Wayland 1.4 Released — Touch, Sub-Surface Protocol, Crop/Scale Support 128
An anonymous reader writes "Version 1.4 of the Wayland protocol and Weston reference compositor have been released. The Wayland/Weston 1.4 release delivers on many features and includes promoting the sub-surface protocol to official Wayland, improved touch screen support, a crop/scale protocol within Weston, security improvements, and other fixes."
OMG NO NETWORK TRANPARENCY!!!1 (Score:5, Funny)
Just to preclude about half of the coming threads.
Re: (Score:3, Informative)
And the best thing is that X is no longer network transparent, it's network capable in a way similar to vnc. With the current rendering methods (shm and dri2) you no longer send commands, you send image buffers.
Here, listen what a X developer has to say about that http://www.youtube.com/watch?v=RIctzAQOe44#t=1111 [youtube.com]
Defining new primitives at runtime (Score:2)
the [team] behind wayland are the exact same developers who [messed] up X11 by trying to move it to sending image buffers instead of primatives
It was either that or include a virtual machine of some sort in the X server so that new primitives could be defined, much as HTML5 does with JavaScript and the 2D canvas API. How would you have preferred to implement that? At least image buffers are slightly less likely to cause a security hole than a Lua, Forth, Java, JavaScript, or whatever VM.
Re: (Score:1)
Bollox, the developers of X11 are over 30years older now. I think you are confusing the current maintainers of the two most popular X11 implementations with the actual developers who came up with the original ideas. The extensions over the past 15 year rise of Linux popularity have had to restrict themselves to the design choices made over 30 years so. It is plenty overdue a revamp, silicon has changed to much in that time.
Re:OMG NO NETWORK TRANPARENCY!!!1 (Score:5, Insightful)
All I want is to ssh -C -X and open remote apps and use them like local ones, seamless cutting and pasting. It is quite handy in a LAN. Remote desktop preserving state is useful for monitoring, this is useful for office work. Different scenarios.
So, the race now is between new faster compositors who need X protocols layered for compatibility and features and javascript obfuscated apps replacing networked native applications...
Re: (Score:1)
Sigh, this gets old. I use it everyday and it works! So yes, it is network transparent.
Re: (Score:1)
How is it a complaint when I say that it works for me? The precise technical explanation is that direct rendering does not work over the network. For 99% of all applications this is irrelevant.
Re: (Score:2, Troll)
> And the best thing is that X is no longer network transparent, it's network capable in a way similar to vnc
Then why is VNC so god-damned awful in comparison? Even on a fast local wired network VNC sucks great big donkey balls. VNC on a local network is less usable than either RDP or X (with compression) are across the Internet.
The disaster that is VNC on MacOS is why I find the idea of Wayland so repugnant.
Re: (Score:2)
VNC is highly dependent on how clever/fast your image scraping and compression is. The TigerVNC client connecting over a LAN to a TightVNC windows server is so much faster then some of the other clients its unbelievable. Even over the net it's much quicker then others (TemaViewer's VNC client is terrible).
The problem is very much "how quickly can you get a copy of the pixels" "how quickly can you compress them" and "how many screen sections do you need to update".
The slightly disappointing thing (or thing I
Re: (Score:2)
Follow up: And it looks like sub-surface support is exactly what I was just complaining about - a way to update small portions of your window and tell the compositor that's what you're doing. Make your editing windows subsurfaces and only send the necessary updates.
Re: (Score:2)
I have NEVER seen an X app that wouldn't run just fine over the net and neatly appear on my desktop as if it were running there. Can you name one?
Unlike vnc, it doesn't make me display the whole desktop in a window and then the app window within that. It also doesn't require an X server on the remote side. If the app uses a tray icon, that works normally as well.
It's trivially easy to do all of that through sh as well for ease and security. When Wayland is up to that, give me a call.
Re: (Score:2)
I have NEVER seen an X app that wouldn't run just fine over the net and neatly appear on my desktop as if it were running there. Can you name one?
Anything which uses OpenGL has about a 50:50 chance of working at all, and an even lower chance of working correctly. The protocol allows for it, but it just doesn't work most of the time.
