two listed below as a non-complete list:

1. Ever try actually using X11 for anything even remotely graphically complex over even a rather decent broadband connection? You could also gouge your eyes out for a similar effect! Before you say how great X11 is over your cable modem:
a. If it were so great, then Nomachine would never have come into existence and NX would not exist.
b. If I had a dime for every time somebody says that X11 is great *because he is forwarding X-terms over it using an @#K%JJ SSH tunnel* then I'd be rich and they'd be put into a mental asylum where they belong. I'm talking about *real* graphical applications being shot over a broadband network here, otherwise there is no point to "network transparency" to begin with.

2. Real simple scenario that I've known can't work for over 10 years and for which there is no solution available using X:
      a. I run a program remoted to my desktop. Yay network transparency (blah blah blah).
    b. I get up from my desk and grab my notebook/tablet/smartphone/etc. and I want to simply transfer the remotely displayed application to the other device.. *cannot be done*.
Note how I spotted this problem 10 years ago? That was long before everyone was carrying around smartphones/tablets/etc., I was way ahead of the curve and this issue has only gotten more important over time.
  c. What's really hilarious is how many people have called me stupid or moronic for thinking that actually have *real* network transparency over X instead of the crap version from 1985 we are stuck with now would be a good thing.. and many of these same people lovingly brag about how they use screen all the time....

That's 2 issues.. there are many more. People who seem to despise any OS other than Linux for "not innovating" really tick me off when they try to kill the first real piece of innovation in the Linux graphics stack that we have seen in this century.
     

Re:Better than usual from Phoronix

[markdavis]

There are none right now (since there is really no Wayland). But the whole point of Wayland would be to compile the apps without X11 support. Otherwise, why bother with Wayland at all?

Re:Better than usual from Phoronix

[Anonymous Coward]

And yet, this is contrary to the "Unix Way"- where a particular piece of software does one thing and does it well, but also integrates nicely with other software with clear messaging.

X11 is overly featureful and bloated- jack of all trades and master to none.

Re:Better than usual from Phoronix

[0123456]

a. If it were so great, then Nomachine would never have come into existence and NX would not exist.

X11 was designed for LAN use, hence the excessive reliance on round-trip messaging. As NX and other proxies have proved, there's nothing particularly difficult about fixing the protocol for WAN use.

People who seem to despise any OS other than Linux for "not innovating" really tick me off when they try to kill the first real piece of innovation in the Linux graphics stack that we have seen in this century.

X11 was innovative. There's nothing innovative about Wayland, it's throwing away everything that separates Unix graphics from the rest of the world.

Re:Better than usual from Phoronix

[Paul Jakma]

    b. I get up from my desk and grab my notebook/tablet/smartphone/etc. and I want to simply transfer the remotely displayed application to the other device.. *cannot be done*.
Note how I spotted this problem 10 years ago?

Actually, I was able to do this just fine, 10 odd years ago with iPaqs handheld and GPE, which was X based. I could bounce GPE applications from my iPaq to main computer, so that I could use my main keyboard and display with the application (much handier than the pen based input on the iPaq). There's nothing in X that stops this from being implemented. It's the clients (i.e. there toolkits) which have to learn to switch between servers. Alternatively, it can be done done with an X server proxy - at least architecturally. (I can't actually remember which way the GPE solution did it).

Re:Better than usual from Phoronix

[Belial6]

I'm an actual fan of Lotus Notes/Domino, and I will agree that that is definitely one of it's warts.

Re:Better than usual from Phoronix

[Paul Jakma]

You seem to have completely missed the point of my reply. Again, apps for Wayland will almost certainly be written to use a toolkit library like GTK+/GDK or Qt. The app itself will *not* be written directly for Wayland - that'd be insane, because that'd require the application to completely re-implement its own child-window and input handling, as well as widgets (i.e. re-implement the stuff Qt, GTK+, etc already provide). So the app will be Wayland agnostic. Instead it will be GTK+/GDK or Qt that does the Wayland rendering. These toolkits *already* support multiple rendering backends, already support X11 (via Xlib or XCB) - and this can be runtime selectable.

So, again, the applications will be written using some toolkit, like GTK+/GDK or Qt, like they already are today. The toolkits will determine what display protocols are supported. These could easily allow you to choose whether to render locally or via X11, at runtime, as they already today are capable of supporting rendering choices at runtime. Hell, they could even do it transparently, without asking you, based on whether DISPLAY is set or not.

Re:Better than usual from Phoronix

[Anonymous Coward]

Lotus Notes.

That's a name I've not heard in a long time.

A long time.

(Lotus Notes is a terrible idea. Period.)

Re:Better than usual from Phoronix

[Anonymous Coward]

So, because I can't display a heavily graphical app over a broadband connection, we should remove the ability to do xterms and the like as well?

And for the record, there are networks fast enough to make this happen. It's not only usable by "broadband" connections (implying home usage).

Re:Better than usual from Phoronix

[shiftless]

It's the clients (i.e. there toolkits) which have to learn to switch between servers.

Sounds trivial.

Re:Better than usual from Phoronix

[shiftless]

X11 might be old, but the architecture is still completely valid. Instead, people are trying to away and destroying forward compatibility. All they chant is how it will be possible to run Wayland apps in an Xserver running under Wayland. But that does NOTHING for apps that are designed for and compiled for Wayland. THEY will have zero network transparency and will not run on remote Xservers.

SO?

Re:Better than usual from Phoronix

[CajunArson]

WRONG.
I'm saying that remote applications should be displayed *the right way* and Wayland does nothing to prevent the ability to display remote applications *the right way*.

Go look at the chrome remote desktop extension: https://chrome.google.com/webstore/detail/gbchcmhmhahfdphkhkmpfmihenigjmpp

When a stupid browser extension does a better job of cross-platform remote GUI management than X, it is time to do something better than X instead of pretending that we reached a magical Utopia in 1985 and that anybody using facts and logic to disagree is some sort of religious heretic.

Re:Better than usual from Phoronix

[Anonymous Coward]

"Broadband" is a very broad term - generally anything faster than ISDN (something less than 1 Mb/s at very most, IIRC) would be called broadband just as my fiber to premise that's capped to 100 Mb/s and capable of many times that is still broadband as well.ly, Furthermore my home gigE LAN does something like 400 Mb/s solid. Retards will claim that this is this way more than needed to run X11 over network. And they are retards exactly for that. Proper GUI applications such as Clementine or Firefox are just not as responsive as local instances at all times and are even almost really usable over merely "broadband" tunnel to my uni that, too, has fiber to premise but the bottom really falls through if you dare to try streaming even 720p (1280x720 YV12, I believe) video without audio overhead. It's impossible even with LAN that's limited by how fast the modern integrated network adapters can transfer data over a short cable! Using Xv acceleration instead of pure X11 makes it somewhat better but it will still stutter unbearably and 720p is not even FullHD (1080p or 1920x1080). And because X11 doesn't do audio, you also have the joy of tunneling PulseAudio (doable and done by this AC but the experience is something around 4th level of hell with scant documentation) or learning about the "prim and proper (non-Poettering)" unix audio in the form of Network Audio System but, really, you would have to be a blind hobgoblin with sever case of STD to go near that one.

Re:Better than usual from Phoronix

[Anonymous Coward]

I fully agree, CajunArson. Just wanted to let you know that you are not the only one out there (there's a whole two of us now).

Re:Better than usual from Phoronix

[jbolden]

Ultimately there are 3 situations:

a) Running an application on a machine sharing ram with the video card.

b) Running applications on a machine close enough to the video card that the latency between them is lowish and the bandwidth is plentiful and performance is irrelevant.

c) Running applications on a machine where either the latency is high or the bandwidth is limited

X11 does terrific for (b) in exchange for damaging (a) and (c). X11 was designed in a world where (b) is common. The entire X11 approach is optimized for (b). If you live in the world of (b) Wayland is going to be a downgrade.

Re:Better than usual from Phoronix

[jbolden]

I would imagine by Gnome 5, KDE 6 they will be Wayland only. The reason will be the enhanced possibilities / features once the video buffer and the application are sharing RAM. That's what's going to drive the move to Wayland only apps.

