Slashdot Mirror
GTK+ 3.8 Released With Support For Wayland
kthreadd writes "Version 3.8 of the GTK+ GUI framework has been released. A new feature in GTK+ 3.8 is support for Wayland 1.0, the display server that will replace X on free desktops. Among the other new features are improved support for theming, fixes to geometry management and improved accessibility. There is also better support for touch, as part of an ongoing effort in making GTK+ touch-aware."
"A new feature in GTK+ 3.8 is support for Wayland 1.0, the display server that will replace X on free desktops." Who said this is going to replace X on 'free desktops'? As far as I have been hearing, this is just another in a long line and because it hasn't done it yet, it is not justifiable to say it will.
when you pry it from my cold, dead, fingers!
the display server that will replace X on free desktops!
yea I know it takes some time to get stuff right, but call me when this thing gets out of duke nukem forever mode k
thanks
Has there been a Gtk release where theme compatibility was the only thing that broke? That sounds amazing!
(Why yes, I do write Gtk apps for a living, how did you know?)
There's no -1 for "I don't get it."
Yet more "waaah, they're taking my X remoting!" crap. Wayland will enable an X server to run on top of it just like Windows does, just like OS X does - whilst enabling a far more efficient and modern rendering pipeline.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
I haven't used gtk+ much (for my own programming, that is) but I do use glib. A lot. God's gift to C programmers: glib.
If you're a zombie and you know it, bite your friend!
If not for the announcement of Mir, this would have taken at least 5 years
VNC? What if more than one user wants to use something on the remote machine? It's the age of the "cloud" where you can get a 64 core machine for around $9k, hook it up to fast storage and let a lot of people use it for very complex tasks. What is someone want to run things on multiple machines and doesn't want to juggle half a dozen full "desktops"? Sod this MSDOS single user non-networked approach - it's 2013 FFS!
Also that block diagram implies speed hits from the complexity and ignores that the wayland server+compositor is going to be doing a similar number of things internally as both the X server and compositor, so it doesn't prove your point and I doubt the person that drew it intended it to be used to try to prove that point.
It's been a long time and a lot of claims - why no benchmarks for identical task yet instead of handwaving and "X sux!!11!"
VNC? What if more than one user wants to use something on the remote machine?
They they start another Xvnc process.
What is someone want to run things on multiple machines and doesn't want to juggle half a dozen full "desktops"?
Yes, that is a real problem.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Doesn't matter. GTK dev's (and GNOME dev's) can't be trusted. I'll use QT from here on.
Didn't QT announce support for Wayland, too?
Most of the objections raised about network transparency seem pretty silly IMO. Network transparent apps are a niche feature even in Linux and there are alternatives even in the short term, not to mention longer term possibilities.
What is someone want to run things on multiple machines and doesn't want to juggle half a dozen full "desktops"?
I wonder for how many people this scenario would even apply though, or why it should mean the experience in a Linux desktop should be hampered by X11 just to facilitate it.
I wonder for how many people this scenario would even apply though, or why it should mean the experience in a Linux desktop should be hampered by X11 just to facilitate it.
That begs the question, is the experience in a Linux desktop hampered by X11? I've so far seen no evidence that it is.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
So why stick all the old cruft of X right back on?
So you can fire up the X cruft when the user starts an application that uses the X cruft, and you can shut down the X cruft when the user has closed the last application that uses the X cruft. As more GUI toolkits are ported to Wayland, fewer will require X to be running.
I don't know what kind of graphics cards you put in your servers (or why)
I know why: abusing pixel shaders.
Bandwidth is a far easier problem to solve, particularly for longer connections.
Are you talking wired or wireless? And how long of a connection? In a way, the sustained bandwidth of a typical connection is limited to 6 to 15 kbps. That's the 2 to 5 GB per month cap on a mobile data connection, times 8000000 kilobits per GB, divided by 2629746 seconds in an average month.
How many more years to we have to wait for a Win32 port of GTK+3? There are several projects which only have old versions ported to Windows because their newer builds target GTK+3 and that's not available yet.
Hail Eris, full of mischief...
E pluribus sanguinem
Wayland does not guarantee a network transparent transport.
However, users' demands will guarantee that distributions make such a transport available through a command analogous to sudo apt-get install wayland-network-transparent-transport.
Would your Kindle allow you to run remote X sessions?
That depends on whether or not Amazon left a feature essential to the X server for Android out of the Kindle Fire. Did it?
About 50 people where I work for a start. Different servers do different tasks so application windows are vastly better than a stack of slow VNC desktops and even large images are best dealt with on a node with 32GB instead of a desktop with 4GB.
In what way exactly? Wayland hasn't progressed to a working demo state yet so how is X hampering people more than what is available?
It appears we've hit the problem of the person who is dismissing X out of hand does not actually understand why people use it, which I suppose it why your suggestions have failed to address the questions.
Until Wayland can match the features and performance of X on a Pentium100 in 1999 or X on a Kindle today I'm going to call out all these "X sux" idiots that go on about performance problems. Benchmarks say a lot more than "just do something incredibly inconvenient and Wayland does the job", so how about some benchmarks from tests instead?
That modular behaviour that you are pretending is a flaw is probably why we are still using X after all these years - like grandpa's axe we've put a better handle on and changed to a better head three times but we still call it grandpa's axe. At least one positive thing along those line with wayland is it can use more than just it's own toolkit even if the window manager etc is all built in.
Well if the inclusion of JFS meant that the experience for the majority using EXT3/4 or BTRS was degraded then yes. But it isn't, so it doesn't matter. But for the majority of people using a local desktop, they are suffering a degraded experience for the sake of a fairly esoteric workflow (one which can be accomplished anyway even if the desktop did move to Wayland).