Re: (Score:2)
Re: (Score:2)
Well, I've been trying it since the late 90s and it's been seriously hit-or-miss, even on mainstream hardware.
Re: (Score:2)
Re: (Score:2)
I have also seen some issues there, but in every case, the app wouldn't run locally either. That is, the network transparency part wasn't the issue.
Re: (Score:2)
Oh dear.
Look, I don't care whether or not X technically has pure network transparency any more. All I and just about every other multi-system-Linux user want is to ssh to another computer and have individual programs launched from that shell show up on my screen as if I was sitting at the remote computer.
If Wayland does that then there's no problem.
X does this right now, though admittedly these days addons like xpra are needed to make it usable over slow links.
Re: (Score:1)
As long as ssh -X doesn't work, it's a no-go for Wayland. I don't care HOW it is implemented though!
I'm cautiously optimistic, but not ready yet (Score:4, Interesting)
The wayland repository continues to mature and moves slowly. This cycle again only saw a few wayland changes, most of which where fairly unexciting:
- SHM Buffer SIBGUS protection. We added and couple of utility functions to help compositors guard against broken or malicious clients who could truncate the backing file for shm buffers and thus trigger SIGBUS in the compositor (Neil Roberts).
- Subsurfaces protocol moved to wayland repo and as such promoted to official wayland protocol (Pekka Paalanen).
- wl_proxy_set_queue() can take a NULL queue to reset back to default queue. (Neil Roberts).
- A few bug fixes, in particular, I'd like highlight the fix for the race between wl_proxy_create() and wl_proxy_marshal().
- A few scanner error message improvements and documentation tweaks and polish.
I'm hoping the Maui Project [maui-project.org] (which uses Wayland [slashdot.org]) can continue to gain momentum as Wayland does and that it becomes a viable option in the next few years.
Re: (Score:2)
Re: (Score:2)
Re:Wayland! (Score:5, Funny)
So...there's that.
Re: (Score:3)
Today's moderators don't seem to have a clue, the above post is not funny but deeply insightful.
Re: (Score:2)
Re: (Score:2)
I am pretty sure he will slip a systemd dependency into wayland somehow. Wayland is developed by xorg developers and will eventually replace xorg, so yes. :-)
Shut the fuck up (Score:1)
Please.
It's VNC-like, not VNC. VNC sucks donkey balls, but that's because it's done in an absolutely insane way where it actually does continuous screen-grabs (of the whole screen), encodes them as JPGs and then sends them over the wire.
The wayland version of this is much, much less braindead.
Re: (Score:2)
VNC sucks donkey balls, but that's because it's done in an absolutely insane way where it actually does continuous screen-grabs (of the whole screen), encodes them as JPGs and then sends them over the wire.
No, it doesn', unless you have a really, really braindead implementation which doesn't track which pixels have changedt. But don't let not knowing what you're talking about interfere with your rant.
X's way is the only feasible way. (Score:1, Insightful)
You're right on the mark. This is why the Wayland attempt is going to fail. They're going to do what the VNC crowd has already tried, and they're going to fail in exactly the same way. There's only one correct way to do this, and that's the way that X did it. Either you do it the same way as X and succeed, or you do it any other way and fail miserably.
Re:X's way is the only feasible way. (Score:5, Informative)
The same way as X did? You mean produce a hell old protocol which was network transparent yet utterly incompatible with SHM requiring every modern distribution to have the compositor render the frame and then send it over the network as a bitmap? Wait what? I just described VNC.
Actually no I didn't. VNC uses compression, X doesn't. Hence remote X running on any system produced in the past 20 years actually holds the crown as the only system SLOWER than VNC over the network. It's only saving grace is that you can send individual apps and don't need to export the entire desktop. Then there's the underlying problem of a protocol which forces a shitload of talking to the local server before even being capable of sending a bitmap to the remote one.
As one X developer described it, it is not possible to do remote desktop any worse than X.
Re: (Score:2)
Re: (Score:2)
Hence remote X running on any system produced in the past 20 years actually holds the crown as the only system SLOWER than VNC over the network.
