Xorg is pretty bad, in the sense that it has so many tack-on features that it's unmaintainable. That's a problem from a developer's perspective (and there aren't people who want to develop it anymore, literally almost everybody has given up/moved on).
From a user's perspective, X might have some convenient features (X forwarding. I love it!), but the two big wins for Wayland are enhanced security, where applications can't just read your input unless you allow them tom and lag/tear free rendering.
A lot of people complain about Wayland issues, and they're justified complaints when you take one thing that currently works and substitute in its place something which might not work yet, but Wayland's time is near. It's design might be
Another set of complaints is how some Wayland things work on one implementation, but not another. This is partly a complaint about how the protocol is defined, and another about how different implementations might introduce non-standard extensions. Personally, I don't care about this so long as things get standardized, but of course this is one of the weaknesses of FOSS - different people, different opinions, disjointed work and effort. At least with Windows and macOS they get one implementation (for better or worse) and that's that.
If you want to compare things, I think Wayland can be considered equivalent to DWM on Windows. And with PipeWire (the Linux analog to macOS's CoreAudio) en route, I think Linux desktops are going to get really strong in the next year or two. Hopefully I'm not spreading bogus info in this comment.
I don't know too much about either so this may sound really dumb BUT
xorg is basically dead, even it's developers agree that it's time to move on. it's necessary to move to something that can updated and be maintained and be secure. It's approach to display managing is really different so it really is a significant change, but it also makes it much lighter. It's been in development for about a decade now and it's finally almost there. With full gnome support, to the point where wayland is the default on some major distros (I want to say fedora is one, and ubuntu is going wayland this year) NVIDIA has also made some changes recently to support it more (though I still think it's rough from what I've heard) and now kde ready for wayland, I really think this year will be the tipping point. At least I hope.
I believe manjaro uses wayland as default (though it's been a minute since I last had a linux pc, so don't take this as gospel).
This could be placebo, but whenever I use Wayland, it feels like there is a bit less input latency. Everything seems slightly snappier, which is very nice.
Interesting, used KDE way back. I'm kinda hooked on GNOME 3 now.
The main thing stopping me from giving KDE a chance is its use of Qt. The Qt company gives off extremely scummy vibes, and the fact that KDE needs a special licensing agreement with them is gross.
I'm not bothered by what the Qt company does, since the special licensing is airtight and guarantees Qt in its fullness is available to KDE under GPLv3 or BSD even if they go under or go full evil. If it ever happens, it would take months for upstream Qt changes to affect my desktop, plenty of time to migrate to MATE or whatever.
Screen recording/sharing and keyboard automation primarily. And there are one or two Compiz accessibility plugins which aren't replicated that I used to use.
Aside from that, I'm tired of the project's lack of coherency and its insistence on deferring to the compositor. The project will be as old as Xorg before it will work 1:1 with what it intended to replace. It's been what, 12 years now?
that makes sense, especially keyboard automation because I have no idea how that's going. But I do think that even if it does take until it's as old as xorg it won't have the same problems xorg does now with it being too bloated and almost impossible to maintain.