Re:Better than usual from Phoronix

[markdavis]

High latency/low bandwidth can be fixed using X11 with schemes like NX, which are very effective:

http://freenx.berlios.de/
http://en.wikipedia.org/wiki/NoMachine
http://www.nomachine.com/

Re:Better than usual from Phoronix

[complete loony]

Remote access is not what Wayland is for. The implementation of X11 has historically been responsible for too many different services that should have been separated a long time ago. From mode setting to compositing, to the interface for separate window management, the implementation of X11 has been showing its age for a long time. Wayland is about compositing your local desktop. Taking over a number of roles that X has historically served, while still allowing you to use X or other network protocols for remote access.

Re:Better than usual from Phoronix

[mpol]

Ehm, isn't there xmove that lets you move X11 apps from one X11 server to another?

Re:Better than usual from Phoronix

[TeknoHog]

There is also xpra, which is basically "screen" for X. It is not perfect, but gets the job done.

Re:Better than usual from Phoronix

[Anonymous Coward]

It sounds like you're wanting to use "xmove", which allows you to repoint a running application at a new display.

Re:Better than usual from Phoronix

[Rich0]

And what happens when those toolkit maintainers decide "hey, nobody uses X11 anyway, so let's drop all this crufty code?"

After all, if the use-cases for X11 are so minor that it isn't worth porting them over to Wayland, then they're going to be so minor that it isn't worth having anybody else support them either. You're just pushing the same decision further down the chain.

Or maybe QT keeps support and GTK+ drops it, and then you have this wonderful experience where half of your apps work and half don't.

Re:Better than usual from Phoronix

[Coryoth]

Ehm, isn't there xmove that lets you move X11 apps from one X11 server to another?

Well presuming you want something that had development discontinued in 1997 and has been essentially unmaintained ever since ...
Even the man pages have strong warnings that xmove is old, unmaintained, and not guaranteed to work at all.

Re:Better than usual from Phoronix

[Anonymous Coward]

If you need remote xterm, go ahead, but xpra doesn't support OpenGL so it's virtually useless for about 50% of my software unless I want to see pixels being sent over network. I expect xmove has the same problem - only core X11 is supported not the trillion extensions we need to force X11 from year 1984 to 2000.
ALSO DETACHING AND REATTACHING X CLIENTS SHOULD HAVE BEEN A CORE FUNCTIONALITY! Lack of it is responsible for oh so many extensions over the decades. Resizing, reconfiguration, X server crash recovery and more, all of those would be non-issues if clients could be disconnected from a running X server and reconnected to another instance.

Re:Better than usual from Phoronix

[tqk]

I'm an actual fan of Lotus Notes/Domino ...

You really ought to seek professional help for that.

ssh X11Forwarding even in Cygwin

[digitalaudiorock]

One thing I love about X has always been the ability to run gui apps remotely via ssh using your local X via X11 forwarding. For those who haven't tried (or haven't tried lately) it's even pretty easy to get a shell running within X in Cygwin and run remote gui 'nix applications under Windows...too cool.

Re:ssh X11Forwarding even in Cygwin

[lindi]

X11 forwarding is not very nice since the remote server can fully control your client machine in many cases (e.g. in Debian/Ubuntu "ForwardX11Trusted" defaults to true). I prefer xpra instead.

Re:ssh X11Forwarding even in Cygwin

[Anonymous Coward]

X11 forwarding is not very nice since the remote server can fully control your client machine in many cases (e.g. in Debian/Ubuntu "ForwardX11Trusted" defaults to true). I prefer xpra instead.

oh noes! however will be change the default setting?! if only there were text editors and human-readable config files...

Re:ssh X11Forwarding even in Cygwin

[Paul Jakma]

Pedantry: The X *clients*[1] may be able to snoop on all your keystrokes, including to other applications, and read what is on your display.

1. The X clients are the applications you run, which connect to the X server. The X server is responsible for co-ordinating drawing to the display. X client may run remote from the X server. The X server typically runs local, on the machine you're sitting in front of, and whose display screen you're looking at. I.e. in X, the server is the local bit (on your machine) and the clients are (potentially) remote, running on other machines - this often confuses people. ;)

Re:ssh X11Forwarding even in Cygwin

[Anonymous Coward]

If all you need is ssh+X11 under windows, I like mobaxterm a lot more than cygwin these days.

Re:ssh X11Forwarding even in Cygwin

[barjam]

And it is awful. All remote desktop access to unix/Mac is awful. X, Vnc, no-machine. Windows excels at this which is funny because they started out not having this functionality at all and unix folks would make fun of them for lacking it. Now it is the unix variants that largely lack a usable technology in this space. Of course it is rarely needed as unix servers excel at being administrated via shell and windows sucks at that.

Rdp over ssh works well and is many times faster than X or VNC.

Re:ssh X11Forwarding even in Cygwin

[phantomfive]

If you like RDP that much, then use it on OSX or Linux. There are clients for both.

Personally at work I use both RDP and VNC to remote into various servers, and neither one is particularly better than the other. I get the job done with both.

Re:ssh X11Forwarding even in Cygwin

[mpol]

One thing I really like is the aspect of 2 cut-and-paste buffers. When I explain it to tech-friends of mine they really are amazed.
There's a primary buffer and a secondary buffer. The secondary buffer is like the buffer in Windows and MacOS. You have Ctrl-C for copy and Ctrl-V for paste. Or you can use the context menus.
The primary buffer however is everything you selected, and then ofcourse only the last selection. Pasting is done with middle mouse button.
This way you can use 2 buffers. Like you Ctrl-C the username, and select the password. Place focus on the other app. Paste the password from the primary buffer, and Ctrl-V the username. All in one go.
I just don't want to live without it.

Re:ssh X11Forwarding even in Cygwin

[digitalaudiorock]

Speaking of the primary buffer, I can't live without my parcellite copy history app. There are others like it of course, like Klipper etc....but I use parsellite because it's nice and lightweight and works well with fluxbox. On the increasingly rare occasions that I use windows, I have a hard time getting used to NOT having my highlighted text copied automatically.

Re:ssh X11Forwarding even in Cygwin

[Anonymous Coward]

There is an RDP xserver that works fairly well (Not the hack that uses Xvnc) only problem with it is clipboard sharing to rdesktop (It works to the Windows client).

The Sun Turbo VNC / Virtual GL combo when connecting to a sparc with an XVR1000 works really well (Sadly only as far as Solaris 10 though).
Sunray is pretty close to perfect. (I guess Citrix to UNIX stuff works properly as well).

Windows (As of 2012) is best managed from the CLI i.e powershell (Or remotely).
Funny thing is the way powershell works in Windows 2012 is closer to what UNIX used to be like. (Learn the basic principles and then you can do anything these days you need to learn too many commands one by one).

Linux as it is now is awful and pretty much as inconsistant as Windows has traditionally been.
(And it is polluting the other UNIX's (Solaris/*BSD) - dbus / hal / avahi / pulseaudio).

Windows is getting better Linux is getting worse. (Even the Windows 8 graphical monstrocities are less awful than gnome3/unity).

Re:ssh X11Forwarding even in Cygwin

[Miamicanes]

OK, big, honest question... you have a computer running Mint 13 with Cinnamon. Or KDE, if it makes things easier. You want to connect to it over the LAN from a laptop running Windows using a win32 X server. What, exactly, is the proper way to do it? Is there any way to use X11 remotely in a RDP-like manner (passing primitives over the LAN instead of scraping bitmaps like VNC)? Or do you basically have to set up each and every single application you want to run remotely over X, and configure the Win32 X server to be your local window manager, font server, etc? I've been told that if you wanted to genuinely run Gnome or KDE remotely over X, you'd theoretically have to port KDE or Gnome to Windows, then integrate them locally into the win32 X server, and hand-configure the windows X server to mimic the look and feel of your desktop settings on the remote computer running Linux.

It seems like Linux (and by extension, X) has no real RDP-like middle ground between X and VNC that allows casual, adhoc remote desktop rendering without having to settle for bitmap-scraping.