It's only a problem if you're sending bitmaps back and forth across the network, whereas other drawing primitives are fast. The only problem is that too many half-competent developers have decided that the only way to do things was to draw on local bitmaps and then push them to the display (possibly with some extra fetching the display buffer back in the other direction a few times too, just for fun). That's just never going to be all that fast on a realistic network due to the latency involved.
The proper f
Re:X's way is the only feasible way. (Score:4)
As one X developer described it, it is not possible to do remote desktop any worse than X.
Well this is the problem and why Wayland is greeted with such suspicion.
The claim is "oh it's better than X because it's being done by X developers", and then you have claims like that which are disingenuous bordering on an outright lie.
Just about everything does remote windowing worse than X. Even if you accept that X sucks, everything else sucks harder. The thing that seems to be by far the best is NX, and that's basically juiced X, not X with all the bits removed.
Re: (Score:2)
Re: (Score:3)
Yeah, you guys are using the wrong version of VNC / RealVNC.
Look up tightVNC or even tigerVNC if you need something fast enough to do 3D graphics and maybe even full motion video at a few FPS.
Re:VNC "works fine"? What the fuck? (Score:5, Interesting)
> Wayland probably won't be able to perfectly replicate your perfect 1992 X experience, because nobody (who isn't the kind of fucking retard who runs shit remotely just to make themselves feel like one computer isn't enough power for them) gives a shit about doing that kind of thing any more.
What hole have you been hiding in?
Remote desktops are all the rage now. They are very common in corporations and even "regular people" are using them.
The rest of the world has caught up to the 1992 X experience. Now clueless nitwits want to set us back 30 years because they think it's trendy or some such.
Re: (Score:1)
Remote desktops are a stop-gap for people using shitty apps that don't have any better kind of network functionality designed in, or people trying to uninstall browser toolbars off gramma's old pc. Nobody *wants* to use them.
Just like nobody wants to go back to shitty-looking Motif desktops because they're fucking eyesores. But no, we should bin off pointless frivolities such as anti-aliased fonts and go back to 8-bit colour, because some guys somewhere need to visualise petabytes of CFD data from outdated
Re: (Score:1)
Yeah a retard who is trying to figure out why an aircraft wing starts fluttering via a CFD simulation on TACC's Stampede or maybe trying to determine where the asbestos fibers from explosion in the densest part of a major US city are going to disperse to via a dispersion model on NCAR's Yellowstone. Yeah I'm some "kind of fucking retard who runs shit remotely just to make themselves feel like one computer isn't enough power for them", because even with 72286 cores my jobs still take hours to run and the out
Re: (Score:1)
Interesting (not). I disagree with GP on many things, most having to do with VNC as I've used (Tight)VNC for many purposes and have achieved nearly same level of usability over 256K line as I have locally (not for any kind of animated/video context like games, video, etc. but for browsing web for articles, email, coding, etc.) but I also use X forwarding for several reasons and all I can see in your post is a counter argument that isn't and never has been a thing. It's just foul mannered nothing not worth t
Re: (Score:1)
I don't want a remote desktop, I want a single app forwarded. Some of the machines I forward X from don't even have X itself or the desktop installed, just librarues and an app or 2. On the few occasions I want the whole desktop, I use VNC just fine.
As for windows, since when did it work fine?
Re:X forwarding-like feature? (Score:5, Informative)
Yes. It's on the road map.
Not Wayland, but Weston (Score:3)
Wayland as a replacement for the X protocol does not define a lot of the functionality of X but rather puts it to the client. It's up to the Wayland client implementation to define things like remote display.
Weston has two network backends in its code base which are being worked on. One is their own implementation of something similar to VNC much the same way X currently does remote rendering but with the advantage of compression, and the other is an implementation of FreeRDP.
In any case it's not's Wayland
Re: (Score:2)
There was a SPICE backend as well, which also sounded interesting but I don't know what the status of that was.
Re: (Score:2)
So if we'll just shut up and jump, they pinky swear that they'll meet us half way down with a parachute?
Re: (Score:2)
Yes.
But you don't have to jump. Whoever creates your distribution will jump in your stead.
Re: (Score:2)
No distro I use will do that.
If the underlying toolkits the apps are built on will work with either Wayland or X without a re-compile, that would be somewhat acceptable but a bit bloated.
