Dale <[email protected]> writes:
One big problem they
likely have, making it work with the current X11 system as
well. If it
was designed to work only on software written to work with it,
it could
be easier. Having to deal with both and make Wayland compatible
with
both, that could be making things harder. I'm not a dev so just
my
thinking.
Xwayland runs an X server on top of a Wayland compositor, and - as
i know from direct experience - that's often sufficient for many
programs that aren't explicitly Wayland-aware. (For example, The
GIMP 2.x series.) That said, i know there are indeed programs
where this is insufficient; i can't remember any off the top of my
head, but perhaps KiCAD?
For the record. I'm not against Wayland. It's just not ready
for my
use yet.
Disagree.
My experience has been that statements like this aren't helpful,
because people assume "Not ready for my specific use-case(s)"
means "Not ready for anyone's use cases". That's a strong claim,
and demonstrably incorrect. And it goes both ways: there are
people for whom Wayland is a significant improvement over X, and
so say to X users, "No, you're wrong, it _is_ ready." But again,
that's incorrect; different people have different use-cases.
i started doing my own investigations for Wayland because i
realised that it's coming down the pipe, and i wanted to add and
update the wiki's information about Wayland, to make sure that
it's factually correct. There haven't tended to be enough
volunteers willing to work on maintaining X, until the recent
appearance of XLibre; most of the Xorg devs don't want to have to
wrangle the X code base anymore, and prefer to work on
Wayland. (This includes Xorg dev Matthieu Herbb, the primary Xorg
dev for OpenBSD + Xenocara; the last slide of his 2023
presentation at
https://2023.eurobsdcon.org/slides/eurobsdcon2023-matthieu_herrb-wayland-openbsd.pdf
says "X11 is fading away / Wayland is the way to go for graphical
desktops").
i've now been using Wayland basically exclusively for a couple of
years, only rarely facing any significant issues. It's basically
ready for _my_ use case. And there are people for whom Wayland
solves screen tearing issues they experience under X (e.g. while
gaming). That's not an issue _i've_ faced, but it's a significant
issue for others.
Still, there are certainly various use-cases that it's not ready
for. Here's a KDE community wiki page:
https://community.kde.org/Plasma/Wayland_Known_Significant_Issues
Here's a in-depth post by a blind user about the state of
Wayland's accessibility functionality - not adequate, but at least
being actively worked on:
https://fireborn.mataroa.blog/blog/i-want-to-love-linux-it-doesnt-love-me-back-post-4-wayland-is-growing-up-and-now-we-dont-have-a-choice/
(And for their pains of trying to give a factual assessment of the
state of Wayland's accessibility functionality for _their_ needs,
got called a 'Wayland shill'. Go figure.)
Wayland-on-Nvidia also seems to be an issue for a number of
people, with some saying "How is Nvidia hardware not supported in
2025??" But the fundamental issue here isn't Wayland people not
caring about Nvidia hardware; the long-standing issue with Nvidia
on Linux is that Nvidia continues to not play nicely with FOSS
devs, guarding how their proprietary drivers work, making it
significantly more difficult for FOSS devs to add working
support. And this has been an issue on X as well, at least in the
past; not sure where things are currently at re. the Nouveau
driver. i deliberately don't use Nvidia hardware for this reason,
but i know that there are people in situations where Nvidia
hardware is the only practical option for their use-case(s).
Many, if not most, of us use Gentoo because it's more easily
adapted to a variety of use-cases that other distros don't cater
to. Gentoo tries to provide choice as long as there's the
volunteer capacity to do so. For some of us, Wayland is more than
ready; for some of us, it's really not.
Alexis.