Re:ssh X11Forwarding even in Cygwin

[Miamicanes]

AFAIK, there are ways to connect to a Linux box "using" a RDP client, but at the nuts & bolts level, all they're doing is scraping bitmaps just like VNC and wrapping them in RDP to kludge it into working.

Their only advantage is that you can use the RDP client you already have instead of having to download and install a VNC client first.

Re:ssh X11Forwarding even in Cygwin

[firewrought]

Two separate clipboards makes for a confusing UI. Not that it isn't handy once you learn to exploit it, but... from a polish standpoint: ugh.

FWIW, Windows users might want to try ditto... I think it's called dittocp on sourceforge. It remembers the last few hundred things you copied to the clipboard, lets you text-search it, save/group snippets for future use, and paste any of it quickly. It takes a little bit of learning, but it's incredibly handy for coding and writing sql.

Re:ssh X11Forwarding even in Cygwin

[armanox]

I've found that NoMachine (or just the NX protocol in general) runs at least as fast as RDP.

Re:ssh X11Forwarding even in Cygwin

[armanox]

Before I can answer I have to ask what you're trying to do. Are we running the whole gnome-session (or KDE or w/e), or just individual apps? And do we want it in a Window,.full screen, or managed by Windows? Because all are possible with X11 (either XDMCP or X over SSH) and some with NX (which runs much nicer then VNC or X over network). (See this pic for an individual app having been called over SSH to run on Win32 X).

Re:ssh X11Forwarding even in Cygwin

[Anonymous Coward]

I think what you are after is XDMCP. Start the X server on your local machine and have it connect to a desktop session on a remote machine? By default this will, however create a whole new desktop session for you that won't be the same as the one that would be displayed on the remote machines connected display, if it has one.

Try it with "Xnest -query <ip address>" to run a whole desktop in a window. You will need to enable XDMCP on the remote machine first (usually), check the settings panel for GDM or KDM.

Re:ssh X11Forwarding even in Cygwin

[caseih]

I beg to differ with you that "it is awful." On a LAN, X11 over ssh just works, works fast, and is the most transparent of the solutions you mention. I use it every day to access apps on my virtual machine, my server, and other machines in the house. And when I worked for a university, we used it all the time and it was plenty fast. Just as fast as VNC or NX (on a LAN mind you).

If you need to run remote X11 apps over anything less than 100 Mbit LAN, then yes NX is the ticket. It can display individual apps or a full desktop, and it's fast. It's really an optimization of the X11 protocol to eliminate round-trips, though to do this it has to run an X server proxy (an X server really) on the remote system. I use NX frequently to access a graphical app on a server half way around the world.

Until VNC is per-window (application aware), it's simply not a solution. RDP is decent, but of course windows-centric.

Re:ssh X11Forwarding even in Cygwin

[evilviper]

All remote desktop access to unix/Mac is awful. X, Vnc, no-machine.

NX works extremely well. Needing to share an SSH key for the NX user is awkward, but not a big drawback. My only compaint it that there aren't NX clients for more platforms... Namely: Android. But with a working Xserver on Anroid, somebody just needs to be interested enough to make it happen.

RDP isn't great by any means... it's a window in a box, and doesn't have anything like SSH+X11 which makes it so easy to run commands remotely, and just have the single application display locally.

And I should mention, RDP is just a pared-down Citrix MetaFrame, a service which is also available for Unix servers.

Re:ssh X11Forwarding even in Cygwin

[tqk]

The X server typically runs local, on the machine you're sitting in front of, and whose display screen you're looking at. I.e. in X, the server is the local bit (on your machine) and the clients are (potentially) remote, running on other machines - this often confuses people.

Ya know, I've been hearing that mantra since my first days running X, and I've never found it confusing in the least.

Re:ssh X11Forwarding even in Cygwin

[dbIII]

You are describing something limited to one to one - nothing but a mirror of a desktop or virtual desktop, and then comparing it to the one to many solution that is X. In any situation where a user wants to run something on more than one host a one to one solution is going to be inferior no matter who it is from.
You can get around it with X and then RDP or a well written application on a web server to tie stuff together and present it as a web interface, but RDP alone does not replace something like X.
Also you need specific stuffing about, reserved machines and careful reading of licences to present a remote desktop to MS Windows users while running an application via X on the same machine that somebody else is already using is trivial. I don't want to have to set up an MS Windows server just so one user can use something remotely on ocassion in a way that MS Windows could do better in NT3.51 instead of the nerfed bullshit it provides now.

Re:ssh X11Forwarding even in Cygwin

[Rich0]

Yeah, google for untrusted x11 forwarding failed and read about the misery you get when you change that default. That feature is essentially broken with modern versions of the various pieces of software.

Re:ssh X11Forwarding even in Cygwin

[Rich0]

That's the thing that bugs me. There are plenty of things about X11 that could stand improvement, and the fact that any client can evesdrop on your keyboard app is just one of them. The problem with Wayland is that they are tossing the baby out with the bathwater.

Re:ssh X11Forwarding even in Cygwin

[Anonymous Coward]

AFAIK, there are ways to connect to a Linux box "using" a RDP client

Well, then you must not know very far.
It works well but does not have printing support.

Re:ssh X11Forwarding even in Cygwin

[Anonymous Coward]

Hell, I've used X11 over a 28k modem. It's way better with DXPC or some other protocol proxy, but it is (was) doable.

I'm amazed at how little modern systems with massive CPU and RAM have sped up responsiveness.

I used a tek4010 emulator to run CAD over 2400 baud to a microVax with 4 seats in college (1988). The room only had 4 seats, but there were more serial lines and a dial up so I could have a 5th seat to do my assignment. I could go to a student lab with 9600 baud serial to get near interactive use.

When I got my 1st X11 setup in '92 it was way better. SunOS 4.1.3 Sparc 1+, 16MB RAM. A 486 w/ SLS Linux, 16MB RAM and ATI VL could do graphics faster, but couldn't multitask as well. Switched networking and 100T LAN didn't exist for another few years, but it was enough to run Xterminals with a few Xterms running emacs and LaTeX and DVI viewers for a few secretaries. We would've killed for OpenOffice! Or NX.

Re:ssh X11Forwarding even in Cygwin

[spitzak]

You seem to have gotten the wrong story.

What *does* work is what it sounds like you are asking for. The Linux machine can run the Gnome/KDE program and display it on the X server on your Windows machine. It will show up in a window on your desktop.

Possibly you were confused by what is a problem: if you wanted to take some KDE or Gnome program and *run* it on your Windows machine (ie *not* over the network from your Linux machine), you would think all that is needed is the X server and to recompile the program. Or maybe you think you might also need cygwin installed before you compile. You will quickly find that you also need to install *huge* amounts of stuff, pretty much the entire KDE/Gnome desktop set of libraries, before the program will compile. In fact X is probably the only thing you *don't* need (GTK can be compiled to call WIN32 directly). This is a certainly a PITA but is not exactly the problem you described.

Re:ssh X11Forwarding even in Cygwin

[jwhitener]

If you like that, you should try putty + vcxsrv.sourceforge.net + mRemote.

mRemote can connect using ssh, http, citrix, rdp, vnc and more protocols. When you set up a ssh connection for a server, you can choose a putty profile to use. And then putty is setup to talk to your xserver provided by vcxsrv.

From the article

[Anonymous Coward]

It only took three years to go from X1 to X11

Of course. First X1, then X10, then X11 (to be followed by X100, then X101, then X110, then X111, then X1000, etc.)

You must be new here.

Re:From the article

[Anonymous Coward]

It only took three years to go from X1 to X11

Of course. First X1, then X10, then X11 (to be followed by X100, then X101, then X110, then X111, then X1000, etc.)

You must be new here.

X6 and X9 were also released: http://en.wikipedia.org/wiki/X_Window_System#Release_history

Re:From the article

[Gherald]

Actually, clearly X0 came first.

You must be new as well.