So... what is it? (Score:2)
Re: (Score:3)
Yeah, I'm mostly with you - though this has been bubbling on /. for ... a couple of years, now? Still - you're right and the editors suck.
Wayland is an alternative display system that could maybe someday replace X.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:2)
Re:So... what is it? (Score:5, Informative)
Some guys who worked on Xorg/X11 for years are redesigning it to be better. They got some good ideas but it will be a while before it can actually replace Xorg/X11. Here is something: http://www.youtube.com/watch?v... [youtube.com]
Re:So... what is it? (Score:5, Funny)
Some guys who worked on Xorg/X11 for years are redesigning it to be better.
Yes.
'Xorg sucks, but this new interface it will be much better. Trust us! We wrote Xorg!'
Re: (Score:2, Informative)
How about:
"For 20 years we've been developing the code that allowed _actual_ applications to do what they do while being shackled by this protocol dreamed up by people who , for the obvious reason, didn't have a clue where things were going. Now we think we have a _pretty good_ idea of how this should work, unlike the aforementioned, and the hordes of muppets on the internet, so we're doing it. But no one's forcing you to use it. You're more than welcome to get coding on x.org or xfree instead of whining on
Re: (Score:3)
No one wrote X-org. It was a fork of XFree implementation of X11. They just removed some 500k of worthless lines of code from XFree and that became X-Org.
The intent was to try and make do, it was not to fix the underlying problem.
If you go to a chef and give him a cake made entirely of cow dung and asked them to make it taste better with the restriction that he can only use dung from an animal, how do you think the cake will turn out?
The cake is XFree86, the animal dung was added to make X-Org, and the requ
Re: (Score:1)
> it couldn't possibly be any worse than X.
But it can be worse. It can do less and will have no decent device driver support.
The developer priesthood needs to venture forth from it's echo chamber once in awhile and actually observe real end users.
Re: (Score:2)
By doing less it reduces attack surface and increases maintainability. No more wondering why Xorg won't start because it can't find some useless raster font. Everything is done in toolkits that render to buffers these days.
The following quote from the Wayland FAQ is entirely true:
Wayland uses EGL and GLES2, which means that any driver that exports those (virtually all of them, given th
Re: (Score:2)
void main() {} is pretty reliable too. No exploits, no mysteries. It does nothing every time, just like it's supposed to.
But it will never be a popular productivity tool.
Re: (Score:2)
Xorg actually runs in production today. I'm using it right now.
Re: (Score:2)
Right now think back to what XOrg is.
XOrg is XFree86 with some 600000 lines of code removed. Did you notice? Do you know they wanted to remove more and change things as well but couldn't because of the arse backwards (in the modern GUI sense anyway) way the protocol is defined?
You sir need to stop literally interpreting comments. Not being worse or doing less does not mean that the end game isn't a fully functioning windowing system for Linux. I hope we talk again in a few years where we can list the many t
Re: (Score:2)
All I ever hear when I ask if Wayland will support a useful feature is NO and you will like it because we're gonna cram it down your throat.
It should be no surprise when I say no.
I'm all for trimming the fat, but so far what I see here is wholesale amputation. Chopping your legs off is not my idea of a sensible weight loss plan even if it technically meets the goal.
Meanwhile, I find X working just fine right now, and I don't find it to be slow.
Re: (Score:2)
I'm all for trimming the fat, but so far what I see here is wholesale amputation.
They aren't trimming fat. They are in bed copulating and making something new. It'll take many months before it's ready and likely several years to get good.
THAT'S WHY THEY ADDED THE COMPATIBILITY LAYER.
You want to keep using X, do, that doesn't stop you using Wayland, and that doesn't stop Wayland developers still working on it.
Meanwhile, I find X working just fine right now, and I don't find it to be slow.
This is a case of not knowing what you're missing? X may not seem slow, but when you put through a debugger to see what it is actually doing there are a LOT of points where X delays
Re: (Score:2)
If the developers would focus on making sure the resulting system has the functionality that users indicate is necessary (such as simple network transparency) rather than blowing it off or claiming someone else might somehow figure out a way to kluge it in later, most of the objections would go away.
