The publishing of v1.0 of a protocol has little to do with the stability of software implementing that protocol. The software in question is in fact KWin, the KDE window manager. It is the most mature and featureful Wayland compositor, and the KDE devs have done a lot of work over the past few years to get it competitive with its X11 backend. However even they themselves still consider this a transitional phase: https://www.phoronix.com/news/KDE-Wayland-Is-The-Future
Also, time is in any case less relevant that developer hours. Were I feeling snarkier, I might point out that GNU/Hurd was first released in 1990 - perhaps we can finally switch to that...
I'm still baffled that on a platform where fragmentation is a serious issue it was decided the best course of action is to have every DE implement it independently. We are going from every DE sharing certain settings and configs to them all doing it differently, maybe not even implenting certain aspects even on the same OS.
The presenter (an X11 dev) argues that it was sorta already the case. That, nowadays, X11 WMs are thick, do much of the work themselves, don't use most of the X server's capabilities, and it's just a weird man in the middle between applications and the WM.
I don't have the experience to confirm or deny this, though
Yeah, this is the biggest headache: so many things have gone from "you can do this with basically any WM and DE" to "only works on GNOME, or KDE, or this obscure compositor, hope you don't want some combination where the intersection is zero!"
I'm mostly seeing this sentiment from people who have little to no experience using Wayland. I've been using Sway as my primary environment for 5+ years now and it's been working great. What are the actual specific features that you believe are missing? Copy/pasting works, screenshots work, video recording works, screensharing works, before you try to parrot some ancient stuff. Post-Sway Wayland (even if you don't use Sway itself) is really a whole different beast. Drew and emersion and co. really sped things up and a lot of new compositors have sprung up since then with similarly good or better support for things. Splitting out wl-roots from Sway so that other compositors didn't need to build as much from scratch probably helped, but some compositors are still doing their own thing successfully enough. Hyprland seems to be pretty popular these days. Even Pewdiepie is using it, last I heard. They even added a tearing protocol for people who think they need to have tearing to get just a bit more performance out of their games and such. On the note of games, the Steam Deck's Game Mode uses Valve's Wayland compositor, Gamescope, and has been greatly successful as well. Some people are even running Gamescope on top of other environments on desktops because of how well games work in it.
The only thing I've heard isn't working in the last year or two is stuff related to accessibility for blind users. Screen readers and related software aren't working as well as they should yet. I don't mean to dismiss that as unserious, by the way, though I would wager that the vast majority of people don't use/need those features and should've been fine to switch over years ago. I hope accessibility improves ASAP as well for those that need it (and sorry if you're one of those people).
Maybe your use of Linux is limited to those things that happen to work. As far as I understand, Wayland is deliberately incomplete and a number of problems people have stem from design decisions.
> These problems exist because Wayland’s design omits basic functionality that desktop applications for X11, Windows and macOS have relied on for decades—things like being able to position windows or warp the mouse cursor. This functionality was omitted by design, not oversight.
Wayland might work for your or my personal selection of work cases, and currently my stance is "when it's done it's done, let's wait and see". But it might also end Linux support for entire classes of software and use cases.
