I’ve been using Wayland for 5 years. There were a few bugs in the beggining, but now it works great. These threads are such a waste of time.
I have over 100 confirms X11 developments
That’s great dude. Why don’t you go maintain it then, apparently nobody else wants to: https://www.phoronix.com/news/RHEL10-Removing-X.Org
Wayland took too long
Look up how long btrfs has been in development, or at audio subsystem churn. These things take time, because it’s mostly volunteers working on them.
Systemic complexity has doubled in the last two years
What does this even mean?
Mir was better
It turns out the Canonical dumping random stuff over the wall is not the same as creating a legitimate open source community around a project.
Unfixable amount of race conditions
As if there’s never been a synchronization bug in X… But also System76 and others are writing Wayland compositors on Rust anyway.
My main thought is “this is a screenshot of a wall of text that’s hard to read”.
You should switch to Wayland
As an enduser my only noticeable issue with Wayland is that Auto-Type with KeepassXC doesn’t work.
Posting pictures of text is annoying. The letters are too tiny, i can’t read it comfortably. That is my first thought.
I remember reading through that thread when it came out and those are extremely worrying points. Wayland has extremely deep core issues. #2 there alone is horrible.
There are and were alarm bells ringing all around btw with Wayland. From a software developing perspective the approach is terrible. You cannot solve super complex problems by throwing away 30 years worth of code and redoing everything from scratch. You’ll just run into the exact same issues again. Which no, haven’t gone away as the technology advanced as many people would like to believe, we’re still using displays and networking and keyboards and mice.
There is a lot of legacy in X but there’s also a lot of accumulated experience and battle-hardened code. The obvious path would have been to keep the good and remove the bad.
Wayland will eventually since those issues but it will take just as long as it took X, because that’s what happens when you start everything from scratch again.
This is filling me with deja vu because it’s exactly what some of us went through with X, trying to piece together a working desktop out of dozens of pieces. But when you point that out you get “ha ha grandpa that’s old stuff, this new stuff won’t have that problem because [insert magic here]!”
Keep in mind that when Wayland started it was supposed to be a mini-server, to be used for the login screen only. Then the idea came to make it usable for stable, controlled and simple devices where there isn’t a lot of user configuration or hardware variation.
How it got from there to “let’s use it for everything on the Linux desktop and ditch X” I’ll never understand.
Which no, haven’t gone away as the technology advanced as many people would like to believe, we’re still using displays and networking and keyboards and mice.
Which X.Org was not designed to support.
Do you mean not initially designed to support? Because at least for displays and networking (in the sense of being able to send X events over the network) that seems wrong, a network capable display server is basically X’s entire purpose? And for keyboards and mice there are extensions now, so x.org as a standard now very much supports those by design. Actually to my knowledge Wayland basically just forked their keyboard standard, the X Keyboard Extension.
Most of the post is an “argument from authority”: Trust me, I have a PhD and maintain my own X server, and I assure you that Wayland is a pile of shit!
OP claims that “actually nothing will actually run” because the stable Wayland protocols lack so much important functionality. In reality, many people use Wayland every day, and multiple large distributions use it as the default display server. This doesn’t inspire confidence in OP’s knowledge.
Admittedly, the first bug they linked is a real issue and it should be fixed, but it’s not a Wayland design flaw. It’s an (arguably important) feature that hasn’t been implemented by all compositors yet. With the second bug OP laments that Wayland compositors are implemented in C, an unsafe language. This is true about X.org too, so I don’t really see the point. Arguably Wayland improves on X11 here, because someone could develop a new Wayland compositor in Rust, while in X11 this is a core part of the display server.
With the second bug OP laments that Wayland compositors are implemented in C, an unsafe language.
That’s not what they’re saying. They’re saying wlroots is full of race conditions, which will be very hard to fix because they’re part of a fundamental design problem.
That is a serious problem, but advocating X11 will not solve anything. Wayland is being improved every day, while X.org is in deep maintenance mode.
And let’s not pretend that X.org is perfect. Race conditions at least can be fixed, even if it takes a lot of time and effort. Worst case, someone will rewrite wlroots in Rust. But in X11 any application can kill other applications, install a key logger, pin itself to the foreground, etcetera. This is by design: it’s what makes window managers,
xkill
andxeyes
work. It’s also a huge security flaw that can never be fixed.That security argument is like advocating wearing a motorcycle helmet when walking down the street. It sounds like a great idea and super safe, but it’s also super impractical and the things it’s supposed to protect against are extremely unlikely.
