Wayland and X.org are both part of freedesktop. Whatever maintenance is still happening on X.org is mostly being done by people who primarily work on Wayland. There isn’t some kind of holy war going on between The Wayland Developers who want to kill X.org, and The X.org Developers who believe it is great and want to keep it. They’re nearly all the same people, and they all want X.org to die. AFAIK there isn’t anybody who is actually clamoring to *do the work of maintaining X.org upstream*. There are people who don’t want it to die because Wayland doesn’t yet have the features they need or the NVIDIA proprietary driver doesn’t work well on Wayland or whatever, but AFAIK, none of those people is actually volunteering to maintain X.org long-term.
↫ Adam Williamson
There’s really no clearer summary of the current state of affairs than this.
The distributions are part of this as well. I had issues with MATE-Panel and some of it was caused by Debian’s packages not building panel applets in-process. I ended up logging bug reports with upstream and Debian over it. There’s still a lot of broken things on MATE/Wayland e.g window buttons aren’t scaling with the panel. so they run over menus. The notification applets also aren’t drawing on Debian. This could be because of Debian’s packages, the version of code they currently run, or it might be something MATE hasn’t implemented yet. The problem is an end user doesn’t care why it’s broken. They care that X11 currently runs MATE better than Wayland does.
I wouldn’t call users only caring if something works or not a problem. If anything, maybe that’s all they should care about. End-users using knowingly broken software creates unnecessary headaches for themselves and distractions for the developers. Their use-case may not afford them the luxury of constant willful problems.
@Darkimage Nailed it!
Wayland Devs need to stop badgering people to use broken software and just get on and fix it, the desktop is a field of dreams, build it and they will come!
Ah yes, the usual gnashing of teeth. Except there is no reason to. “Wayland isn’t ready!” Yes it is, but some projects still need to catch up. So some desktop projects still need to port to Wayland. They will do so and for the foreseeable future they will have a functional display server in X.org. Some applications need to be ported, but they will do so or be superseded by Wayland equivalents. X.org may be on life support, but it won’t disappear overnight.
When it comes to “end users” and the claim they “don’t care why it doesn’t work”… It’s simple, if you didn’t buy your machine pre-installed and kept the vendor provided distro, you went off the beaten track and are now responsible for making it work. A vendor will make sure they ship working configurations. Any major alteration you make in the system is up to you, but also your responsibility. It’s the same for Windows. Built your own system? You get to keep the pieces if it breaks.
Stay on X11 as long as it isn’t feasible to switch. On the other hand, don’t expect all pieces of legacy tech to be ported. There surely will be some packages that will fall by the wayside. Which is fine. We aren’t using a.out anymore either. If Wayland is so shiny that you can’t resist, make your equipment and your choice of software Wayland compatible.
r_a_trip,
It can’t be declared ready, at least not in good faith, when it’s not working. ONLY when everything a user needs is working, that’s when it will be ready. It’s really no more complicated than this.
There’s a chance everything will be working well enough for me in 2024…but every time someone has said it’s ready and I’ve tried it…it absolutely was not ready for me and I feel strongly that user needs need to be respected rather than dismissed.
And you don’t understand that Desktop Linux is moving away from the “bit and pieces” approach towards being more integrated. In other words, if Wayland works satisfactorily on Ubuntu, Fedora, and Steam Deck, then X.org can be abandoned.
And you know what, that’s a good thing. Desktop Linux needs to grow up into a commercial product and not be a playground anymore.
And that’s your first problem. You think this is just about Linux.
The BSDs needs to form a contingency plan for X11, because Red Hat is going to end support for X.org. Red Hat are the last ones putting serious money in it. Don’t get caught without a viable display system. Scrambling while your default display platform is already crumbling is not a situation you want to be in.
@r_a_trip:
OpenBSD forked X11 a long time ago, their version is called Xenocara and they fixed a lot of the security issues with X11. I’m not a OpenBSD developer, just an enthusiastic user, so I’m not privy to any plans to deprecate Xenocara if such plans exist, but I doubt it is going to happen in the foreseeable future.
There is a Wayland port in the works for OpenBSD, and there is also Arcan being developed as a display server/game engine combination that is quite different from both X11 and Wayland. As for the other BSDs, I have no clue as I don’t use them at all.
kurkosdr,
Sometimes you have an anti-linux vibe, so I’m not sure if you’re setting me up here or not. But…do you realize the irony of what you are stating?
