I love Linux and vastly prefer it to Windows, but whenever people tell me Linux is vastly more stable than Windows, I think of the whol X11/Wayland saga.
I still scarcely know what these are. In fact I actively don't want to know about compositors and whatnot. When I want GUIs, I just want to see them.
I can list many crappinesses of Windows, but stuff like this kinda just works.
I've been using Gnome for years, but, honestly, it just isn't good: seems like it's optimized for very basic use. Something as simple as adding launcher to a panel now requires an extension.
Also Wayland has some problem on my system (Thinkpad / Intel Xe) where it randomly just goes slow, this makes it an easy choice to try things other than Gnome.
I haven’t booted into an X11 environment in maybe 4 years. Wayland has been fine (Fedora + Gnome, Fedora / Arch + Niri). I think this is one of those issues where hardcore users overestimate how much anyone else cares or will notice.
I used X when I started with Linux in my high school. When I finally stopped tolerating macOS almost a decade ago, I started with Wayland only, and never looked back. I even ditched Krita in favour of Gimp, solely because Gimp supports Wayland (from version 3, and on 2.99 before that). To this day, I don’t understand X, like at all. It’s all some super complicated pile to me. While I don’t understand Wayland either, it’s super simple to interact with as a user. That’s all I care about, especially given that Wayland is better with security, multiple displays, has a smaller code base, and whatnot. Each time I see someone says Wayland is unusable to them, I wonder whether that’s some edge case they have, or is it just some habit. I’m happy I don’t need X for anything, so I just wait till most apps would support Wayland.
Remember x11 was initially a client server model, where multiple clients could access a single x11 server. This brought a lot of complexity we don’t use these days
> Remember x11 was initially a client server model, where multiple clients could access a single x11 server. This brought a lot of complexity we don’t use these days
It's worth noting that the "client" and "server" are flipped from what was typical: your screen on your desk is the server and the client is a program running on some expensive machine on a rack somewhere.
It's still really cool to be able to spin up a GUI app on a remote machine, and use it like it's running locally.
For certain jobs, I've done development for Linux while also having a Windows box for other things. Opening Linux GUI apps remotely on my Windows desktop is nice and allows me to consolidate my displays. This is an edge case, for sure. How well does Wayland support this?
If you want that feature, then the display server doesn't need to be the one to support it when the display server lets applications obtain and control window positions.
Wayland doesn't let you do that, and it's a deliberate choice.
Transition to Wayland opened so many user experience regressions. Many are solved today, or at least partially solved but...
There is still no possibility to have proper remote sessions when using Wayland. On any Window Manager and any distro. It's such a shitshow when you go into details. Nothing works, including third party tools (like NoMachine) and I could find no real hope for actual solutions being designed.
The best you can go with "remote session" on Wayland is viewing a desktop session that was already opened by someone directly on the computer. You can partially work around this by... setting your account to be automatically logged in with no password :D And even then it's a crippled experience.
A basic feature I used for the past 25 years and helped me to learn linux and offer safe space for others to learn it as well. To work around work computer limitations. To use your best hardware wherever and whenever you want.
I currently had to ditch both my favorite distro and WM because of that. But at least we can make screenshots nowadays, so I guess it could be worse.
Kde's new Plasma Login Manager / Plasma Login (backend/frontend) have been coming along nicely (replacing sddm), and include remote login support. Just announced this past spring but very active. https://blog.davidedmundson.co.uk/blog/a-roadmap-for-a-moder...
The "Transition to Wayland" from a user experience pov is the slowest car crash of all time. We are like 1.5 DECADES in at this point.
I have a simple application written in QT6. It works on Windows, macOS, and X11/Linux. On Wayland/Linux, applications cannot move their own windows anymore, because "security". Good luck finding this in the QT documentation, it is there, but only at 3/dozens of places were it would be necessary, and 2/3 of those dont mention the word "Wayland". Great fun.
Remote assistance, NOT remote logins. It can be used as support when someone is already went to that computer, authenticated and has a full gnome session opened.
So you literally CANNOT log in remotely :) If you are lucky, you can assist remotely to a session someone opened locally on that machine.
And it's like that on any other WM. KDE also has a deceiving option in settings that suggests full remote desktop, while it doesn't allow that.
I don't want to argue on semantics. Currently you can't start a graphical session completely remotely using any protocol (RDP, VNC, no machine, whatever).
There are no headless sessions on Wayland. At all.
You want proper headless session, set up X11 distro and use xrdp - it's really easy. But on wayland "remote support" to something that is already displayed on screen is all you can get now.
What I want is to be able to start a session remotely after a reboot, and continue that same session when I get back home. And conversely start a session while at my desk at home and resume that same session remotely. Without any weird limitations.
In other words, how RDP works on Windows.
So you're saying that is still not possible I take it.
I was working on a carousel library a few months ago. I had made a few stress-test demos so that I could catch obvious issues while I was adding things and tweaking things.
One carousel there had 16K slides.
On Windows both Chrome and Firefox managed that fine. They scrolled from start to end and back without issue and you could see, I think, all the frames in my 60Hz screen.