But ok, more security isn’t a bad thing. But why not make it an option, like SELinux for example? That way users can choose a degree on a scale between security and convenience that suits their use case and circumstances. Why make it all or nothing?
It’s a valid concern IMO. Any application on X11 can install a key logger, record your screen, and influence other applications in a myriad of ways. With open source software from a trusted repository, this is not an issue, but an increasing number of people run random binary blobs from Steam, the Snap Store and Flathub. I am 100% certain that some less-conscientious publishers are already using X11 features to build ad profiles of their users; it’s a matter of time before the first ransomware will appear. The only sensible way to prevent this, is to confine applications to their own space.
But ok, more security isn’t a bad thing. But why not make it an option, like SELinux for example? That way users can choose a degree on a scale between security and convenience that suits their use case and circumstances. Why make it all or nothing?
Wayland simply doesn’t have protocols for most of this stuff. (Applications are supposed to use D-Bus and portals.) Developing new protocols that offer X11-like functionality is a large investment and will also need changes in the toolkits and apps to make it work.
Race conditions in wlroots is fixable, and not a reason to choose x11 over Wayland.
Out of curiosity, how much mail do you tend to store at your provider?
And if it’s a non trivial amount, what do you plan to do when they announce it?
It does give anti-SystemD “why make new when what we got now is good vibes”.
Their Java bashing was more a criticism of design patterns than Java, but fell into the meme bashing of tech based on one example. Find an old bug and say tech is dreadful as a result.
Sounds like a heap of crap. X.Org developers moved to Wayland, they were the ones who made it happen. Now, I wonder where this dude with his XOrg Forks and PhD and shit was during all that 15 years it took to conceptualize wayland.
You all need a lesson in taking everything people say, including and most importantly their qualifications with a huge grain of salt.
Wayland has been working perfectly for years now. Many of the supposedly “impossible to implement” functions of the old hunk of junk Xorg were either found to be bogus anyways or have been made available on Wayland.
Sincerely– Someone who’s been using wayland since 2016
Wayland has been working perfectly for years now.
Not for every one. For example, I still get random black screens with only mouse trails, windows disappearing, and videos not playing properly. Why yes, I do have an Nvidia card, thank you for asking.
After finally realizing nobody is interested in EGLstreams, Nvidia seems to be on track to make their drivers less of a disaster for Wayland support, so thankfully it is bound to become better
I just want you to know, this isn’t a failure on anything other than Nvidia trying to force their own crap on everyone and failing
I am of two minds:
- He’s not wrong
- It doesn’t matter at this point
It’s a mess, but honestly so are a lot of critical FOSS projects (e.g.: OpenSSH, GNUPG, sudo). Curmudgeons gonna curmudgeon. There was a point of no return and that was years ago – now that Wayland’s finally becoming useable despite itself it’s probably time to come to terms with the fact that better alternatives would have arisen had anyone thought they could truly manage it.
it’s probably time to come to terms with the fact that better alternatives would have arisen had anyone thought they could truly manage it.
This is the most important takeaway. There’s a lot of people whining about Wayland, but Wayland devs are currently the only people actually willing to put in the work. Nobody wants to work on X and nobody wants to make an alternative to Wayland, so why do we keep wasting time on this topic?
Some old people just do not want to learn new things and will try to justify it however they can. Happens with Rust, happens with Wayland.
This is not a case of old people. Wayland genuinely lacks features. People of all ages find it trash.
It is designed to lack features. It has a purposefully limited scope. Bashing it for it’s goal is weird.
I believe this is called cope
Copy paste?
I took wayland a decade to become usable. It tells me all I need to know about it simplicity and usefulness.
How do you define “usable”, and how long did it take X to get to that same point?
2 years from the original release to multiple ports, commercial applications, licencing to external groups and people actually asking to use it.
Isn’t Linux about choice? If you don’t like Wayland/SystemD etc. then you can just not use it lol
it is but if 20 years later there are no apps that support xorg… well, you wont have the choice of running xorg
This has Systemd vs Runit vibes. No matter how many anti-systemd folks scream to me about how horrible it is for XYZ technical reasons, every Linux distro I’ve ever used for years, desktop and server, has used systemd and I’ve never experienced single problem that those users claim I will.
Same here with Wayland. All the major desktop environments and distros have or are implementing Wayland support and are phasing out X. The only reason I’m not on Wayland on my main computer already is because of a few minor bugs that should be ironed out in the next 6-12 months with the newest release of plasma.
It’s not because Wayland is unusable. I try switching to Wayland about every 6-9 months, and every time there have been fewer bugs and the bugs that exist are less and less intrusive.