X is old and creaky and I know it. I’m not against a wayland replacement, but that doesn’t automatically make wayland suitable. Growing up and not being just a playground requires broken use cases to get fixed and get feature parity with windows and macos (to say nothing of other linux desktop APIs). Things like remote desktop should have been supported 7 years ago and should have been ready long before covid. But due to bad planning and direction by project leaders the implementation of certain critical use cases has been unnecessary rough leading to “wayland doesn’t work for me” backlash. Not for nothing but microsoft and apple would face the same backlash if they tried to pull this “use our new desktop, but ignore the fact that we broke use cases and applications that used to work”. I’m sure you’d agree that this generates anger and frustration among users.
I can get behind that statement, because it means Wayland is absolutely not ready for prime time yet. It is broken in one way or another on all three of those platforms. Put Ubuntu or Fedora on a system with a Nvidia card and watch Wayland fall on its face. Try to connect a Steam controller (you know, made by the same company that makes the Steam Deck) to a Steam Deck, and watch Wayland fall on its face{1}. In any of those situations, fall back to X.org and *everything just works*, no fuss no muss.
So you are absolutely right, Wayland is currently not ready for prime time. Thank you for proving the rest of us right!
{1} https://old.reddit.com/r/linux_gaming/comments/oraq9t/if_someone_was_wondering_if_the_steamdeck_runs_on/h6jocvc/
I don’t know how moving from a display server used by 98% of the users to everybody doing his own shit and not sharing the work (except the wlroots users) is more integrated.
I understand the need for specialized servers like Greenfield but somebody has to convince me why regular desktops environments like KDE and GNOME having completely different code bases is a good thing.
Seconded!
We used to have one display server (x11) and now we have > 25 implementation of ‘modern’ display server protocol (wayland).
Another problem that’s confused me is: ‘There’s no manpower to maintain X11.’ OK. But looks like there’s a bunch of manpower for those fragmented protocol.
Final Cut Pro doesn’t work on Linux, so Linux isn’t ready for every user. Nor does full Office Suite, so Linux really isn’t ready for every user. The goal of Wayland is not to be a drop in replacement for X11 with provide full compatibility with any piece of software that was written for X11. It’s not Wayland’s responsibility to create full backwards compatibility for every piece of software. The software will have to be rewritten to work with Wayland, not the other way around.
Patric,
I don’t accept the argument that wayland isn’t ready equates to linux not running windows software, but taking your argument for what it is… By your logic, wayland promoters should stop portraying wayland as good enough for everyone. Is that what we are seeing? Or is team wayland trying to tell everyone to switch to wayland even though some problems still remain?
IMHO wayland hurts itself more by breaking things that arguably should have been included from the get-go. Respecting user needs goes such a long way and I honestly believe that under better project management these issues would have been sorted out nearly a decade ago. It is what it is, it’s just a shame that community needs get deferred for so long.
The point I was trying to make about Final Cut Pro and MS Office is that no one complains that Linux needs to contort itself to make sure these are compatible on Linux. Apple and Microsoft would be the ones responsible for building these to work on Linux. Another example: you’re not going to scream at Sony if an Xbox game doesn’t work on Playstation.
In the same way, if an application that was originally written with X11 then the software creator has to make it compatible with Wayland.
I’ve been using Wayland since Fedora 34 on GNOME and KDE. I have not logged into an X11 session all but a handful of times. Those all were to use screensharing in Zoom, which Zoom needs to fix not Wayland.
Patric,
Sony users aren’t screaming at Sony to play xbox games, but they absolutely would scream at sony for breaking previously supported features.
Major companies including canonical, zoom, raspberry pi foundation had to postpone wayland deployment over missing features.
There’s still a lot more X software than wayland software; making xwayland more feature complete would go a very long way in addressing everyone’s needs. But even ignoring what wayland devs could do to improve the situation for users, the software compatibility issues you bring up are not limited to X software. Unfortunately wayland created feature fragmentation even among wayland implementations. So remote desktop software that works on GNOME won’t necessarily work on other wayland desktops. This was an obstacle for me last time I tried wayland a few months back and honestly is one of the reasons I am unable to switch to wayland. If they manage to fix this, then great! I’ve literally been waiting years. But until then I stand by what I said originally:
Disagreeing with this is actually part of the problem holding that’s holding wayland back. Wayland promoters shouldn’t be blaming users, they simply need to fix the problems that are causing adoption barriers. Then users including me will go adopt it.
@Alfman,
This is what is so absolutely frustrating about Wayland zealots. They will say “Wayland is ready, and if it’s not, let’s blame the users!” Then the users say “we only say it’s not ready because the software we use doesn’t work with Wayland yet.” The zealots change their tune and then say “Wayland is ready, and if it’s not, let’s blame the software developers!”