On GNOME and X11 (dual boot, so same hardware) Chrome was fine but there were issues with Firefox. I was curious so I logged out and logged in with Wayland. On Wayland Firefox was fine too, indistinguishable from Chrome.
I don’t understand hardware, compositors, etc., so I have no idea why that was, but it was interesting to see.
Anecdotally, I strongly doubt this is true, although my environment is probably quite biased. I know a ton of people who use Gnome, some who use KDE, and I think roughly all of these people use them with Wayland. The standalone-WM users I know are also mostly on Sway or other Wayland ones. The only real X11 holdouts seem to be people using X11-only DE's, such as Xfce or Cinnamon.
Xfce is working on Wayland session support. It is working now with some limitations (limitations on what you can embed in the panel are all that's left, I think).
Makes sense to move to something actively maintained. I don’t use GNOME, but I have been on Wayland for years in both KDE and tiling WMs. It works great while on X11 I would often get weird flickering and stuttering issues (some NVIDIA bug I could never track down). Anyway, if the X11 die-hards want it to survive, they need to organize an effort to maintain it, not yell at everyone who wants to build something better.
I am sympathetic to people who have a working setup and just don’t want to mess with their configuration anymore. Unless you’re on OpenBSD, though, that ship has long sailed in most *nix distros (even “stable” Debian). Long-term stability is underrated but hard to achieve.
Sadly, I recently had to switch back from Wayland to xorg because clients are getting so memory hungry. My eight year old gpu only has 2gb of vram, which I constantly run out of. Some part of the gfx stack should handle swapping out vram to main ram but it apparently isn't.
Another nail in the coffin. Bye, Red Hat. Bye, NGOME. Non GNU OldIBM Mediocre Environment.
I wish GTK4 dies in IBM hands too, for the good. XFCE can go back to a community supported GTK3 anytime.
x11 is in maintenance mode at this point and Gnome is not going anywhere. Gnome is used (and financed) by major distributions.
Nothing new is being created with x11 and the people from freedesktop don't seen to be thrilled to maintain it. I don't think should change just for the sake of changing, but I'd start looking to migrate whatever you use that depends on x11.
I had assumed that XWayland is a drop-in replacement fo X11, and will be available indefinitely.
I regularly write code which relies on a working X11. I have written a virtual machine which makes X11 calls to do 2D graphics and event handling, as well as applications which compile to the virtual machine code. If X11 and now XWayland cease to be available, not only would I have to rewrite large parts of my virtual machine, but also rewrite all the 2D graphics code in applications. All so that I can stand still when the rug is being pulled from under my feet. I'm sure there are others in a similar predicament.
I may be naive about this, but as X11 just works, and has done for decades, it should require little to no maintenance, so why the need to withdraw it? I don't expect, or require, any additional functionality.
For my part, I have no intention of moving off X11 for the next decade at least. The only app I use that I don't fully control is a browser, and the worst case fallback is to run the browser in a Wayland compositor that runs on X.
What a weird question. I have no use compelling use-case for python2. I have plenty of use-cases for X, such as the fact that none of my software other than my browser has Wayland support, including my window manager.
Nothing weird about it, it makes perfect sense with what you are posting. I find it weird that you react like that to a simple question. Just probing what kind of old software you use, cause it tells a lot depending.
It was weird because you implied (and still are) that it'd mean I was hanging on to old, superceded software that has adequate replacements. I'm not. So this is telling us more about your assumptions.
I'm using my own terminal, wm, and file manager. They use X11, and I have no interest in changing that, because I have no need to as long as X11 works on my hardware and that won't change anytime soon. Everything I don't do in a terminal, I do in a browser.
EDIT: To add some more context for why I have no interest in changing that:
1) my wm is 1568 lines of code at the moment. If anything, that is more than I'm happy with. With Wayland I'd need to write my own compositor. Way too much work even with reusing e.g. wlroots.
2) My file manager is more of a basic desktop launcher. That is fine, and intentional. I may add some features to it. But the reason I'm using that rather than any of the over a dozen options I've tested is that most of them either never had or have ripped out spatial features, and the ones that had some spatial features didn't act the way I wanted them to. I want Amiga-like semi-spatial features of being able to selectively snapshot icon and window placement ("semi-"spatial because traditional spatial would imply a single instance of a window for a given path; I just want default placement to be the same as last time I snapshotted it). Wayland on purpose refuses to allow that, and so I'd need to hack on a compositor or write my own to be able to support the most important feature to me in the file manager.
I'm not going to tolerate my usability being reduced just to switch away from software that does what I want it to, to software that offers me nothing new that I want and takes away features I do want.
I'm curious, what were you hoping to learn about their use of python 2? If you had specific questions, it would be helpful if you ask those, instead of trying to ask through euphemisms.
If it was maintained (security fixes and platform support only), but with no other changes, it would be a very tempting alternative to python3 for the many times API stability is valuable.
And wayland is in broken mode. KDE keep changing the default back to wayland after each update, and every time my linux systems are broken until I switch back to x11.
