Slashdot Mirror
Ubuntu Still Aims For Wayland in Quantal Quetzal
jones_supa writes "While there's still more than one month until the Ubuntu 12.10 feature freeze, Ubuntu developers continue to work towards their tight schedule of having Wayland serve as the compositor for the Quantal Quetzal release due out in October. Canonical's intends to provide smooth transitions from boot to shutdown. Wayland is also used for session switching and other operations, avoiding traditional VT switching, providing a consistent monitor layout, using the greeter as the lock screen, ensuring that locked sessions are actually secure from displaying, and showing the greeter while the session loads. Phoronix remains skeptical about Ubuntu making the deadline."
but i'm still waiting for x to start...
Ubuntu is sparking a new 'unix war', dividing the linux ecosystem. It first pushed hard for its own Unity now with Wayland that breaks all current X apps. Theyr'e only in it for themselves.
Without X we will lose network transparency among many other great features. Let's not even mention the lack of gpu support to say the least.
Who in their right mind would trust the Weyland-Yutani Corporation given what they did to the colonists on LV-426?
Is there anyone who still thinks the Ubuntu release team prioritizes either usability or low number of bugs now? That's a serious question; I have no idea why this is considered a novel or even notable thing at this point. New Ubuntu release, leading edge software that's not ready for prime time is included, the release is at best beta quality software by any reasonable standard. Same story in every release going back to at least the 8.04 PulseAudio debacle.
I read the article and the Ubuntu Wayland wiki. The Ubuntu developers have not set any deadline, they don't appear to have set a specific goal, they're just continuing their work on Wayland as usual. The article appears to have just pulled the release target out of the author's arse and then claimed the developers won't make it in time. From the linked page:
"When will Wayland become the default on Ubuntu?
This has not been decided. This decision will be made at a future Ubuntu Developer Summit (UDS)."
So apparently the developers won't reach their goal that they haven't set. How is this a story?
Although the more detailed answer is that GTK/Qt apps will need recompiling with an updated library. If you use X11 directly, then you have more work - except that you can also run an X server within Wayland to support native X11 clients.
I've been impressed for some time with how well the Wayland developers have thought about backwards compatibility. X11 needs replacing; complaining that its different to X11 doesn't help solve the problems.
Slashdot - News for Nerds, Stuff that Matters, in ISO-8859-1 Has just realised that beta makes this signature redundant
That would be sad.
But beyond that: there's one big deficiency of current X applications--if the X server dies, so do all graphical programs.
That's quite surprising when you think about it.
After all the graphical programs are X clients. Why would a client up and die just because some server died?
Does your browser die when a webserver dies?
And please, no pretending that X on Linux doesn't crash. It does, and this is the 4th time I've restarted this laptop today. Hanging hard with VirtualBox.
To sum: If the graphics server crashes, I'd like to see it automatically restart with Upstart, and then the clients automatically reconnect.
I'm not a lawyer, but I play one on the Internet. Blog
I have no comment on 'wayland good' versus 'wayland bad', but it is certainly a known quantity and it has been waiting in the wings to replace X for a while.
As you said, the dream of replacing X has been around for a very long time. No one has done it because it is a ton of work and because the changes are bound to piss off some developers and users (see "pound sand" post above).
I'm impressed that the Ubuntu folks are going to take this on and I wish them well. If they succeed then the rest of the distro's will probably follow suite.
I agree. Wayland needs to happen. We've wasted so much time, effort, and knowledge keeping X Windows up and running. I sometimes think about how much further along desktop Linux might be if we threw off the shackles of X Windows years ago. There's a reason Android and MacOS do not use X Windows.
Kriston
Canonical has plans to include Wayland as a technical preview in 12.10, not as a replacement for X11. This means that they have to actually get it working at a basic level before putting it in the repositories. While Canonical is pushing Wayland, they've already said that it's still several years away from becoming a viable replacement for X11. This is just Canonical trying to push forward the development of a peice of software they believe in.