Re:From the article

[Anonymous Coward]

Well, X0 could be a place-holder for "0 lines of code". This makes sense for two reasons:

0) Minor version changes before the major release of X0 would want to be negative.
1) Logically speaking, "X" would be a cleaner completion to the sequence than "X0", which could be confused with the name without the version number.

Re:From the article

[TeknoHog]

I'll drink to that!

Re:From the article

[Anonymous Coward]

Of course that pattern only lasts until some witty GNUtard releases XX and the regexes on ICBMs all over the world end life as we know it on this planet.

Re:From the article

[Misagon]

It ran on a high-end Laserdisc player from Pioneer?

Re:I can only hope...

[nurb432]

You have no clue what X11 is all about, and are a fool.

The olden days

[gdav]

The first time I used it was in 1993 when NCSA Mosaic came out. There was a copy on the university Sun box, my office boasted a spare 286 running DOS - some packet drivers and Vista-eXceed for DOS and I was away!

Re:The olden days

[antdude]

For me, it was in my university's CS lab. I never used Linux/UNIX's GUI before (only dial-up shell accounts during my high school days) and I think I also used NCSA Mosaic web browser on the school's fast pipe (fractal T1!). Years later, I used my own X11 and Linux on my home computers. :)

Re:The olden days

[Anonymous Coward]

I remember when I first came across with a Linux desktop. I started using X with HP's workstations and had been accustomed to the mouse and keyboard controls of the HP so well that the result I got from the Linux desktops at the beginning were mostly annoying beeps. I was considered pitiful, or an idiot at the computer class. I just logged in to the desktop environment I knew by heart for that release of frustration.

Re:The olden days

[rmdpgh]

... I also used NCSA Mosaic web browser on the school's fast pipe (fractal T1!). Years later, I used my own X11 and Linux on my home computers. :)

Dude! A fractal T1? A T1, within a T1, within a T1???? That must've been some serious bandwidth! I'm freakin' jealous! :)

Re:The olden days

[antdude]

Hahaahah, yeah and it was slow too during peak hours. :O

Wayland?

[Anonymous Coward]

Wayland trying to unseat X11? Does anyone even use Wayland yet? Shouldn't the summary say Wayland is vaguely trying to get a toe-hold?

Whenever Wayland comes up there is always a crowd of people talking about network transparency. Personally, I'm not at all concerned about that. What I am interested in is whether Wayland interface actions (such as clicks and keyboard input) can be scripted as they are with X. Programs like xdotool make it possible to simulate user input on the display and that can be quite helpful. Not sure if Wayland has anything similar, but I hope it does.

Re:Wayland?

[jedidiah]

Everyone has their own reasons for liking "the old things". When snot nosed kids come along repeating 15 year old propaganda from guys that can't even do their own interfaces right, they get all excited and want to tear everything down. New is good and old is bad as an article of faith.

They're full of themselves and think they're smarter than everyone else and thus don't bother to actually consider the end user and what people's requirements might be.

You end up with something that's shiny and new and worthless.

This kind of drama plays out in corporate environments all the time.

Re:Wayland?

[Anonymous Coward]

I just want something to make proper use of my 6 core CPU and $500 GPU. Keep up or get out of the way.

Re:Wayland?

[Anonymous Coward]

Honestly gaming under wine/X has gotten pretty good. I can run LOTRO/DDO @40-60fps on a Radeon 4770 with 512 megs of ram just fine. Now where the problem comes in is ineffeciently coded apps, such as Star Trek Online, certain EA games, and other apps that have very specific failures within wine's emulation.

Nowhere within it has X performance been a major issue. In fact every time I've had graphical slowdowns it's turned out to be because of either ALSA, another app (such as chrome or firefox bloating), or running multiple GL apps.

Honestly fixing ALSA so it doesn't constantly have underruns and lag would do more for Linux systems than replacing X with wayland. (I now have a 10 year old SB Live in my gaming system because under linux and linux only, the soundcard stuttering causes significant and horrible lag while gaming.)

The secondary issue: ATI cards don't ramp up their fans fast enough leading to some microstutter if your room isn't cooled (solvable by running 'aticonfig --pplib "set fan 0 100"' to run it full-bore.)

With those two 'fixes' applied, almost every game is playable on what is now an unsupported videocard (AMD having dropped Pre-HD5000 card support with Catalyst 12.6 on linux and 12.8 on windows.)

Re:Wayland?

[unixisc]

Speaking of which, I wish that Wayland would work on the sound problem just like it does on the compositing problem, and just take it away from ALSA/PulseAudio and whatever else there is. On the Linux side, as is.

Re:I can only hope...

[gigaherz]

I do have a clue, and I still want it to die.

Original O'Reilly Manuals

[arthurpaliden]

I still have my original copies of the beta 8.5x11 in spiral bound O'Reilly X11 manuals. My wife was going to thrown them out last year but I saved them.

Re:Original O'Reilly Manuals

[chromatic]

I rescued mine from the castaway bin at HP in '98 or '99 and even used the xlib manual once or twice.

The next time I move, the desire not to carry them will outweigh the nostalgia factor. Pun intended.

age becomes me

[at10u8]

I remember the first vendor demo workstation arriving with X running on it. I remember GraphOn X terminals, and NCD X terminals. I remember rewriting the Keck CCD image display program not to send each image 3 times and getting live readouts over 28k modem to my living room.

Re:age becomes me

[sconeu]

Back in the day (1992), we used to run SCO Unix on a 486 box with HDS X terminals connected to it.

Future

[StripedCow]

Any speculations on the future of windowing systems?

Personally, I wonder when we'll get the first windowing system based entirely on HTML5.

Re:Future

[bobs666]

I wonder when we'll get the first windowing system based entirely on HTML5.

That would be a web browser. Browsers can run in lots of windows. You can have that already. (right?)

NextStep tried using postscript to render window content. I never got to use one, ... as I understand the problem was the hardware was under powered. Perhaps an idea before its time.

Re:Future

[Paul Jakma]

Well, GDK (low-level library used by GTK+) has support for HTML5, apparently: http://blogs.gnome.org/alexl/2011/03/15/gtk-html-backend-update/

Re:Future

[Randle_Revar]

>windowing system based entirely on HTML5.
Shoot me now.

Re:Future

[Randle_Revar]

> That would be a web browser. Browsers can run in lots of windows. You can have that already. (right?)
Pretty sure they mean the WM is written in HTML

>NextStep tried using postscript to render window content
Yes, and Quartz 2d uses something somewhat related to Display PostScript (though it is PDF, not PS).

And Sun wrote NeWS entirely in an extended PS, including WM functions, which NeXT never did with DPS

Re:Future

[Misagon]

Yep, on NeWS the client could upload Postscript code for widgets that would execute as entities on the server, responding to user input without any round-trip to the client as in X (and Wayland).

NeXTStep and MacOS X Quartz used Postscript only for rendering. BTW, the core of PDF is an extended subset of Postscript, so a lot of it is the same.

Re:Future

[Anonymous Coward]

OS X uses PDF, not PostScript (PDF is royalty free; PostScript is not). And it doesn't actually use PDF for rendering screen graphics.

In a sense...

[Anonymous Coward]

wayland is simply the latest and shiniest x.org fuckup.

I always thought the acceleratedx people were being unprofessional in their public criticism, and they are. But diving into x.org makes it painfully clear why; it's really really hard staying professional looking at what they think passes for "architecture". It's no coincidence that other pile of steaming crap code (yeah, it's proprietary, but there's been leaks, oh yes there's meen leaks) runs circles around the unix world when it comes down to graphics. The x.org code base really is that bad. While wayland feebly tries to fix past mistakes, it still comes from the same people with the same bright ideas they always had. No, I'm not holding my breath. Why do you even ask?

Re:In a sense...

[unixisc]

I had no idea that Wayland is from x.org

change is good

[Vince6791]

Why are so many people bashing wayland? It's like when Window users were bashing Windows 7 for not being more like Widows XP, but they adopted it and not going back to the old XP. Change is good especially with technology. If ubuntu and opensuse incorporates wayland into their distro's, no biggie, there are other distro's that people can still use with X11. But, wayland will work fine in conjunction with X11 for software compatibility. This project was started by someone at Red Hat.