I don't want assurances that someone else might solve the problem. That's as good as saying fuck you.
Put simply, unless and until Wayland can actually demonstrate remote display working as seamlessly as X runnin
Re: (Score:2)
Regarding the network transparency that's not what they said at all. All they said is that network transparency is a client problem and not a function of the Wayland protocol. This is consistent with the entire philosophy of designing a display server protocol that defines as little as possible to make it the most flexible protocol possible. Defining every detail at the protocol level got X in the shit it's in today. Wayland gives the client the choice, and believe it or not Weston has two different remote
Re: (Score:2)
Sorry, failed to meet my bare minimum requirements BY DESIGN, so it will never see any use here.
All you did was rephrase, but it still comes down to exactly what I said, I just stripped the candy coating off of it.
If the developers actually gave a damn about any eventual community that might use Wayland, they'ed burn that transparency in as a requirement for a client end enforce it. At the very least, they shouldn't try to pass off a vague possibility as if it were an assurance (as I have seen done here on
Re: (Score:2)
But has it? When have you in your day to day life actually cared what the underlying protocol of your software is? Do you say man I wish I could play Halflife multiplayer but UDP so failed by design?
Putting anything in a protocol is EXACTLY what got us into the mess that we're in. Funny story, X requires so much intercompatibility within the protocol that you can actually run GLX Gears using a printer as the output device. Not even slightly practical for the GUI end and a major fuckaround for the printer si
Re: (Score:2)
What your missing is that practically everyone involved in or advocating for Wayland has been blustering on about how it doesn't include network transparency as even a design consideration and that they intend to force it on people (just look at the quality of advocacy right here in this story). They practically demanded opposition from nearly everyone and so that's what they've got. They poo-poo me so I poo-poo them. That's how it works.
When they got pushback, they hand waved about how it might not be enti
Re: (Score:2)
But no one is hanging a sign on you, unless you're developing a client in which case yes you're getting a sign hung on you saying kick me.
This all comes as part of changing the low level specs of the protocol. Yes feelings will get hurt, but as an end user (power user at that) why should I care whether network capabilities are provided by Wayland or the client? Wayland has been clear on it's purpose and what will be defined in the protocol, they never said that any of this will impact the end user or that t
Re: (Score:2)
But no one is hanging a sign on you, unless you're developing a client in which case yes you're getting a sign hung on you saying kick me.
Of course not, the Wayland developers have hung the sign on Wayland. They practically beg for people to want it to fail with their attitude towards completely reasonable user use cases.
You should care about the protocol. ALL X Apps work with network transparency because otherwise they don't work at all. That is actually a very helpful trait.
You clearly did not take my suggestion of looking at the attitude projected all over this discussion. You clearly didn't actually look at the crazy claims that X doesn't
Re: (Score:2)
I think the core problem here is that people are overly attached to something they a) don't understand, and b) shouldn't care about. That's where I think the attitudes come from. I as a user don't care that my Slashdot session uses TCP or that I have an ethernet cable in the back of my computer. All I care about is that Slashdot comes up. Much the same way I don't care how my remote application is displayed as long as it works and is easy to achieve.
To use a somewhat crap analogy can you imagine where we'd
NO! That's misleading (Score:2)
What did happen is a guy who wrote an extension to X recently decided he'd do his own project which differs from X in many ways.
It's really about putting stuff into a framebuffer instead of the more complex X framework. That pushed a lot more complexity back onto the writers of the applications but th
Re: (Score:2)
Err, I don't know why you put it in such thin terms. Qt has pretty thoroughly integrated support for Wayland via QtWayland [qt-project.org], which lets you write your own Wayland compositor using the QtCompositor class.
Architectures like Wayland directly benefit toolkits like Qt because it directly services what Qt was doing already: rendering in a buffer and displaying it.
Re: (Score:2)
Because there is still some way to go before there is a fully functional Wayland environment.
I tried to go for maximum bland to avoid pissing off the thin skinned but it appears that even faint praise is taken as some sort of attack. Am I supposed to wave my arms and shout "X sux - go Wayland!" like the mindless fanboys?