The Ubuntu folks aren't developing Wayland. Wayland is being developed pretty much entirely by Intel employees. Ubuntu is just planning to integrate it.
Sure. X11 needs replacing because its hardware acceleration is worse than Wayland. X11 is actually many programs acting like layers: window mgr, compositor, greeter, locker, etc will be only 1 under Wayland. And still the design is simpler than any one. It's not that the code is being added, but that a no-longer-useful X11-style abstraction is being let-go.
Science & open-source build trust from peer review. Learn systems you can trust.
It will work if SomeProgram is an X program, just as it does on OS X or Windows if you have an X server installed. It won't work if SomeProgram is a Wayland program. Wayland eliminates a number of process boundaries in X, moving the window and compositing managers into the main executable. This is done for performance reasons, presumably by people who have never profiled an X server and therefore not noticed that these round trips are not a bottleneck in modern systems, and at the expense of stability (if your compositing manager crashes in Wayland, your display server also dies, with X11 it can be restarted, usually without any data loss).
I am TheRaven on Soylent News
X11 needs to be replaced? Like my roof needs to be replaced because it works and have no holes in it? But it is old, I think about 15 years, so yeah it needs to be replaced.
X11 is old, but it works. I can have multiple monitors just fine, 3D effects are working fast, games are working. Also, it is supported by all hardware vendors like AMD, Nvidia, Intel. Also I have network transparency with no additional costs. Multi-User works just fine with very easy Ctrl+Alt+F1 up to F12. I don't know why you want something else. In KDE there is also a user switcher GUI way. Boot is smooth in Fedora 14, 15 and 16.
As long as there are no drivers from AMD, NVidia and Intel, Wayland will be a wet dream of a few developers. I do not need to go back to the time where the only graphics mode was Framebuffer. My impression is from the Ubuntu developers that they like some kids who are just pushing things like they want without any though about anything. Just to be different or "cool".
First the totally unnecessary changes in Gnome with the Close/Minimize buttons; then the not usable Unity; and now the Wayland, which will be usable only after more 5 years in development and testing.
http://www.mueller-public.de - My site http://www.anr-institute.com/ - Advanced Natural Research Institute
X11 is more like a roof that was originally a 40-foot timber yacht. You've turned it upside down and fixed it to the tops of the walls. Gradually, over the years, you've patched in the holes until it only leaks when the wind is in the West. You've figured out how to get a flue up through the thing. You've nailed a TV antenna to it, and sealed around the cable with silicone. When you put it there you never bothered to take the decks out, so it's almost impossible to get into and work on and the structural elements, optimised rather for the sea than for housing, make it not very useful for storage. You still have to repaint it with pretty expensive paint every five years or so, else it starts to rot, and for some reason it attracts lots of confused-looking seagulls.
Anyway, look at all the features! It's got a winged keel, a 200hp diesel engine, and a gorgeous timber and brass wheel. All the fittings are marine-grade stainless, the rigging was all almost brand-new when you installed the thing and in her day she'd do 27kt reaching across a good wind. Don't actually use much of that any more, of course, but still...
Technically the keel still violates local planning ordinance, and technically it still smells quite a bit of fish. But it's been there for 15 years and it works. There's no need to replace it.
What's that love? You want to build an extension? Ah.
Slashdot - News for Nerds, Stuff that Matters, in ISO-8859-1 Has just realised that beta makes this signature redundant
And, in case you missed some of the metaphor there, the basic problem with X11 as a display system for desktop systems today is that none of the apps written for it actually use X11. Your multiple monitors work because of Xrandr - which isn't part of X11. Your 3D effects work by bypassing the X server and making OpenGL calls - not part of X11. It's supported by all the hardware vendors - and most of that support is in kernel drivers, not in the X server itself. VT switching works... more... or... less... until you'd actually like to switch VT before you get back from your coffee break, or your X server doesn't recover correctly. And, in case you missed it, they're not provided by X11. And I'm not sure what boot times have to do with it?