Re:change is good

[Osgeld]

well until one of you LN's can give me a reason TO change other than its here, it does stuff, you cant really tell the difference, use it you Luddite! I am really indifferent to wayland.

Re:change is good

[armanox]

It is a big deal - Red Hat essentially defines the Linux standard for the corporate world - so what they do the app vendors support.

Re:change is good

[jbolden]

I don't know what an LN is. But the main feature would be things like video integrated throughout the OS, like it is on OSX. That sort of high speed system with dozens of complex animations of videos is impossible under X11.

Re:change is good

[Osgeld]

LN = linux nerd

and care to explain why any of that matters? complex animations sounds like fluff that eats system power to present a box with items in it, and the last OSX I used was 10.2 and I wasnt taken back by its abilities in the mouse + desktop arena, there was a mouse pointer and a desktop, should I read more into that simple idea?

again why? whats the benefit for me to hose everything thats been working fine? toys? animations? I dont want my desktop to get in the way of applications.

Re:change is good

[Anonymous Coward]

Because client-side decorations suck.

Re:change is good

[jbolden]

The why is making video as smooth as text leads you to use video like you use text. Prior to the early 2000s I never cut and paste video or images, today I do very casually. I do that because of tremendous work to make it seamless enough. Those animations on my desktop allow my desktop to expand or contract things based on what I want to do. So scrollbars don't take up room on my windows until I want to scroll. The dock is invisible until I'm pulling stuff from it and then the icons expand so I can easily click them. etc...

But ultimately if you want 1990s capabilities Wayland offers you nothing. X11 works great for simple desktops and simple applications. The problems with X, like any system only become apparent if you are going to push it. The question is not how a system works at 1% utilization but at 30,50,70,90%.

Re:change is good

[sjames]

Not all change is good. If I come down with the flu, my health will have changed.

The problem with Wayland is that it is far too willing to throw away capabilities of X11 that people actually use frequently. If the new system wants to gain traction, it needs to be at least as capable as the old one.

First release of X ?

[Anonymous Coward]

The article mentions that X version 11 was released in 1987.

Does anyone know when the first release of X was?

Re:First release of X ?

[Mostly a lurker]

The first use of the name X was for X1 in June 1984. It originated at MIT. The name X was used to distinguish it from the earlier and rather different W (now you know why a Window system was abbreviate as "X": it was basically "W" mark 2). The obvious follow up question is when was W released? I cannot remember (perhaps never knew) and am too lazy to search the Internet to find out.

Re:I can only hope...

[schaiba]

How did you reach this conclusion?

DANGEROUS VIRUS

[Anonymous Coward]

First, a little history. The X window system escaped from Project Athena at MIT where it was being held in isolation. When notified, MIT stated publically that "MIT assumes no resonsibility...". This is a very disturbing statement. It then infiltrated Digital Equipment Corporation, where it has since corrupted the technical judgement of this organization.

After sabotaging Digital Equipment Corporation, a sinister X consortium was created to find a way to use X as part of a plan to dominate and control interactive window systems. X windows is sometimes distributed by this secret consortium free of charge to unsuspecting victims. The destructive cost of X cannot even be guessed.

Don't be fooled! Just say no to X.

Re:DANGEROUS VIRUS

[unixisc]

This was from the Unix Haters' Handbook. I loved that book - bought a hardcopy when it was out, and now have a softcopy.

Not HTTP-friendly, blew opportunity

[Tablizer]

Something like X-windows could have allowed for desktop-like GUI's over HTTP, but it doesn't handle latency well because it micromanages individual keystrokes and characters. Contrast with an HMTL text-box.

We wouldn't have to fart around with DOM and JS and Microsoft IE unstandards if they did it right.

Re:Not HTTP-friendly, blew opportunity

[Anonymous Coward]

Actually, IMHO, the problem with X11 is that most remote access connections have been split-speed. In 1995, I was running X11 over 128K ISDN, and while it was sluggish, it was very usable. Today, with 10Mb/1Mb split-speed ADSL, its absolutely unusable, which means that I have to use and RDP client into the LAN and then run X.

Re:Not HTTP-friendly, blew opportunity

[Anonymous Coward]

X11 Over HTTP? How did that get modded up?

For those with poor math skills, X11 was out well before the web even existed, and X10 is even older. Why would it support a protocol that didn't even exist yet?

Re:Not HTTP-friendly, blew opportunity

[Tablizer]

I meant creating an HTTP-friendly update.

You Must Be Kiddin

[Anonymous Coward]

X11 can run over dial-in speed (28K and even less) in a useful way (e.g. rendering an editor app). Show me how to do that with your Redmondian shite.

Re:You Must Be Kiddin

[shiftless]

Why in the world would I ever do that?

Re:You Must Be Kiddin

[Miamicanes]

The difference is, if you tried doing it with a 32-bit alpha-blended desktop, VNC would have to shovel raw bitmaps over the network when you moved that semi-translucent terminal window that's partially-obscuring kEdit. RDP would just say, "move window XXX to a new origin of (x,y)", and all the alpha-blending window-translucency magic (and dropshadows, and everything else) would be rendered locally.

RDP between two computers running Windows works decently well over slow network connections, even with large displays and 32-bit alpha-blended eye candy. The same can't be said for VNC.

Re:You Must Be Kiddin

[shutdown -p now]

GP was talking about networked X11 though, not VNC.

Though, as I understand, networked X11 with modern widget toolkits also involves a lot of bitmaps copied around.

Re:You Must Be Kiddin

[tqk]

X11 can run over dial-in speed (28K and even less) in a useful way (e.g. rendering an editor app). Show me how to do that with your Redmondian shite.

Why in the world would I ever do that?

Now, you wouldn't but back in the '90s when that's the best dialup POTS would do, you'd be happy with it. I was impressed with 9660 at that time. It worked admirably. I worked from home over it, and people were pissed to find I wasn't even in the office doing that work. They thought I was. Ha haaaaa!

Re:You Must Be Kiddin

[Anonymous Coward]

RDP is quite simply one of the best remote protocols out there, in practice, because it's fast, but you generally need a LAN, because once you get down to isdn speeds, it really sucks with window redraws and such. X11 forwarding sucks in my opinion, for several reasons, and the window redraw speed is not great at all on anything less than the fastest LAN.

But despite rdp being one of the best out there, it's Windows. And Windows still sucks no matter how fast the remote protocol is.

I've been doing remote admin on hundreds of UNIX and Linux servers for the last 4 yrs and the only access was ssh. worked like a dream, and was super efficient. I just moved jobs back to Windows world, and Windows rdp administration is just making me cry every day. Yeah, it's fast on the LAN, but "Windows administration" is just so frustrating and inefficient and unable to do some of the things Im used to from UNIX.

Indeed

[Anonymous Coward]

Running a powerful shell like bash. Or a powerful editor like Emacs. Still much more efficient than most of the blinky-shiny idioms of Windoze. It requires quite some investment in mastering these sharp tools, but then you are easily 500% more precise and 1000% more productive than the point-click-drag bozos.

But, "why do we need a difficult caterpillar, if all idiots master a shovel in five minutes ?"

Re:I can only hope...

[schaiba]

And why on Earth was I modded troll?

It Will Do As Much Damage as Anti-Mohammed Videos

[Anonymous Coward]

And that means, it will do no *real* damage. If you cannot live with diversity, please relocate to North Korea. Diversity is a good thing and those who need "simple, coherent views of Linux" should take Windoze or OS Thought Police X.

Diversity and competition are the basis of progress. If Wayland indeed brings something useful it will survive, otherwise it will die the death of gnome and KDE. Those two tried to emulate Windows, when all you need in reality is xfce. No, I mean this 100% serious. Miguel Icaza will pursue his love for M$ approaches, but who cares ? Real Unixers will use xfce, vi, LaTeX, wxWidgets and stuff like that.