What is broken for you? At this point, starting from roughly KDE 6, Wayland has been pretty much flawless for me. KDE 5.27 was pretty much fine already as well.
I've never seen similar issues using a variety of terminals on Gnome, Sway, or Niri. Haven't used Konsole or Plasma, but I wonder if it's maybe a driver issue?
Do you have more specifics? I just tried it on my machine (Fedora 42, Plasma 6.5.1 Wayland, Konsole 25.08.2, Radeon 780M) and it seems fine for me. Does it only occur occasionally/under specific circumstances for example?
It would be sad if, after all those years, it was still missing anything significant. Maintenance mode sounds like a good thing, not something that makes me tempted to switch to some less stable alternative.
I don't know the implementation details but I can't really complain about the state of wayland today. It used to be annoying to get working many years ago (worse because I had a nvidia gpu). But today I drive a nigthly build of niri, run it by just spawning an dbuss session and everything works. Bluetooth audio, screen sharing, fractional scaling, no tearing, no font blurring. Every utility I needed has been created and works quite nicely (e.g.: wdisplay). I can even play video games with HDR support.
I have a more stable experience with wayland today than I had with x11. Which to be fair was not only because of wayland but because desktop linux as a whole has made a lot of progress in the last years
I don't think it's true that anything is architecturally or fundamentally broken in Wayland (though if you disagree, I'm very curious what you think is so deeply broken).
Most of the issues and slow adoption were because the core protocol was deliberately kept extremely minimal, and agreeing on all the needed extensions took a long time. Don't take it from me, but rather from KDE developer Nate Graham: https://pointieststick.com/2023/09/17/so-lets-talk-about-thi...
As such, anyone who tried it early probably had to deal with a pretty large amount of non-working stuff, but by now the platform is capable of most features people require and the biggest remaining bottleneck is that software needs to use these new APIs.
Window positioning is one that on its own is sufficient to make me ignore Wayland, as it means that without my own compositor with my own extension, I can't get a file manager that will behave how I want it.
Most people won't care, but for a number of us Wayland is stubbornly refusing to support functionality we see as dealbreakers.
That's fair! I believe that window positioning also works on XWayland, though, so running your file manager that way should still work with the rest of the system being Wayland (and Gnome has no plans to drop XWayland afaik).
I believe the main holdup is a desire for Wayland to be usable with e.g. VR interfaces where there is no simple 2d grid.
Out of curiosity, how do you want the file manager to behave? And did you write your own or are you using an existing one that works that way?
It's managing the desktop too, so I'm not sure that'd work unless running Xwayland in "rootful" mode, in which case I might just as well run X.
The VR stuff is a poor excuse - just fail on that scenario. Nobody that cares about window positioning will have an issue with that.
My file manager defaults to re-opening a window for any directory to a previously snapshotted location, like the Amiga Workbench did. And, yes, I wrote my own. It's a few hundred lines of of a quick and dirty Ruby hack talking directly to a pure Ruby X11 binding, which is anothe reason I stick with X - I can throw things together quickly for X. The amount of ceremony, or big additional dependencies, needed for Wayland is ridiculous.
What's the substance behind this claim? It keeps on being repeated but I don't get what it's actually about. Is there anything proprietary about Wayland that I'm not aware of? What's the difference between proprietary drivers using X11 and Wayland?
I don't think I understand what you mean. Do you mean wayland is not usable with nvidia proprietary driver? I remember that being annoying but possible many year ago (with sway --my-next-gpu-wont-be-nvidia thingy).
But if you use really old nvidia gpu you can have a mixed experience with wayland. Which is a fair problem to complain, but you can't blame that on wayland and call that lack of freedom. That problem was caused by the lack of freedom coming from nvidia gpus and how locked down they are and how nvidia for many year has been hostile towards linux desktop.
The “what’s the harm” here is the systemd conversation all over again basically. If you pipe everything through a single point of failure black box users have already lost, when you combine it binary blob drivers that shouldn’t exist it’s worse. Linux is doomed in achieving its most important goals which are user freedom, not someone’s idea of pretty UI imposed at the expense of that. If that’s what users want they should buy a Mac. If you want to get locked out of your OS for eye candy we have that.
Where is this Wayland black box then? If anything, Wayland made this situation significantly better: the X11 server was exactly this 'single point of failure black box' you are describing. Wayland replaces this with a much simpler protocol with multiple independent implementations (notably Mutter/gnome-shell, KWin, wlroots-based ones such as sway, and Smithay-based ones such as niri).
It’s achieved developer and very tech savvy IT pro freedom. If you can deal with command lines and debugging systems you are not a user. You’re a computer professional.
If OSS wanted to bring freedom to users its primary focus would be radical simplification and UI/UX.
What nonsense, especially if framed in a such an absolutist way leaving no room for nuance. Millions of people are storing their data on self hosted or 3rd party run managed OSS services on multiple platforms. Sure, it is not all perfect, UX might be an issue here and there but compared to having to relearn using some proprietary app redesign every other year because some Product Manager needed a promotion, a lot of OSS stuff is perfectly usable, stable and secure.
