The Regressed State Of KDE Plasma On Wayland, But Things Should Get Better

[schmidtbag]

I've been using KDE with Arch for a couple months and it is mostly stable. Plasma desktop is currently the glitchiest aspect about it.
CPU usage had overall gone down since switching away from X11.

[Charlie68]

What I do not understand is what is the problem? Kde on Wayland is development and I think it will take time, so normal that there may be regression. What is the news?

[leipero]

I don't track all the distros, but Wayland is working great on openSUSE Tumbleweed with Gnome desktop. And I'm almost certain it was available with Gnome on Arch last time I used it.

Actually, Arch was first distribution to use default GNOME wayland session, It's not because of Arch, it's because of GNOME, since Arch use "vanilla" packages, not patched ones.

[boxie]

What I do not understand is what is the problem? Kde on Wayland is development and I think it will take time, so normal that there may be regression. What is the news?

the problem is that the carpet was yanked out from underneath the Kde teams feet with changed between Qt5.7 and Qt5.9 - and the news is that wayland will take a bit longer to get out the door

[cipri]

I have stressed already so many times, that in my opinion one should remove the QT dependency.

[Vistaus]

I remember I was on ubuntu 10.04 when it was rumored that ubuntu 10.10 would have default wayland. We're in 2017 and apart from Fedora nobody uses wayland! Then Ubuntu talks about Mir and suddenly wayland becomes indispensable. Something will not come back and anyway, years will pass before the wayland will be used by many distributions, if this happens!

Wrong. Fedora is the only *major* distribution using Wayland by default but by no means it's the only distribution to use Wayland by default.

[Luke_Wolf]

the problem is that the carpet was yanked out from underneath the Kde teams feet with changed between Qt5.7 and Qt5.9 - and the news is that wayland will take a bit longer to get out the door

Exactly, it pushes things back. Not in a world ending way, but it's the kind of thing that tacks months onto development time.

I have stressed already so many times, that in my opinion one should remove the QT dependency.

You do realize that what you're proposing is not just difficult but basically impossible to do right? It would be far easier to rewrite everything from scratch in Rust than it would be to remove Qt as a dependency. Qt isn't just a toolkit, it's a platform. To remove Qt is the same level of difficulty it would be to remove the Java BCL or .NET Framework from a very large codebase from the respective languages. Pretty much everything would have to be rewritten and tested... possibly even completely changing the idiom of the code. It is so difficult that the balance is seriously heavily in favour of just a from scratch rewrite without the dependency in the first place. If we're just keeping things C++ though what's the advantage? What value does rewriting it without give you? The answer isn't a whole lot of positive, maybe some bad architectural design choices here and there revisited and fixed, and you're probably going to have a much buggier situation for many years until everything is properly shaken out again.

[AdamOne]

i'll use an older Qt-version, i just want wayland, please!!

[R41N3R]

From my point of view since Qt 5.9.1 the latest default Plasma Desktop with Wayland works quite well. You will find issues mostly when adjusting the desktop too much, but I'm sure this will improve quite fast. Often issues you see in KDE are related to graphic drivers, input and Qt bugs. The whole stack needs small improvements. But this Qt bug Martin talks about is/was really bad and I don't understand how Qt could release two versions in such a bad state where one of the biggest Qt projects basically has been regressed that much. They didn't even consider a bug fix release for Qt 5.8. Not nice at all. I hope Qt will take this bad example seriously.

[Charlie68]

... what I meant was that I do not rush into wanting wayland that at user level changes little or nothing. I am not saying that wayland is bad, but simply that in this development phase it is normal for dev to encounter problems and regressions. It will take years before the wayland will definitively replace Xorg. GNU / Linux is not just Gnome, but Kde, Xfce Lxqt, etc. Without counting driver issues. But to the end user these things matter little, the important thing is that things work well that either Xorg or Wayland does not matter.