Oh Yeasi

[Anonymous Coward]

I have seen Windows, MacOS, X11 (HPUX, AIX, Linux) and I am quite happy with being an X11 user. Yes, Motif was really nasty, but it appears that modern libs like wxWidgets are nice APIs. Commercialware comes and goes, while X11 is here to stay. Feel free to be ass-raped by M$, Google-NSA, Oracle, Apple and all the other mega-corporations. I'll pass.

Do YOU NEVER THINK ABOUT THE POOR SHAREHOLDERS ?

[Anonymous Coward]

The shareholders of Intel and Micron ? They live by the fact that all three years computer hardware is made obsolete by Windows Lala 20XX and Miguel's latest Gnome-crapola. I always thought you would be more responsible !

The captcha brigade says "vomited".

XOrg's X11

[jd]

...is just a "reference implementation". It is NOT the be-all and end-all.

If someone were to produce a wholly new windowing system that had a compatibility layer for the standard X11 API, support for the X11 configuration files, and the option of sending X11 packets over a network, you would have something that was compliant with the reference implementation. It would be a superset, but the reference specification would be 100% implemented according to the standard, agreed?

Indeed, since the current reference implementation is highly modular, you could replace one module at a time with something that solved the problems inherent in X but which remained 100% backwards compatible.

Let us call this new implementation X12, since it's a stepwise upgrade, similar to (but less crippling than) the upgrade from X10 to X11.

What would I imagine this X12 to look like?

Well, X is still horrible for games, so sprites and shaders make sense. (Nothing stops environments like Gnome or KDE from implementing their own, but to make games viable, you've got to have one API that always works even if you have other APIs for each desktop environment.)

Also on games, but also for multimedia, sound would be good. The challenge is that you want a universal "front end" API where you can switch between engines (such as PulseAudio or Jack) without having to change the code. You'd simply get the characteristics you want. The reasoning there is that different sound systems do have different characteristics and you want a different set for different circumstances. But, again, manufacturers don't care about your freedom to choose, they care about being able to sell to the most people with the least variation in the codebase. No problem. If there's a single universal API that forwards what applies, translates those things that are translatable, and ignores the rest, then the manufacturers are happy and the freedom people are happy. Everyone is happy. That's good.

For scientific and engineering work, you get the best results by converting from vectors to pixels at the last possible moment. Metafont/metapost have a good way to describe shapes (though you'd want to "compile" these descriptions into bytecode for efficiency), and transformation matrices aren't complicated. It would take a bit of work to get the system to work efficiently in 4D, but it would make life a lot easier.

The legacy X11 protocols aren't very efficient OR secure. They're needed because there's so many X11 terminals out there, but X12 should only use X11 to talk to X11 systems. X12 to X12 should be designed from the start to be secure, compact and extensible. It should also be transport-independent. Why should X care if it using TCP, UDP, DCCP, SCTP or something yet to be developed? Or what version of the IP protocol is involved? (If it IS involved! What's wrong with X over Infiniband?) So long as both sides of the connection know what to use, and things that have to be reliable are kept reliable, it's just a port as far as 99% of the code is concerned. Only when it comes to building it in the first place will there be any difference and that can all be hidden in an abstraction library.

Re:XOrg's X11

[kermidge]

Sounds like a good idea to me.

I'd settle for any improvement that'd solve a problem I've had - once or twice a year a major x11 update makes it necessary for me to re-install Crossover, then Steam, then my games. (I've tried abbreviating that process with snapshots, archives, local copies of relevant stuff but that never works. Maybe I'm doing it wrong, or I haven't found the right combination.)

For that matter, I'd be happy if my curiosity didn't drive me to try to understand things I either don't want to learn about - I just want them to work well enough to not get in my way, or that are intrinsically over my head anyway. If Wayland works, and eventually works well enough that it brings me less hassle I'll be pleased. The times I need to remote to a someone's desktop for support (usually Windows) I tend to mostly use Teamviewer - it's simple enough that both of us can use it and it works.

Gnome & deIcaza

[Anonymous Coward]

..are a bunch of people who don't really have a unique idea of their own. They are constantly trying to copy or use some fancy-sounding technology of the day. 100% redundant.

Re:I can only hope...

[Kergan]

And why on Earth was I modded troll?

While posting your ~400 comments, have you never noticed how modders tend to mod pro-Linux/Android/GPL and anti-MS/Apple comments up, and anti-Linux/Android/GPL and pro-MS/Apple down?

Re:I can only hope...

[nurb432]

And why was i modded funny? I was serous. Anyone that is against X11 doesn't fully understand it and only thinks about 'the local desktop' and not the bigger picture.

Sure, it can be improved there is no doubt, but tossing out what it is and what it does is just plain stupid.

Re:I can only hope...

[No2Gates]

It's like X10 only better... Or like the volume control that goes to 11.

1. Thinking 2.Babbling

[Anonymous Coward]

How would your funny contraption handle the problem of immediate feedback ?? HTML interfaces only got ergonomic (as in Google Apps), when they ran Javascript in the local machine and were able to handle every single keystroke and mouse action.

X11 does not have any kind of program-controlled executable code which can be run in the X11 server. So they must send keystrokes and mouse movements quite directly to the program. IIRC programs can control how many events and what type they want to receive.

I could not venture into a criticism of JS and HTML, but I guess it is sufficient to say that there exist many, many better languages than JS. Think of memory-safe Pascal or Ada. (Yeah, I am one of these old farts who was educated by Pascal programming and still loves it over the breadwinning C++, Java, JS shite).

Re:1. Thinking 2.Babbling

[Tablizer]

It doesn't have to be all-or-nothing. A given textbox could be set to "immediate mode" or "delayed mode" or perhaps both (sending but not acting) depending on designer choice, for example. Most of the event activities we use JavaScript for could be declarative; that is, built into the GUI language.

For example, expandable trees/menus have been a staple of desktop GUI's for years and are pretty much standard and expected (tree and slide-down mode), but are not built into HTML, requiring JavaScript.

Re:I can only hope...

[shiftless]

I have a clue. The fool is you. X11 is an antiquated piece of shit.

Re:I can only hope...

[shiftless]

What else IS there besides "the local desktop"? X11 is next to useless over any kind of restricted bandwidth connection, and not really any more useful over a LAN connection. What good does it provide in today's world?

Re:I can only hope...

[nurb432]

Um, you (they) have a right to spout stupid statements but i don't have the right to call you on it? Really?

I suggest you go look up the definition of a hypocrite. Sounds like your picture will be there in all its glory.

Especially For You

[Anonymous Coward]

.. and the other 700 million bozos, the God of Monkeys has thrown Metro on this planet. Metro will come with a full-featured API for Chinese intelligence so that they can download all your sensitive files.

Re:I can only hope...

[icebraining]

But not to have it not modded troll.

Re:I can only hope...

[blade8086]

Plenty useful for me over a LAN .. what do you base this off of?

Oh right - it was useful over 10Mbit lans, but not now with GigE ???

Yes, some applications (which require heavy LOCAL graphics processing) will work better DUH LOCALLY
but this has always been the case.

Sorry. Wayland is a POS

[tlambert]

It needs APIs the hardware people are unwilling/unable to provide and it wants capabilities for the hardware that are out of the hobbyist (used computer) range.

It doesn't address the major EXPORT_SYMBOL_GPL problem that is intended to (unsuccessfully) blackmail the video card vendors into exposing their patent and copyright violations (not the stated intent, but certainly the effect).

And it fails to address the "look and feel is in the app not the window manager" problem.

Bases covered?

Re:Sorry. Wayland is a POS

[Paul Jakma]

Uhm, it's hardware vendors who are paying for the work on Wayland, and the Linux graphics stack generally. Most notably Intel.

Wayland is the bridge to the future

[BalkanBoy]

"Unseating X" is not what Wayland will do, at least not anytime soon. Even X's developers realize that X's architecture has gone a little stale given the current desktop use cases, so they are working to make X a Wayland client. X is likely not going to go away for another 10+ years at least, provided every X app ever developed gets converted into a native Wayland app... And that's a LONG time off on the horizon.

Wayland NAILED it where every attempt to replace X outright prior to it, failed miserably. Wayland is the future only because it allows X programs to run -unmodified-, while at the same time providing a new, more performant window server.