The X Window System (X11) is a protocol with multiple implementations. Sure, the X.Org Server (Xorg) was the most popular by a huge margin, but there were quite a few others (e.g. XFree86, Xming, XWayland), though over time most were discontinued for one reason or another.
X11 and Wayland do differ in an important way: in X11 window managers (GNOME, KDE, i3, whatever) all sat atop the Xorg server; whereas in Wayland there’s only the compositor, so GNOME, KDE, Sway, whatever, all essentially include their own equivalent of Xorg (which could be fully integrated, or factored into a library, such as Mutter, KWin, wlroots).
X11 desktop environments are dying, but it doesn't mean that X11 is dead. XWayland is still a thing so you can still run your X11 apps on Gnome.
The big reason why I want to keep X11 besides backwards compatibility is the ability to run GUI apps remotely, even from a server that has zero graphical capabilities. But these do not really apply to desktop environments. If you want to remote a full desktop rather than individual applications, there are better options (VNC, RDP, ...).
I think waypipe should generally do what you want, though TBH so far when I need to run a remote GUI app, I just have them display to XWayland out of habit.
Rumors of Gnome's demise seem greatly exaggerated to me. It's still the default DE in nearly all major distributions, and it doesn't seem to have incurred major mindshare or marketshare hits recently. I feel like most of the 'complainers' already abandoned the Gnome ship with the release of GNOME 3.
Really the only high-profile 'switch' in recent times I can think of is that Fedora promoted KDE to be first-class ('edition') alongside Gnome, instead of delegated to a more second-class spin. And while KDE is a bit more conservative in this regard, I believe that in the long term KDE also wants to go Wayland-only at some point.
Personally I did switch from Gnome to KDE some time after Gnome 40, since I quite liked 3.x but the UI overhaul 40 did wasn't really my thing. It also helps that KDE got a lot better in recent years.
We'd have had a lot fewer problems if Wayland had been X12, in the sense of taking an approach of gradual iteration, even if they went at it fairly aggressively.
E.g. instead of the Xwayland approach, you could've already ditched "half" of Xorg if you stripped most of the server-side drawing primitives and server-side font support and moved them to Xlib, handling it client side, and then made it clear someone else would need to take over maintenance of Xlib, and "started over" with a stripped down Xcb.
You could've validated further restrictions by letting clients opt in to them with extensions before "flipping the switch" and restricting them by default when the damage was acceptable.
Even if we then eventually reached a point where there'd be a schism, odds are it'd be far smaller. And certainly far less time would've been wasted.
X11 isn't going anywhere because distros will ship XWayland for a long time to ensure compatibility with existing X11-only applicatin.
xorg-server is gone from the linux desktop. Gnome and KDE use wayland shells by default, and that's what users get when they download a Debian/Ubuntu/Fedora/whatever ISO.
I love Linux and vastly prefer it to Windows, but whenever people tell me Linux is vastly more stable than Windows, I think of the whol X11/Wayland saga.
I still scarcely know what these are. In fact I actively don't want to know about compositors and whatnot. When I want GUIs, I just want to see them.
I can list many crappinesses of Windows, but stuff like this kinda just works.
Does this mean we finally get proper drag-n-drop support back? [1]
[1] https://gitlab.gnome.org/GNOME/file-roller/-/issues/4
I was mostly surprised by the "Gnome 50" part. Last I remembered Gnome was still version 3. Turns out they jumped from 3.38 to version 40
They are not planning to go Gnome 4, hence Gnome 3.40 became 40. Just like Emacs went from 1.12 to 13.
I've been using Gnome for years, but, honestly, it just isn't good: seems like it's optimized for very basic use. Something as simple as adding launcher to a panel now requires an extension.
Also Wayland has some problem on my system (Thinkpad / Intel Xe) where it randomly just goes slow, this makes it an easy choice to try things other than Gnome.
[flagged]
Comments like these are less than worthless. If you're going to contribute, say something meaningful.
[flagged]
Gonna tell him what?
I haven’t booted into an X11 environment in maybe 4 years. Wayland has been fine (Fedora + Gnome, Fedora / Arch + Niri). I think this is one of those issues where hardcore users overestimate how much anyone else cares or will notice.
I used X when I started with Linux in my high school. When I finally stopped tolerating macOS almost a decade ago, I started with Wayland only, and never looked back. I even ditched Krita in favour of Gimp, solely because Gimp supports Wayland (from version 3, and on 2.99 before that). To this day, I don’t understand X, like at all. It’s all some super complicated pile to me. While I don’t understand Wayland either, it’s super simple to interact with as a user. That’s all I care about, especially given that Wayland is better with security, multiple displays, has a smaller code base, and whatnot. Each time I see someone says Wayland is unusable to them, I wonder whether that’s some edge case they have, or is it just some habit. I’m happy I don’t need X for anything, so I just wait till most apps would support Wayland.
Remember x11 was initially a client server model, where multiple clients could access a single x11 server. This brought a lot of complexity we don’t use these days
> Remember x11 was initially a client server model, where multiple clients could access a single x11 server. This brought a lot of complexity we don’t use these days
It's worth noting that the "client" and "server" are flipped from what was typical: your screen on your desk is the server and the client is a program running on some expensive machine on a rack somewhere.
