Linux Mint’s Cinnamon is one of the last desktops to still not support Wayland, and is relegated to only being compatible with legacy X11 environments. With the next release of Cinnamon, however, this is finally going to change.
We worked really hard on Wayland and we got to the point where it feels solid and the experience is almost on par with X11. Wayland support will no longer be considered “experimental”. In the next version of Cinnamon, both X11 and Wayland will be fully supported.
↫ Clement Lefebvre on the Linux Mint blog
The next release of Cinnamon, version 6.8, will be part of the next release of Linux Mint, scheduled for Christmas of this year.
I really enjoy using Cinnamon, so will be very interested to see what it’s like. I don’t care about X11 vs Wayland. I just want it to work and not be slow.
Shifu,
I also have Cinnamon desktops in the house. I also don’t care about wayland/X11 as long as things work.
I needed to make use of xrandr to set things up correctly for HDMI on the TV: this tool and some of it’s features do not work with wayland. I don’t mind so much rewriting scripts to work with wayland…but I’ll be annoyed if wayland doesn’t do what I need.
Another huge issue with wayland for me in the past has been remote screen sharing. Same deal here, I can switch tooling if needed but things had better bloody work when I do, haha.
What I don’t understand (and it’s probably because I’m a just a user), is why every project has to do so much work to get Wayland running on their distro and performing basic functions, when Wayland itself has been in development for 18 years? All I can say is it better not be laggy by the time I get to it.
Because Wayland isn’t a piece of software per se, it’s just a protocol determined by a governing body with bureaucratic processes that remind me more of politics than software development. They release the Wayland specification (they also release Weston, which is software, but it’s basically just a proof of concept sandbox) and then all of the desktop environments and window managers for Linux that want to support the Wayland protocol have to either write their own code from scratch, or else try to shoehorn in another library that implements basic Wayland-compatible functionality such as clipboard copy/paste, window management, keyboard layouts and shortcuts etc. that used to be handled by the Xorg server. If this seems like an insane duplication of effort and unnecessary fragmentation, it’s because it is.
rahim123,
I agree, we lost some of X11’s benefits. X11 typically meant all desktops could support and work with the same features – KDE, Gnome, Cinnimon, Mate, etc running anywhere on a desktop or raspberry pi – it didn’t matter. Wayland’s approach on the other hand leads to much more fragmentation between desktops, which hasn’t been great for the community. Some people would say “it works for me”, but that meant nothing as wayland doesn’t imply the feature is available and working everywhere.
While I know X11 was antiquated, I think wayland made some regressive decisions. But at least hopefully most of these pain points are behind us. Work has been progressing with desktops independently filling out the feature matrix. Nevertheless, wayland’s design still means that bugs and even performance profiles could be wayland compositor specific. This wasn’t the case with X11 where all the desktop environments pretty much shared the same X11 implementation.
I couldn’t agree more. I’m sure that Wayland works fine for most common apps and scenarios, but there’s always going to be edge cases, especially with proprietary software for things like remote access and screen recording, which for Linux is already a small niche, and then add the fragmentation minefield of Wayland implementations on top of that. What really kills me about the whole raison d’être of Wayland is the talking point “X11 is insecure because any malicious app can read any other app’s keystrokes!” OK, fine, so don’t run malicious apps on your system, problem solved. The real reason for creating Wayland seems to be first and foremost to stroke the egos of people that enjoy governance, I mean just look at this thing:
https://github.com/cocasoftware/wayland-protocols/blob/main/GOVERNANCE.md
rahim123,
Well, personally I think APIs should be secure and not a free-for-all and it’s clear that X11 wasn’t designed with application permissions in mind. However I still find that many of the arguments that X11 APIs can’t be made secure are disingenuous. For example, taking screenshots IS a legitimate feature, but it’s both justified and technically achievable to have access control lists that limit the applications that are allowed to do it. Adding ACLs to X11 would not have needed much effort.
Anyway the point is mute because wayland devs were set on clearing the state and starting over. Replacing X11 doesn’t bother me because I was never particularly attached to it, but I am critical of poorly executed rollouts that cause friction and prolonged delays. Something I feel that linux projects often get wrong is failing to incorporate & respect community needs from the start. Most of wayland’s issues were foreseeable and should have been solved before they were created. Although TBH I’m betting future projects don’t learn from past mistakes, we’ll just keep repeating them over again. Haha.
Copy-paste, window management and keyboard layouts are in the core Wayland protocol. I think shortcuts are too, if only because the compositor receives all events.
On the other hand, window management is, famously, not part of X11. Shortcuts only exist because all X11 apps can listen for keyboard events.