Wayland is the bridge to the future, along with X.

Re:Wayland is the bridge to the future

[jbolden]

The X developers are the Wayland developers. Wayland came out of the people working on X being asked to do things that were too hard or impossible with X.

Re:Wayland is the bridge to the future

[kenorland]

X11 is used in a lot more places than merely the Linux desktop, including embedded devices, super computer apps, etc. I seriously doubt it is going to be replaced for many decades.

desktop over network

[Anonymous Coward]

I'm really worried about being able to run aps over network. I use desktop on my laptop over (unencrypyed) Gb ethernet and it is amazing how well it works. The reason is that my laptop cannot drive the 30" monitor at native resolution. But it can push the data fast enough! You can play flash movies, scroll the web pages really fast, everything works without any noticeable lag at all. I fact, I'm typing this over this setup as we speak.
I know wayland will keep on supporting X, but will the new applications actually work over X? I'm not so sure, i want to use the latest browser *AND* do it over the cable...

UNIX Haters Handbook Chapter 7

[ydrol]

Whenever I see "X11 Window System" I think of the UNIX Haters handbook. A bit dated but well worth a read - even if you like UNIX/linux - http://m.simson.net/ugh.pdf

Re:UNIX Haters Handbook Chapter 7

[unixisc]

I had & have it, and liked how they suggest calling it X-Windows irritates Unix fans

Re:I can only hope...

[fnj]

Your original post was void of any substance, and now you aren't doing your argument any good at all.

X Window System

[HoaryCripple]

I am being a stickler I know, but can we please call it properly the X Window System, X or X11. Conflating the X Window System and X11 into the X11 Window System just sounds silly. Much like X Windows. That is all.

When was W released?

[Anonymous Coward]

It went out as part of Cheriton's V Kernel system and wasn't widely distributed.

Re:I can only hope...

[Anonymous Coward]

Intelligent people with mod points have the right to vote down uninformative, inflammatory frosty piss such as that which has spewed forth from your facecunt. We've heard it a thousand times before and it's still wrong. So, fuck off, you stupid little cunt.

Re:I can only hope...

[Anonymous Coward]

Processing? You really are dumb. An average FullHD video frame is something like 6 MB for RGB24 (this is the X11 format) or 3 MB for YV12 (can lower color fidelity and requires Xv extension hence not true X11). Yes, those are megaBYTES and yes, there's, whole 24-30 of frames per second in case of low framerate videos. That's... 70 MB/s (600 Mb/s) for YV12 and roughly twice that (did we just max out gigE?!!) for RGB24 (what X11 does natively). For reference integrated gigE adapters generally do 400-800 Mb/s bursts with sustained speed being less than that.
tl;dr WHAT PROCESSING, YOU FOOL, THE RAW BITRATE OF UNCOMPRESSED OR LIGHTLY COMPRESSED VIDEO DATA REQUIRES 2 Gb/s BANDWIDTH FOR CURRENT RESOLUTIONS! And something like 4k or even 8k (yes, those are dimensions) is not that far off. In other words, X11 for video has always and will always require more bandwidth that any layman will be able to afford. In real life instead of X11, you would use Samba or NFS if you enjoy your fs lockups and UID/GID syncing.

Re:I can only hope...

[Anonymous Coward]

You haven't actually used X11 over a low-bandwidth connection have you?

I regularly remoted X11 over a 56k modem link (33.6kbps real speed) until late 2010, and it was perfectly functional. I'd rather have been able to use ssh only, but my work relied on me being adaptable so I was. It was perfectly functional. I wouldn't have liked to run FF or GIMP on those machines over connections like that, but keeping eyes on four dozen comms towers over most of QLD doesn't need GIMP either.

MUha

[Anonymous Coward]

You can do all the described things since at least 1993 using X11. It just does not come with dumbed-down GUIs for the ignoramuses who want to waste their time in front of Fuckbook instead of learning some skills which would be useful in *real* jobs. Think of German air space control, which runs on X11 and Motif to the present day.

So For Porn

[Anonymous Coward]

..use static images. It worked for me in the 90s, will work for you today.

Re:I can only hope...

[Vegemeister]

I'm using Firefox over ssh -X on my gigabit LAN right now. It's perfectly functional, and makes my desktop substantially more responsive by freeing up a gig of ram (out of 4) for disk cache, as well as offloading some CPU time.

Re:I can only hope...

[Anonymous Coward]

http://www.tmrepository.com/trademarks/worksforme/
http://www.tmrepository.com/trademarks/thepluralofanecdoteisfact/

Freetard fail.

Wayland, Unity

[kenorland]

If Wayland is to X11 as Unity is to Gnome 2 or KDE 3, it doesn't look like X11 has anything to worry about.

Replacing a mature, working system based on some vague idea of how to do things better rarely works.

You've been mislead - the local machine is simple

[dbIII]

I've been told that if you wanted to genuinely run Gnome or KDE remotely over X, you'd theoretically have to port KDE or Gnome to Windows

You've been bullshited.
X works on a per application basis and if the application can put stuff on a local window it can just as easily put stuff on a remote screen and it doesn't care what is managing the windows at the other end. I've got half a dozen people at work with MS Windows7 and Xwin32 running stuff that is on linux and solaris nodes with no problems, and before that it was XP or win2k and hummingbird exceed or cygwin.
All the X windows software on the MS Windows end has to understand is how to do X. Gnome, KDE, whatever all gets handled if needed by the thing at the other end that is actually running the application in the window you see on the screen.
Back in 2001 I was even running the enlightenment window manager from a linux box and displaying it on a win2k box using cygwin. That was overkill, but it's an example of having something you can use to launch linux applications from your MS Windows box and have it work as if they are all local, while still having the START menu and all your MS Windows applications. Now things like Xwin32 let you seamlessly cut and paste between local and remote applications as if even the non-X applications were X.

Historical question

[__aaltlg1547]

Anybody know why Apple didn't use the X11 GUI in Mac OS X?

Re:Historical question

[unixisc]

OS-X was essentially NEXTSTEP being redone to become the OS upgrade to System 7, IIRC. System 7, like Windows 3.1, was a co-operative multitasking OS, if that, and was seriously behind Windows 95 (which was finally out), not to mention NT, Unix, Linux and so on. Apple had stumbled from project to project, first trying their object oriented OS Pink from their subsidiary Taligent, then they had teams working on their next generation OSs Copland and Gerschwin. Those seemed to take forever, setting Apple even further behind, and finally, they had no choice but to look @ acquiring another company.

One of their options was Be, Inc, and the other was NEXT. They went for NEXT, and got back Steve Jobs. Hitherto, every homegrown Apple OS (except AUX) was independent of Unix - be it Copeland, Pink or whatever. After acquiring NEXT, they decided to make NEXTSTEP the basis on which the new OS would be based. It took them a while though, since they not only overhauled the entire interface, but upgraded several things - Mach 3.0 replaced Mach 2.5 (but the OS did not become a microkernel OS), the userland was upgraded to I think BSD 4.3, DPS was replaced by Quartz, and the UI as well. On top of all this, the whole OS had to be ported from 68040/x86 to the PPC, which at the time was the CPU platform of choice for Apple. This whole thing took a while. To Apple's credit, when they developed all this, they developed an x86 version in parallel, so that when Apple got unhappy w/ the PPC roadmaps of both IBM and Freescale, they jettisoned the PPC for the x86. In fact, Apple has been remarkably CPU agnostic since its PowerMac drive in the 90s - given how it acquired a PPC company and made it an ARM CPU division of its own.

Long story short - since NEXTSTEP never had X Windows in it, neither did OS-X.

Re:Historical question

[Paul Jakma]

Apple didn't have to develop an x86 port. NeXTStep already ran on x86 (wish the person I'd lent my OpenStep 4.2 x86 CDs to had given them back). The majority of the user-space code was already portable, higher-level code. The kernel space code already had x86 ports.