It's still really cool to be able to spin up a GUI app on a remote machine, and use it like it's running locally.
For certain jobs, I've done development for Linux while also having a Windows box for other things. Opening Linux GUI apps remotely on my Windows desktop is nice and allows me to consolidate my displays. This is an edge case, for sure. How well does Wayland support this?
I don't understand how Wayland is becoming the norm when it can't even restore window positions yet.
Why does the display server have to restore window positions?
If you want that feature, then the display server doesn't need to be the one to support it when the display server lets applications obtain and control window positions.
Wayland doesn't let you do that, and it's a deliberate choice.
See e.g.:
https://wayland-book.com/xdg-shell-in-depth/interactive.html
"However, a deliberate design trait of Wayland makes application windows ignorant of their exact placement on screen or relative to other windows."
And: https://hackaday.com/2025/11/11/waylands-never-ending-opposi...
Because the display server refuses to let anyone else set or restore window positions.
Tiling window managers are the future.
Blocked in my location. Did a quick search only to find that it's blocked in many other locations - https://news.ycombinator.com/item?id=35314374
https://web.archive.org/web/20251114122702if_/https://linuxi...
Transition to Wayland opened so many user experience regressions. Many are solved today, or at least partially solved but...
There is still no possibility to have proper remote sessions when using Wayland. On any Window Manager and any distro. It's such a shitshow when you go into details. Nothing works, including third party tools (like NoMachine) and I could find no real hope for actual solutions being designed.
The best you can go with "remote session" on Wayland is viewing a desktop session that was already opened by someone directly on the computer. You can partially work around this by... setting your account to be automatically logged in with no password :D And even then it's a crippled experience.
A basic feature I used for the past 25 years and helped me to learn linux and offer safe space for others to learn it as well. To work around work computer limitations. To use your best hardware wherever and whenever you want.
I currently had to ditch both my favorite distro and WM because of that. But at least we can make screenshots nowadays, so I guess it could be worse.
Kde's new Plasma Login Manager / Plasma Login (backend/frontend) have been coming along nicely (replacing sddm), and include remote login support. Just announced this past spring but very active. https://blog.davidedmundson.co.uk/blog/a-roadmap-for-a-moder...
The "Transition to Wayland" from a user experience pov is the slowest car crash of all time. We are like 1.5 DECADES in at this point.
I have a simple application written in QT6. It works on Windows, macOS, and X11/Linux. On Wayland/Linux, applications cannot move their own windows anymore, because "security". Good luck finding this in the QT documentation, it is there, but only at 3/dozens of places were it would be necessary, and 2/3 of those dont mention the word "Wayland". Great fun.
"gnome-remote-desktop" does exactly that - providing (amongst other capabilities) a way to handle remote logins: https://gitlab.gnome.org/GNOME/gnome-remote-desktop
Remote assistance, NOT remote logins. It can be used as support when someone is already went to that computer, authenticated and has a full gnome session opened.
So you literally CANNOT log in remotely :) If you are lucky, you can assist remotely to a session someone opened locally on that machine.
And it's like that on any other WM. KDE also has a deceiving option in settings that suggests full remote desktop, while it doesn't allow that.
Please just click in the link, read the README! It offers: - remote assistance - headless multi user remote login - headless (single user)
I don't want to argue on semantics. Currently you can't start a graphical session completely remotely using any protocol (RDP, VNC, no machine, whatever).
It's not explicit from the link, but does it allow the headless login to be resumed from console and vice versa?
I tried some solutions in the past but they did not support that, which is a deal breaker.
There are no headless sessions on Wayland. At all.
You want proper headless session, set up X11 distro and use xrdp - it's really easy. But on wayland "remote support" to something that is already displayed on screen is all you can get now.
What I want is to be able to start a session remotely after a reboot, and continue that same session when I get back home. And conversely start a session while at my desk at home and resume that same session remotely. Without any weird limitations.
In other words, how RDP works on Windows.
So you're saying that is still not possible I take it.
I was working on a carousel library a few months ago. I had made a few stress-test demos so that I could catch obvious issues while I was adding things and tweaking things.
One carousel there had 16K slides.
On Windows both Chrome and Firefox managed that fine. They scrolled from start to end and back without issue and you could see, I think, all the frames in my 60Hz screen.
On GNOME and X11 (dual boot, so same hardware) Chrome was fine but there were issues with Firefox. I was curious so I logged out and logged in with Wayland. On Wayland Firefox was fine too, indistinguishable from Chrome.
I don’t understand hardware, compositors, etc., so I have no idea why that was, but it was interesting to see.
Mighty presumptuous to think everybody uses Gnome. I've spent decades using neither Gnome nor KDE.
What a strange title. X11 is still more popular than Gnome, and formulating a wish like a fact doesn't make it so.