Your apps draw using Cairo or similar, not X11. They draw onto a buffer provided by the compositing extension, not X11. The buffer gets put into video memory by the compositor, not X11.
So why exactly are we keeping X11 hanging around? Why not get rid of it and halve the size of the display server code base, making it much easier to program against in the process? Why are we carting around a heuge amount of code that is of no modern relevance except to be able to claim that it is an X server? The maintenance burden of the current X server is too much and any thought of adding new capabilities horrific.
Otherwise, your arguments amount to, "No-one supports it yet so it's a waste of time." Good on Canonical for pushing it - if anyone has a vague chance of getting vendors to support it, it seems they might.
Slashdot - News for Nerds, Stuff that Matters, in ISO-8859-1 Has just realised that beta makes this signature redundant
Wayland apps are like Mac apps. They have no native remote desktop capability. You have to use a 3rd party hack like VNC to get such a feature.
I've seen how VNC runs on a Mac. Not impressed.
Running X on a Mac won't let me run iTunes across the network. The same is true of Windows.
A Pirate and a Puritan look the same on a balance sheet.
That's an awful lot of hyperbole in one post with no actual information.
The thing that would be best would be X12, not scrapping X11 completely. There's lots of good and useful things in X, which Wayland doesn't have.
Oh, and by the way, please, if you want OSX or Android, you know where to buy them so go and do so. Stop trying to turn my very nice desktop system into an inferior commercial alternative.
SJW n. One who posts facts.
From the FAQ: "The Wayland architecture integrates the display server, window manager and compositor into one process."
The Wayland window manager will be built in. If you want to change your window manager, you have to change the display system itself.
Particularly egregious is the choice to have Wayland clients provide window decorations. That means that no matter what window manager you have compiled into Wayland, you can't centrally control the appearance of your desktop.
What happens if I want to use a Wayland distribution that provides a tiling window manager, but my apps are coded with window decorations? What happens if I want to use Fluxbox's built in window tabbing, but none of the clients support that?
Give me Classic Slashdot or give me death!
Welcome to *nix. You must be new here.
I rather like that the window manager, greeter, and locker are separate processes. I can replace them. I can mix and match according to my taste. I could even write my own if I wanted. This is the Unix way.
Example: you probably use some ubuntu-y, gnomey, kde kind of bullshit. That's fine for you but I have different tastes. So I use xdm to log in, wmaker to manage windows, and xscreensaver to lock my screen. Result: my X11 install is several hundred megabytes smaller than yours. But if I want your setup it's just a small apt-get away. This diversity, choice, and flexibility are why I like the *nix desktop in the first place.
But if I read you correctly, you want it to be all 1 big monolithic process? What is it, you're concerned about context switches between these processes on today's hardware? What the hell planet are you from? The X design has worked well for decades on much less powerful hardware than I have in my pocket today. X works fine on all of my devices, just like it did in the 90s, just like it does on my Nokia N900 running at 600mhz. I will be sticking with X for the time being.
But Ubuntu is for home users, how many home users are running remote desktops?
How many home users are using the command line? Why don't we rip that out too?
You've got all this stuff that frankly isn't need for the ones the OS is being pushed for, and which makes them system more brittle
What evidence is there that network transparency causes X to be "brittle"?
see the classic rant from thom at OSNews about how X crashed when doing simple tasks for a loooong list of people having trouble with it in the home user space
Sounds like a driver problem, not an X problem. He has a point in that X should be able to handle driver problems more gracefully, but I don't see what this has to do with network transparency. Can't we have both network transparency, server side window decorations, and robustness against driver faults?
why not simply let X be for servers and Wayland be for home users? Choice is good, right?
Fragmentation is bad for choice. If I want to use a certain app, I have to use the display platform that app is written for. And there is no bright line between server software and home software, so anyone who isn't an average user(and on average, almost everyone is not average) is going to be locked out of using some software.
Give me Classic Slashdot or give me death!