Zealotry of any kind is unhealthy and nearly always damaging to the subject, whether it’s religion or software. It’s always a blame game and never a strategy to work with the users and the software developers.
And before someone misunderstands me, I’m not talking about the Wayland developers here; from every indication including the message that is the subject of this article, they *want* to have feature parity with X.org and they *want* Wayland to work properly with all the existing software out there. It’s the armchair devs who couldn’t code their way out of a wet paper bag that are so zealous about Wayland “already being ready, fuck anyone who dares to say otherwise” who are the real problem. The actual Wayland devs probably hate that kind of zealotry but they can’t do anything about it except continue working on and improving the project. More power to them, and I hope they reach their goals before the inevitable replacement for Wayland comes along.
@Morgan
I am one of those “Wayland zealots”. We are Windows users witnessing the gradual enshittification of Windows and looking for the exits. Think of how we were still using Windows Mobile during the HTC HD2 era but we were also looking for the exits because we knew the time to abandon that ship was fast approaching.
Thing is, all successful Linux or Unix platforms had to ditch X11 for something better: MacOS, Android, Enigma2, and even the Steam Deck (note: Steam Deck uses Wayland for game mode, aka the mode that matters). Graphics and UI performance matters a lot and X11 drops the ball badly in that department. In other words, we consider Wayland a precondition for Desktop Linux filling Windows’ shoes, and if some things are missing for this or that use case, we don’t care, we are going to lobby for Wayland anyway.
In case you were wondering why Wayland zealots look a lot like Windows fanboys… now you know.
kurkosdr,
I think it’s logically to say that both windows users and x users wanting to switch are kind of in the same boat. None of us are against wayland, we just want it to work. This is a very reasonable precondition regardless of where we are coming from.
Honestly though? What doesn’t work on Wayland anymore? If it’s obscure DEs (aka anything else than Gnome and KDE), then we Windows users don’t care, we aren’t used to having a choice of several DEs, so that’s a conflict of interest that we have with old-timer Desktop Linux users. Similarly, if remote desktop doesn’t work with obscure DEs, we don’t care.
That also replies to the whole “Wayland forces DEs to re-invent stuff”: We only care about Gnome and KDE, so those DEs being able to tailor things to work best for the particular DE is a feature, not a bug.
BTW another philosophical difference Windows users have is that the OS should try to support every hardware in existence. For us, the hardware vendor brings the driver that makes the hardware work, not the OS vendor. So, we don’t care about Nvidia either. It’s nice to have, but the Steam Deck and presumably any subsequent hardware will use AMD GPUs and any Steam OS-compatible builds will be spec’ed with an AMD GPU.
kurkosdr,
I glad wayland works for you, really. But you’re wrong to assume those of us holding out expect unreasonable things. For me it’s remote desktop on KDE and flawless support for my nvidia hardware.
I’ll quote you…
When wayland reaches the point where it does everything I need out of the box, that will be fantastic! But when some critical things don’t work out of the box, then it’s kind of obnoxious to tell other people they need to switch or that somehow they’re the problem, that’s just silly. There shouldn’t be a controversy, more people will adopt it when their use cases get fixed and hopefully 2024 is the year this happens for me.
It seems you judge Linux as a commercial product. Unless you get it with either a Linux preloaded machine or with a support contract, it isn’t a commercial product. A distribution is an amalgam of software, selected to make an operating system, to the best of the abilities at the time of minting. Linux users, outside of the commercially supported realm, are very much part of the development process. That hasn’t changed over the course of it’s life.
Wayland is ready. It’s an assortment of software projects wich are not. Xfce, Mate, Budgie, Cinnamon et al. are all still catching up and they aren’t ready. They could have been if they hadn’t been asleep behind the wheel. Wayland was announced more than a decade ago and it was clearly designated as the successor to X11 on Linux. Most projects were dragging their feet. Now they need to scramble, now that Red Hat has made clear they will be pulling the plug on X.org as the last serious bug fixer.
Also, comparing X.org at the end of its life against Wayland at its beginning isn’t a fair comparison. X.org works well, most of its niggles have been solved on the surface of its long, long existence. However, internally X.org is a mess. Paradigm-wise the technology doesn’t fit in the current landscape anymore. The core development team is more than ready to burn it at the stake, instead of piling on work around upon work around to keep it working on hardware that isn’t designed with it in mind anymore.
I do remember all the niggles that came, first with XFree86 and afterwards X.org along the way. The X.org team did a stellar job on keeping X.org going and making it as pain free as possible, but X.org hasn’t been that bastion of pain free use from the get go. We’ve gotten used to it disappearing in the background. Wayland will do the same soon enough, but not if we keep it at bay waiting for it to be perfect.