Re:Historical question

[unixisc]

Yeah, but all the changes that Apple made to NEXTSTEP that made it OS-X had to be made on the x86 side as well. I believe it was something they called Yellow Box But yeah, they had to make all those changes, AND move them from a 68040 to a PPC - I think it was a G3 at the time.

Re:Historical question

[__aaltlg1547]

I don't see the relevance of the hardware support since NEXTSTEP had to be ported to PPC, and both it and X11 had been implemented already on x86. Getting X11 to work on PPC couldn't have been too much of a problem because Apple made X11 available on their OS X machines too, but it wasn't used as the primary GUI or even installed by default.

And...

[Anonymous Coward]

Arkanoid for Mac OS turned 24 yesterday! Happy Birthday!

Re:I can only hope...

[Anonymous Coward]

You shouldn't put TLDR in middle of the text, as the reader obviously has read up to that point. But you could have replaced it with "to put it briefly", for example.

Re:I can only hope...

[lister king of smeg]

uh why the hell would you try to stream raw hd video with a tool meant for remote desktop management? Your argument doesn't hold. and you are very right about one thing you would use a file or streaming media server for serving up video, but you would mange it with X

Re:I can only hope...

[Vegemeister]

Parent asserted that X11 network transparency was not useful on a LAN, and does not provide any benefit in todays world. I, in fact, use X11 on my LAN and derive substantial benefit from it. A single counterexample suffices to disprove such a sweeping statement.

Freetard fail.

I hadn't realized Wayland was proprietary software.

Wayland vs X11 conflicts?

[unixisc]

One question - if a system has Wayland, and also has X11 services, how does it know to allow the Wayland compositing and disallow X11? In other words, how does it prioritize some parts of X11, but not the others?

BSDs and Wayland

[unixisc]

FreeBSD currently has no plans to move to Wayland, and I suspect the same is true about the other BSDs. Since most of the BSDs target the server, there is nothing wrong w/ that approach. I do hope that PC-BSD, which normally derives everything from its parent project FreeBSD, makes this one of the exceptions (like it's done for a few others, like PBI) and implements Wayland on PC-BSD. On top of that, they can have Qt5 or GTK+ or whatever else there is. I don't blame the BSD guys for not being interested, but this could be a PC-BSD only thing, fwiw.

Re:BSDs and Wayland

[jbolden]

I don't think they'll have a choice. But say 2025 yes I expect all the BSDs to be getting their X11 on Wayland and not having it run native on the video card.

Re:I can only hope...

[Anonymous Coward]

No, X11 is not remote desktop management. X11 is the windowing system and it's whole purpose is to provide windowed GUI be it remote or local (which is one and that same for original X11). Running MPlayer, VLC or even just Firefox with Flash/HTML5 Youtube videos over your network is what it was meant to do. And it's failing because of bad design. And for good measure the horrible protocol is only half of the pain. The other half is that any X11 implementation has to support features no one cares for or wants, such as font rendering (now done by Freetype/Pango/etc) or primitive drawing (usually Cairo) - yes, those cute buttons you click every day are almost certainly drawn by Cairo and X11 only receives the bitmap. This also means that for a modern toolkit X11 is awfully lot like advanced VNC. And because you can't detach and reattach X clients, when X11 goes down, so do the applications and because X11 was configured exclusively before startup they had to spend decades comming up with extensions that didn't work with stuff like xmove or xpra and were PITA in evey possible way to make X11 not suck tarred and feathered monkey balls. And we still suffer from poor networking capabilties (what if I wanted to control This session from another computer?), X11 crashes killing clients and suboptimal performance because of the overhead of the shitty protocol. And I can only feel sorry for the people who spent years fixing X11 unbecommings - did you even know that X11 could not change resolution or pixel format after startup (something Windows 9x could)? And that was less than 10 years ago.

Wayland and DEs

[unixisc]

Isn't KDE 5 supposed to support Wayland? But I doubt they'll be Wayland only anytime soon, since there will be Unixes that may still use X11 (such as OpenBSD)

Re:Wayland and DEs

[jbolden]

Yes support Wayland. That's why I said 6. I'm assuming by KDE 6 / Gnome 5 they will be ready to drop X11 in exchange for the speed advantages of Wayland.

NEXTSTEP and NeWS

[unixisc]

The NeXT boxes were always based on a Motorola 68040 - that was the problem. By the time NEXTSTEP was ported to both the PA-RISC and the SPARC, it was somewhat late - not long after that, NEXT got acquired by Apple. B'cos otherwise NEXTSTEP on PA-RISC, or OpenStep on Solaris would have been a great OS. Instead, Solaris took on GNOME and let go an opportunity.

One thing I've wondered - when SunOS was around and included NeWS, did it include X Windows? I thought that NeWS and Display Postscript were alternatives to X.

Re:I can only hope...

[John Allsup]

The bigger picture is that the non-local desktop should not be achieved at the foundational level of the windowing system, but as an application layer. Not all applications are feasible over a network, so network transparency should only be used for those applications that need it and, more importantly, the architecture of the desktop should not be compromised in the name of network-transparency.

Re:I can only hope...

[nurb432]

Not a troll, just being honest. When i see stupidity i point it out. With luck they realize they are stupid and go back and learn something. Stupidity can be cured.

Hey Miguel

[Anonymous Coward]

..just because you are a retard and don't know how to use X11 properly does not mean everybody else is equally stupid.

I guess

[Anonymous Coward]

..it is a paid M$ FUDer.

http://en.wikipedia.org/wiki/Fear,_uncertainty_and_doubt

Not true if done properly

[Anonymous Coward]

First, you can still use X11 fonts. Secondly, you can store bitmaps in the X server and reuse them by addressing them by handle. But yeah, Mr Icaza probably does it as if it were RDP.

Just demonstrates that the M$ shills are either idiots or they want to destroy the Unix community by dropping their crap onto it. Too bad we can live without Gnome and KDE quite well.

Microshackle feels threatened

[Anonymous Coward]

That is why they send their propaganda troups around to drop excrement. M$ cannot live with competition, they think they must destroy it at all cost. Their mindset is unipolar.

Unfortunately for them, there are people around who will "shit back" in an eloquent and argument-supported way.

Ok, Boy

[Anonymous Coward]

Maybe you are sitting in a cabin in north Dakota and all you have to connect to the interwebs is a narrowband HF link to Buttfuck City, ND. Buttfuck has fiber, luckily. With narrowband I mean 4800 baud. You can use many X11 applications over 4800 baud, but certainly not firefox to check your facebook contacts.

What you can do over that 4800 baud link is to fire and use up vi, emacs and maybe some well-written X11 application of your employer.

Some people actually venture out of their bedroom and go to places where they only have a narrowband link into the internet. They still need to perform some important function in their employer's system. Think of a soldier in a remote place who has some very special knowledge about a very special system. He is not in North Dakota, but in Crackistan, listening on a deserted CIA asset who plans to blow up some of his former masters. But he needs to log into some special system in Ft. Meade or Hanscom.

Finally, saving bandwidth is always an excellent idea if you want to ensure as much stealthiness for your communications link as it is possinble. This comes from a law of a Mr Shannon, who stated that channel bandwidth is log2-proportional to the signal-to-noise ratio. Strong signals mean easy interception for the bad guys. So, setting up highspeed links is a good way to announce you are there and inviting the other guys to blow you up.

That means some people actually like thrifty protocols and they hate wasting bandwitdth, memory and CPU cycles "because we can".

Too much shelf space!

[Anonymous Coward]

My only complaint with X11, is that the manual set takes up about 2 linear feet of my book shelf! If anyone knows of an electronic (not online) version of the books, I would appreciate a link to them! :-) FWIW, I have both the first and second editions, so that makes up for almost 4 linear feet of shelf...

Direct Rendering sucks

[Anonymous Coward]

Direct Rendering is a BAD idea for an engineering environment, simply because you can't send the indirect buffer or prevent hardware attacks.

I'm quite happy to accept the slight overhead that the protection and indirect rendering benefits (like high-performance VNC and switching rendering targets to different hardware on the fly) require.

Is it still a mess?

[swflint]

Is the code still the mess it always was?