Anecdotally, I strongly doubt this is true, although my environment is probably quite biased. I know a ton of people who use Gnome, some who use KDE, and I think roughly all of these people use them with Wayland. The standalone-WM users I know are also mostly on Sway or other Wayland ones. The only real X11 holdouts seem to be people using X11-only DE's, such as Xfce or Cinnamon.
Xfce is working on Wayland session support. It is working now with some limitations (limitations on what you can embed in the panel are all that's left, I think).
Such a statement is pointless with any data to back it up.
Do you have a source for that statistic?
Stats from Debian and Archlinux. i can't tell how reliable they are.
https://qa.debian.org/popcon-graph.php?packages=x11-common+g...
https://pkgstats.archlinux.de/compare/packages#packages=gnom...
Makes sense to move to something actively maintained. I don’t use GNOME, but I have been on Wayland for years in both KDE and tiling WMs. It works great while on X11 I would often get weird flickering and stuttering issues (some NVIDIA bug I could never track down). Anyway, if the X11 die-hards want it to survive, they need to organize an effort to maintain it, not yell at everyone who wants to build something better.
I am sympathetic to people who have a working setup and just don’t want to mess with their configuration anymore. Unless you’re on OpenBSD, though, that ship has long sailed in most *nix distros (even “stable” Debian). Long-term stability is underrated but hard to achieve.
https://archive.ph/2OaGe
Site is blocked by origin server in India.
Sadly, I recently had to switch back from Wayland to xorg because clients are getting so memory hungry. My eight year old gpu only has 2gb of vram, which I constantly run out of. Some part of the gfx stack should handle swapping out vram to main ram but it apparently isn't.
Another nail in the coffin. Bye, Red Hat. Bye, NGOME. Non GNU OldIBM Mediocre Environment. I wish GTK4 dies in IBM hands too, for the good. XFCE can go back to a community supported GTK3 anytime.
X11 is not going anywhere. If anything it's Gnome adding another nail to its coffin
x11 is in maintenance mode at this point and Gnome is not going anywhere. Gnome is used (and financed) by major distributions.
Nothing new is being created with x11 and the people from freedesktop don't seen to be thrilled to maintain it. I don't think should change just for the sake of changing, but I'd start looking to migrate whatever you use that depends on x11.
I had assumed that XWayland is a drop-in replacement fo X11, and will be available indefinitely.
I regularly write code which relies on a working X11. I have written a virtual machine which makes X11 calls to do 2D graphics and event handling, as well as applications which compile to the virtual machine code. If X11 and now XWayland cease to be available, not only would I have to rewrite large parts of my virtual machine, but also rewrite all the 2D graphics code in applications. All so that I can stand still when the rug is being pulled from under my feet. I'm sure there are others in a similar predicament.
I may be naive about this, but as X11 just works, and has done for decades, it should require little to no maintenance, so why the need to withdraw it? I don't expect, or require, any additional functionality.
Yes, XWayland is intended to continue to be available indefinitely.
For my part, I have no intention of moving off X11 for the next decade at least. The only app I use that I don't fully control is a browser, and the worst case fallback is to run the browser in a Wayland compositor that runs on X.
Are you also still running python2?
What a weird question. I have no use compelling use-case for python2. I have plenty of use-cases for X, such as the fact that none of my software other than my browser has Wayland support, including my window manager.
Nothing weird about it, it makes perfect sense with what you are posting. I find it weird that you react like that to a simple question. Just probing what kind of old software you use, cause it tells a lot depending.
It was weird because you implied (and still are) that it'd mean I was hanging on to old, superceded software that has adequate replacements. I'm not. So this is telling us more about your assumptions.
I'm using my own terminal, wm, and file manager. They use X11, and I have no interest in changing that, because I have no need to as long as X11 works on my hardware and that won't change anytime soon. Everything I don't do in a terminal, I do in a browser.
EDIT: To add some more context for why I have no interest in changing that: 1) my wm is 1568 lines of code at the moment. If anything, that is more than I'm happy with. With Wayland I'd need to write my own compositor. Way too much work even with reusing e.g. wlroots.
2) My file manager is more of a basic desktop launcher. That is fine, and intentional. I may add some features to it. But the reason I'm using that rather than any of the over a dozen options I've tested is that most of them either never had or have ripped out spatial features, and the ones that had some spatial features didn't act the way I wanted them to. I want Amiga-like semi-spatial features of being able to selectively snapshot icon and window placement ("semi-"spatial because traditional spatial would imply a single instance of a window for a given path; I just want default placement to be the same as last time I snapshotted it). Wayland on purpose refuses to allow that, and so I'd need to hack on a compositor or write my own to be able to support the most important feature to me in the file manager.
I'm not going to tolerate my usability being reduced just to switch away from software that does what I want it to, to software that offers me nothing new that I want and takes away features I do want.
I'm curious, what were you hoping to learn about their use of python 2? If you had specific questions, it would be helpful if you ask those, instead of trying to ask through euphemisms.
If it was maintained (security fixes and platform support only), but with no other changes, it would be a very tempting alternative to python3 for the many times API stability is valuable.
And wayland is in broken mode. KDE keep changing the default back to wayland after each update, and every time my linux systems are broken until I switch back to x11.