Yes, transitions are not pain free, but they do make things better for the long run. It’s also part of the Linux process. Linux has never been as cushy as it is now, but it hasn’t always been that way. Very bumpy stuff happening during the course of its development. Linux has come a very long way, certainly when I compare it to Red Hat Linux 5.2. KDE and Gnome are ready for Wayland. The rest will follow shortly.
r_a_trip,
I know it’s not a commercial product, kurkosdr suggested it should be in his comment.
There’s a difference between you saying it’s ready and it actually being ready. When it comes to other user’s needs, it’s their call and not yours whether it’s ready for them. In a few months I will probably try it again and it might well be ready then (fingers crossed) but ONLY if the problems have actually gotten fixed and not because anyone here wants to declare it ready!
I have no issue agreeing with you here.
I agree with this too, although with the caveat that much of the pain could have been avoided if only project leads were more mindful & respectful of user needs. Honestly wayland should have addressed our concerns years ago and we wouldn’t even be here because the transition would already be successful.
Sometimes in order to make change actually happen, you have to get out the hammer. Hammer is out. Now everything must change. With that said, there have been “hammer” like moves in the past. And even then, there has been an “option” to continue on with what you have. This one is a bit more serious. Likely some of the lesser maintained products/projects will “die”. Some people will be mad, but, if you “don’t like who’s running, you need to enter you’re own hat into the ring.”
As a lowly end user I just wish they’d focus on one and get stuff working. Nowadays I mainly use Linux inside a VM (I prefer it this way because I like using snapshots and periodically starting fresh without having to wipe the bare metal OS).
I had to move away from both VirtualBox and VMWare because of graphical issues when my work machine got upgraded to Windows 11. I ‘d have one set of issues with X, and a different set of issues with Wayland, and not trivial issues. With one I’d get graphical rendering issues on simple applications like XChat, and the other had such poor performance it became unuseable.
Now, I’m using Linux through WSL for the time being. Maybe when Ubuntu 24.04 comes out I’ll try VMWare or VirtualBox again and see if anything has changed.
“The X.org Developers who believe it is great and want to keep it. They’re nearly all the same people, and they all want X.org to die”
wait, so they want it to die, but they want to keep it, but they won’t maintain it?
did the author get a stroke or something writing this nonsense?
Xorg is in maintenance, no development is happening and everyone is moving to Wayland. it’s not ready for every use case, but it does the job for many people already.
I don’t think it’s fair to dismiss concerns about Wayland. The opposition is often obnoxious, but I don’t think it’s really a defense of Xorg (which has always been a pain to package and maintain). But it did let you do all sorts of cool, hacky things, and new window managers were quite trivial to write. It was fun. By contrast, new features for Wayland are allowed one-by-one by committee, and compositors are huge projects.
I’m sure the transition will happen in the end, but is it necessary to be snarky about it? It think it’s a little sad, a small part of a trend where things tend to become slicker and more beyond our grasp.
Wayland apparatchiks carry on like our complaints are some sort of Xorg counteroffensive, but it’s hardly the case for the bulk of users.
We don’t need Wayland to be “better” as it’s frequently claimed to be, we don’t need Wayland to be “faster”, we don’t need Wayland to be “smoother” and we don’t need Wayland to be “More reliable”. We really just need Wayland to be “equivalent” and most of us will adopt it, after that point every step forward is a bonus.
Those “off the beaten track” hardware solutions I and other nasty subversives use that “Break Wayland” seem to be heavily dominated by Dell, HP with AMD and Nvidia discrete graphics. Foolish me buying such niche hardware from boutique vendors.
Could Wayland work for me, yes I’m sure it can,
Could I script solutions to resolve Wayland’s OOB failure, yes I probably could.
You see, at the moment for Linux I have A/B configurations, I arbitrarily choose one or the other subject to the resource it’s being deployed on. Simple, streamlined, and it works OOB. To get Wayland working I’d have to go from “AuB” to {A…..Z}. I’d actually have to become the bespoke solution provider that I’m already accused of being.
So will I move from the functional Xorg OOB experience to Wayland, not a chance in hell.
I’ll keep trying Wayland at regular intervals as I have been, probably on older Dell or HP workstations with 4Gb RAM and 1 or 2GB discrete graphics cards. You know the stuff that is ubiquitous in commercial operations. I bet if I try Wayland on a HPC with 64 cores, 128Gb RAM and 32Gb discrete graphics it’ll work sweet, probably because that’s what it’s developed on, but for my real world users it just doesn’t and I think that is sad.