X11 involves a large number of context switches thanks to all the processes working to update the display. The intent of Wayland is to reduce that complexity and context switches which means the desktop will be more responsive and lightweight. I'm quite willing to accept that it will take some time for Wayland to mature sufficiently to reach optimal performance and stability, but that aside, it's obviously a smart idea to pursue. Many prominent X devs think so do, such as Keith Packard.
Criminal... got a feeling that they will search for keywords and pass it along various sub agencies or group, each responsible to figure out if it's a lead within their roles.
You have a bandwidth limited connection? That's nice. Pay some more money, and get more bandwidth.
But there is a limit to how much more money for bandwidth the user of an application will be willing to tolerate. Good luck affording enough bandwidth to run something like OnLive for 8 hours a day, 22 days a month.
Just that the fewer the round-trips, the less latency dependent it becomes.
And if the application is running locally in a sandbox, communicating with the other machine only to synchronize data, there are even fewer round-trips. One example is any web application written in JavaScript.
It appears we've hit the problem of the person who is dismissing X out of hand does not actually understand why people use it, which I suppose it why your suggestions have failed to address the questions.
I'm not dismissing X out of hand. That would be a straw man. What I am saying is that it clearly impacts on the local desktop performance (and it's not hard to find comments by leading X devs who state this for a fact). And most of the objections raised for switching to something more efficient concern a feature that not many people use, and even if Wayland were to become the default experience, could be achieved anyway.
Anyone who absolutely cannot abide the change can just use an X11 fork, or vnc / MX / X over Wayland or scratch the itch and implement in Wayland what they perceive to be missing.
Do you think that Wayland or Mir could add speed and smoothness to the Linux 3D-accelerated desktop? Because that's the feature I'm anticipating the most. When you turn on a little bit of effects there's always that slight jerky feeling you start getting. And it eats the performance too much. There's little delays in things happening, or some animation might pause for a little while. The icing of the cake, the fine responsivity of Win/Mac is currently not there with Linux.
Looks like the FreeBSD project is now working on adding Wayland support. I wouldn't be surprised if NetBSD too gets it. As for Solaris, has Oracle really been adding much to it of late?
I see points two and three in that link as listing shortcomings in current Wayland design - then look at the questions raised in the comments "I am dreading the applications taking control of their window decorations" for one - the sort of thing that has made the MS Windows desktop very confusing at times with apps that have their own subwindows.
The ironic thing is all the hate poured on X even though that's where the Wayland direct rendering was taken from.
So sorry to be someone in a "pretty silly" niche that Wayland developers disparage as irrelevant - but the only reason I have a job working with linux, solaris and freebsd is because X allows users to do the many to one situation of having many applications from many nodes open on their desktops at once without a painfully time wasting VNC experience. Once apps come out as Wayland only it's a trip back to the MSDOS mentality - fine for gaming consoles or phones but not much good in other situations.
So instead of giving an answer you'll namedrop someone that gave the idea and not the implementation faint praise?
But addressing window decorations, that confused me too for a while. As far as I can tell, Wayland does not see it as its job to adorn a surface with decorations. So the simplest solution is the client does it and I expect a library or theme engine of some sort will appear which will take care of it and ensure a consistent appearance over the desktop. But client side decoration isn't the only option - KDE is talking of server side decoration implemented inside the window manager that underpins their desktop.
It's not like you can't override the frames in X even if you have to do tricks - look at Google Chrome for example. Or Steam. Or Wine apps running on X. I assume they all disable server frame and paint their pseudo decorations and frame in its client area. While this looks a bit odd, it also brings advantages too. For example Chrome wastes far less space in GNOME 3 than Firefox is because it condenses 3 horizontal strips of buttons, menus and title bar into a single strip. Sometimes an app does need to be in control of it's appearance.
And to reiterate my point, remote desktops and apps don't go away even on Wayland. There is even an XWayland which I expect that dists will ship and run with for a good while since some apps may be more gnarly to port than others (e.g. what does Firefox do about plugins). Just do what you did before or avail of the other options, or scratch the itch, or fork. I don't see the issue at all. It doesn't mean the entire desktop needs to be hampered by a 30 year old architecture with numerous recognised bottlenecks.
I guess it will take up kilobytes of space on disk while the unused code sits paged out.
Is all the unused code guaranteed to be in the same page as other unused code, rather than being in the same page as used code? Because if one piece of code in a given page is used, the whole page has to be loaded into RAM.
You somehow have managed to miss my entire point of wayland only apps being stuck on scraped screens at best, and have also not managed to show that a monolithic approach with internal communication is necessarily going to outperform a modular approach - you have instead been stating that it is faster with no benchmarks or other evidence to back it up.
Second, a "monolithic" approach has obvious advantages since there aren't 3 separate processes sending messages between themselves, maintaining duplicate state and incuring context switches. Does that mean it translates into an efficient implementation? Time will tell. As for benchmarks, are you even serious? Wayland is still in development. It won't be possible to to compare like for like for some time. It does not mean that makes a valid excuse for not moving. Even seasoned X developers want the move and recognise the need for it.
It's not like X11 is going away any time soon. It will sit on top of Wayland at least until the apps which constitute the desktop are ported.
Since you and others are talking about how wonderful wayland is in the present tense instead of the future sense of course I'm serious - it's a polite way of saying either put up or stop being dishonest.
No, dishonest is where someone pretends that someone else talked about wayland in the present tense in order to justify some ludicrous demand for benchmarks. Stop being ridiculous.