What is broken for you? At this point, starting from roughly KDE 6, Wayland has been pretty much flawless for me. KDE 5.27 was pretty much fine already as well.
https://github.com/X11Libre/xserver
The problem is that I find Wayland to be a lot buggier than x11.
For example, terminal transparency using Konsole on KDE flickers for me.
Its nearly there, but not quite. Maybe Gnome has no such issues?
I've never seen similar issues using a variety of terminals on Gnome, Sway, or Niri. Haven't used Konsole or Plasma, but I wonder if it's maybe a driver issue?
Do you have more specifics? I just tried it on my machine (Fedora 42, Plasma 6.5.1 Wayland, Konsole 25.08.2, Radeon 780M) and it seems fine for me. Does it only occur occasionally/under specific circumstances for example?
x11 being in maintenance mode is the best thing that happened to it for my use case. It hasn't crashed in 15 years.
It would be sad if, after all those years, it was still missing anything significant. Maintenance mode sounds like a good thing, not something that makes me tempted to switch to some less stable alternative.
and one without the features I use. ssh -X is a must.
> x11 is in maintenance mode at this point and Gnome is not going anywhere
True.
But does not address the fact that Wayland is a bad solution to X11's problems, and that its architecturally broken from inception.
I don't know the implementation details but I can't really complain about the state of wayland today. It used to be annoying to get working many years ago (worse because I had a nvidia gpu). But today I drive a nigthly build of niri, run it by just spawning an dbuss session and everything works. Bluetooth audio, screen sharing, fractional scaling, no tearing, no font blurring. Every utility I needed has been created and works quite nicely (e.g.: wdisplay). I can even play video games with HDR support.
I have a more stable experience with wayland today than I had with x11. Which to be fair was not only because of wayland but because desktop linux as a whole has made a lot of progress in the last years
X11 is far more stable now since they stopped improving it! I haven't had a crash in 15 years.
I don't think it's true that anything is architecturally or fundamentally broken in Wayland (though if you disagree, I'm very curious what you think is so deeply broken).
Most of the issues and slow adoption were because the core protocol was deliberately kept extremely minimal, and agreeing on all the needed extensions took a long time. Don't take it from me, but rather from KDE developer Nate Graham: https://pointieststick.com/2023/09/17/so-lets-talk-about-thi...
As such, anyone who tried it early probably had to deal with a pretty large amount of non-working stuff, but by now the platform is capable of most features people require and the biggest remaining bottleneck is that software needs to use these new APIs.
Window positioning is one that on its own is sufficient to make me ignore Wayland, as it means that without my own compositor with my own extension, I can't get a file manager that will behave how I want it.
Most people won't care, but for a number of us Wayland is stubbornly refusing to support functionality we see as dealbreakers.
That's fair! I believe that window positioning also works on XWayland, though, so running your file manager that way should still work with the rest of the system being Wayland (and Gnome has no plans to drop XWayland afaik).
I believe the main holdup is a desire for Wayland to be usable with e.g. VR interfaces where there is no simple 2d grid.
Out of curiosity, how do you want the file manager to behave? And did you write your own or are you using an existing one that works that way?
It's managing the desktop too, so I'm not sure that'd work unless running Xwayland in "rootful" mode, in which case I might just as well run X.
The VR stuff is a poor excuse - just fail on that scenario. Nobody that cares about window positioning will have an issue with that.
My file manager defaults to re-opening a window for any directory to a previously snapshotted location, like the Amiga Workbench did. And, yes, I wrote my own. It's a few hundred lines of of a quick and dirty Ruby hack talking directly to a pure Ruby X11 binding, which is anothe reason I stick with X - I can throw things together quickly for X. The amount of ceremony, or big additional dependencies, needed for Wayland is ridiculous.
Goodbye to any trace of freedom left on Linux when you combine this with proprietary graphics drivers.
What's the substance behind this claim? It keeps on being repeated but I don't get what it's actually about. Is there anything proprietary about Wayland that I'm not aware of? What's the difference between proprietary drivers using X11 and Wayland?
I don't think I understand what you mean. Do you mean wayland is not usable with nvidia proprietary driver? I remember that being annoying but possible many year ago (with sway --my-next-gpu-wont-be-nvidia thingy).
But if you use really old nvidia gpu you can have a mixed experience with wayland. Which is a fair problem to complain, but you can't blame that on wayland and call that lack of freedom. That problem was caused by the lack of freedom coming from nvidia gpus and how locked down they are and how nvidia for many year has been hostile towards linux desktop.
The “what’s the harm” here is the systemd conversation all over again basically. If you pipe everything through a single point of failure black box users have already lost, when you combine it binary blob drivers that shouldn’t exist it’s worse. Linux is doomed in achieving its most important goals which are user freedom, not someone’s idea of pretty UI imposed at the expense of that. If that’s what users want they should buy a Mac. If you want to get locked out of your OS for eye candy we have that.
