Although Wayland has been GNOME’s default session since 2016, X11 has continued to linger in the codebase—until now. That changed with the recent merging of two PRs (here and here), which completely removed the X11 codebase from both Mutter, GNOME’s default window manager and compositor, as well as the GNOME Shell itself.
In other words, the GNOME project is finally closing one of the longest chapters in Linux desktop history. With the upcoming GNOME 50 release, scheduled for mid-march 2026, the desktop environment will officially drop support for the native X11 session, making Wayland the sole display system moving forward.
It is pretty hard to improve if you are not allowed to change anything.
Yes, the design of Wayland means that some of the techniques that work on X will not work on Wayland (on purpose). So yes, some apps have to be adapted to use the techniques that do work on Wayland. And no, changing Wayland to support the old ways is not the answer (because they were changed on purpose).
Wayland has been criticized for taking away previous capabilities before providing new ways to do things. That is a fair critique, though somewhat par for the course when replacing old tech. But at this point, almost everything necessary is possible and Wayland users are in the majority (the massive majority soon).
At this point, it really is the apps developers responsibility to support Wayland properly. I mean, they do not have to of course but that means their app will be broken for 80% of Linux users on two years (and more than half today).
You’ll notice the vast majority of the complaints and problems people had about systemd went away when it launched into distros and invisibly did everything init did, often times in a better way. They made big improvements WITHOUT crapping all over everyone’s legitimate complaints and expected features, and things went fine, to the point where I’d say even most naysayers begrudgingly admit they were overreacting.
Compare that to Wayland, which not only didn’t do that, it makes it worse by actively gaslighting power users, mocking them for powerful features in their workflow and demanding they make changes. It’s been active and spreading for years, and the complaints people had on DAY ONE are still there.
It didn’t work for Mark Shuttleworth when he pushed Unity, and it shouldn’t work now. This is not the proprietary Microsoft world where developers can decide they have a new, better way of doing things and take a wrecking ball to the workflows of millions of active users and then scream “my way or the highway”. I do not need others to tell me what is good for me. That attitude belongs in the closed software world.
I don’t particularly care that 80% of users are on Wayland. It’s not relevant - if I cared about what the majority of users were doing, I’d be on Windows. None of that fixes my inability to use Wayland for my daily tasks. I will continue to use X if I have to build it myself and exclusively run outdated desktop environments. I refuse to give up my functionality to add a layer of security that breaks my critical use cases with no option to opt out.
What functionality do you so desperately need that is not available or workable in Wayland?