Re: (Score:2)
There's already fully functional Wayland environments, it's shipping on at least one vehicle IVI system and Jolla's handset. What's missing is distro adoption, but even that's inevitable. Far from the "hope" that toolkits will take up the slack -- they already have.
Are you so sensitive that you take simple replies as attacks?
Re: (Score:2)
Links please for that shipping system. I find it difficult to believe that there has been that much progress in a couple of months and annoying fanboys have been caught out with lies about Wayland here before, so I can't take you at your word until I know you are more than that.
Re: (Score:2)
http://jolla.com/
It's been shipping since the December 2013, using Wayland and Qt5. No X compatibility layer on this device.
That's lazy (Score:2)
Re: (Score:2)
You asked links for a shipping system.
That link has a "BUY" button.
Re: (Score:1)
Re: (Score:2)
Having fun making me run around you horrible little troll? Why the fuck are Wayland fanboys such pieces of shit? Why can't they let the project stand on it's own merit and no resort to silly little tricks to advocate it.
I'm giving the actual developers the benefit of the doubt but what IS it with your people?
Re: (Score:2)
No, you are misleading.
If you take a look at Wayland source code, you'll see stuff like Copyright © 1988-2004 Keith Packard and Bart Massey. quite often.
https://gitorious.org/wayland/wayland/source/0b29a2fec7801d2530bd004ae68eb9242417bafd:wayland/wayland-hash.c#L2-3
As for pushing back work to the toolkit developers, the Qt developers made the software (client side) backend the default back in Qt 4.4, because it was so much faster than the XRender based one for local clients.
And for Qt5, they simply did
So Wayland was started in 2004? (Score:2)
If you are going to lie then try something a bit less obvious.
How the hell did this Wayland project turn into such a ball of hate against X where people decided that any dirty trick in advocacy goes?
Re: (Score:2)
Re: (Score:1)
X will certainly not be gone. Maybe there will be a schism in the Linux community. This might actually not be bad thing. The freedesktop crowd who break my desktop (or some applications) basically every year (I think this started in 2008 or so) can go on with their misguided attempts to redesign everything over and over again in different ways. And people who want a UNIX-like system with backwards compatibility, stability, configurability, and powerful features could have there own distribution.... This wou
Re: (Score:1)
I am perfectly fine with re-designing things. I am not happy about breaking stuff and especially breaking backwards compatibility. The kernel does not do it, libc does not (usually) do it, the C compiler does not do it, network protocols do not do this. Why do the freedesktop/GUI/graphics people do it all the time?
Re: (Score:3, Insightful)
Wayland's an effort to stuff a pointless layer of abstraction underneath X on Linux in order to make performance worse and debugging more difficult.
(Yes, yes, they say it's an effort to replace X. But look at how they're doing the compatibility with X - running a full X server on Wayland in order to run X apps. Then look at how much effort they've put into making Wayland portable to other varieties of *nix.)
Re: (Score:2)
There's some hope there. Initially by design it wouldn't work on anything other than linux but the IMHO braindead choices of depending on linux only features were changed.
Re: (Score:2)
(Yes, yes, they say it's an effort to replace X. But look at how they're doing the compatibility with X - running a full X server on Wayland in order to run X apps. Then look at how much effort they've put into making Wayland portable to other varieties of *nix.)
Yeah I know. Damn that world that won't make a wholesale switch of an entire protocol at the drop of a dime. I was like you suggesting that the entire world should switch to IPv6 overnight and just throw all the IPv4 stuff in the bin. Who needs transitioning periods, or compatibility layers. This is survival of the fittest we're talking about here.
*faceplam*
Re: (Score:2)
Re: (Score:2)
You won't need X if you use QT or GTK3
Yes, I know. Reminds me of Micrografx Mirrors and WLO.
Re: (Score:2)
And for some reason, just like with Motif, they think doing that is a better idea than something flexible.
Re: (Score:2)
...and here is another problem with this kind of half-baked nonsense.
Linux benefits from being just another Linux. It may not be "certified" but it is close enough to the other real Unixen that it can be treated as one of the fold.
Nonsense like this just widens the distance between Linux and other Unix.
If I wanted Apple style nonsense, I would use a Mac.
There is only one paste buffer (Score:3)