Where is this Wayland black box then? If anything, Wayland made this situation significantly better: the X11 server was exactly this 'single point of failure black box' you are describing. Wayland replaces this with a much simpler protocol with multiple independent implementations (notably Mutter/gnome-shell, KWin, wlroots-based ones such as sway, and Smithay-based ones such as niri).
I don't understand how proprietary drivers with Wayland are supposed to be a bigger problem than proprietary drivers with X11, could you explain?
Personally, I've never used a proprietary driver with either.
Open source has never achieved user freedom.
It’s achieved developer and very tech savvy IT pro freedom. If you can deal with command lines and debugging systems you are not a user. You’re a computer professional.
If OSS wanted to bring freedom to users its primary focus would be radical simplification and UI/UX.
What nonsense, especially if framed in a such an absolutist way leaving no room for nuance. Millions of people are storing their data on self hosted or 3rd party run managed OSS services on multiple platforms. Sure, it is not all perfect, UX might be an issue here and there but compared to having to relearn using some proprietary app redesign every other year because some Product Manager needed a promotion, a lot of OSS stuff is perfectly usable, stable and secure.
> Millions of people are storing their data on self hosted or 3rd party run managed OSS services on multiple platforms.
That's open source being used by developers to provide a closed service to users. Users experience it as an opaque closed service.
SaaS backed by open source is actually the most closed model of software, more closed than closed-source software run locally.
Freedom is dead when a single implementation is replaced with several competing implementations implementing an open standard.
Just so it’s clear:
The X Window System (X11) is a protocol with multiple implementations. Sure, the X.Org Server (Xorg) was the most popular by a huge margin, but there were quite a few others (e.g. XFree86, Xming, XWayland), though over time most were discontinued for one reason or another.
X11 and Wayland do differ in an important way: in X11 window managers (GNOME, KDE, i3, whatever) all sat atop the Xorg server; whereas in Wayland there’s only the compositor, so GNOME, KDE, Sway, whatever, all essentially include their own equivalent of Xorg (which could be fully integrated, or factored into a library, such as Mutter, KWin, wlroots).
Only Nvidia use proprietary graphics drivers?
vmxgfx has similar issues.
If you are using VMWare then the proprietary video driver is probably the least of your issues.
X11 desktop environments are dying, but it doesn't mean that X11 is dead. XWayland is still a thing so you can still run your X11 apps on Gnome.
The big reason why I want to keep X11 besides backwards compatibility is the ability to run GUI apps remotely, even from a server that has zero graphical capabilities. But these do not really apply to desktop environments. If you want to remote a full desktop rather than individual applications, there are better options (VNC, RDP, ...).
I think waypipe should generally do what you want, though TBH so far when I need to run a remote GUI app, I just have them display to XWayland out of habit.
Rumors of Gnome's demise seem greatly exaggerated to me. It's still the default DE in nearly all major distributions, and it doesn't seem to have incurred major mindshare or marketshare hits recently. I feel like most of the 'complainers' already abandoned the Gnome ship with the release of GNOME 3.
Really the only high-profile 'switch' in recent times I can think of is that Fedora promoted KDE to be first-class ('edition') alongside Gnome, instead of delegated to a more second-class spin. And while KDE is a bit more conservative in this regard, I believe that in the long term KDE also wants to go Wayland-only at some point.
Personally I did switch from Gnome to KDE some time after Gnome 40, since I quite liked 3.x but the UI overhaul 40 did wasn't really my thing. It also helps that KDE got a lot better in recent years.
Wayland was created by X11 developers, as they decided keeping X11 going was beyond hope.
Feel free to find volunteers to fulfill their shoes.
I think we would have a lot less problems if wayland was called X12 /s
We'd have had a lot fewer problems if Wayland had been X12, in the sense of taking an approach of gradual iteration, even if they went at it fairly aggressively.
E.g. instead of the Xwayland approach, you could've already ditched "half" of Xorg if you stripped most of the server-side drawing primitives and server-side font support and moved them to Xlib, handling it client side, and then made it clear someone else would need to take over maintenance of Xlib, and "started over" with a stripped down Xcb.
You could've validated further restrictions by letting clients opt in to them with extensions before "flipping the switch" and restricting them by default when the damage was acceptable.
Even if we then eventually reached a point where there'd be a schism, odds are it'd be far smaller. And certainly far less time would've been wasted.
X11 isn't going anywhere because distros will ship XWayland for a long time to ensure compatibility with existing X11-only applicatin.
xorg-server is gone from the linux desktop. Gnome and KDE use wayland shells by default, and that's what users get when they download a Debian/Ubuntu/Fedora/whatever ISO.
> adding another nail to its coffin
They've been adding nails to the coffin for 25+ years now. How many more do you think it's going to take?
Damn thing's more nail than coffin at this point!
> X11 is not going anywhere. If anything it's Gnome adding another nail to its coffin
Yup, my feeling as well.
Wayland was sold as a sorely needed fix to X11 long-standing problems.
The fact that X11 had problems that sorely needed to be fixed is indeed true.
The fact that Wayland is the solution is unfortunately not.
Just because something is the next gen project does not mean it actually succeeded in fixing what it planned to.