Any time you get hardcore enthusiasts and technical people together in large community, this will happen. The mechanical keyboard community is the same way, people arguing about what specific formula of dielectric grease is optimal to lube your switches with and what specific method of applying it is best.
At a certain point, it becomes fundamentalism, like comic book enthusiasts arguing about timeline forks between series or theology majors fighting about some minutia in a 4th century manuscript fragment. Neither person is going to change their views, they are just practicing their arguments back and forth in ever-narrowing scopes of pros and cons, technical jargon, and the like.
Meanwhile the vast majority of users couldn’t care less, and just want to play games, browse the web, and chat with friends, all of which is completely functional in Wayland and has been for a while.
All of the technically-minded posts I’ve read about systemd have been positive. The only detractors seem to be the ones with less technical knowledge, complaining about “the Unix philosophy” and parroting half-understood ideas, or worse, claiming that it’s bad because they have to learn it.
I know xorg has problems, but it was good to get some insight into why Wayland is falling short. Every argument I’ve seen in favor of Wayland has been “xorg bad”.
X code is convoluted, so much so that the maintainers didn’t want to continue. AFAIK, no commercial entity has put any significant money behind Xorg and friends. Potentially unmaintained code with known bugs, unknown CVEs and demands for permission system for privacy made continuing with Xorg a near impossibility.
If you don’t want new features and don’t care about CVEs that will be discovered in future as well as the bugs (present and future), then you can continue using Xorg, and ignore all this. If not, then you need to find an alternative, which doesn’t need to be Wayland
Oh, and you might need to manage Xorg while other people and software including your distro move onto something else.
So yeah, “xorg bad” is literally the short summary for creating Mir and Wayland
I get it.
But as far as I can tell, there are just two xorgs now, one of them is just spelled “Wayland”.
To be honest, systemd does treat its own services specially when I don’t think there’s a good reason for it. Systemd as a whole is a huge improvement over the old bash script mess, but there is something to the “Unix philosophy” take.
The solution isn’t “switch back to writing hundreds of sh scripts”, though, it’s “improve systemd”.
As for the X11 vs Wayland topic: there’s still tooling missing for some use cases. X11 forwarding over SSH doesn’t work as well even with Waypipe. Just yesterday I encountered a bug that caused any attempt to run Waypipe to freeze the entire display.
Again, the solution is to fix Way land’s tooling, not to abandon it for X11 because change is scary.
This has Systemd vs Runit vibes. No matter how many anti-systemd folks scream to me about how horrible it is for XYZ technical reasons, every Linux distro I’ve ever used for years, desktop and server, has used systemd
You’ll one day learn the difference between Popular and Correct.
Trump is popular, for instance.
and I’ve never experienced single problem that those users claim I will.
This is a “everyone tells me to get smoke detectors and I’ve never had a fire in all my 23 years of life” comment.
There’s a reason we have building codes, seat belts, traffic lights, emergency brakes, FDIC, and pilots’ licenses.
(Probable) Source for those who prefer actual text: https://lobste.rs/s/acepul/wayland_is_pretty_good_actually#c_bzr8nr
Wayland has mostly positive user reviews because it presents nicely to the user (VRR, scaling, etc.) On the developer front it seems there’s a lot of struggle over things that were solved in X11 but for some reason require a lot of debate in Wayland.
- There’s still no way to universally configure monitors and input devices, so the startup cost to checking out a new “WM” (compositor in Wayland terms) is non-zero - you have to reconfigure everything from the ground up, and for anyone with complex input systems (see: accessibility devices) this will take a lot of time because each compositor insists on using a different format for configuring these things.
- Each compositor is tasked with coming up with solutions for all parts of the user experience (hence the last point) and thus anyone who wants to experiment with making their own WM now has to worry about a billion things that wouldn’t have had to deal with in X11. Yeah, there’s libraries for dealing with that stuff, but it’s not as simple as it was and lot of innovative WMs won’t ever be able to make the jump.
These are the two biggest issues I can see that are entirely chalked up to its design. Technical issues (like the “load balancer” thing that keeps Firefox from crashing on Wayland) will be solved in time. However, the above points are unlikely to ever be addressed. Should they be? I don’t know.
X11 world is just really as bad. It’s just so that Wayland is new and X.Org hacks are so normalized it doesn’t register for people how much work you need to put into X to make it work.
Each compositor is tasked with coming up with solutions for all parts of the user experience (hence the last point) and thus anyone who wants to experiment with making their own WM now has to worry about a billion things that wouldn’t have had to deal with in X11
Well, you would be disappointed to know that it’s somewhat the same fuckery under XOrg as every usable compositor and window manager needs to carry fixes, hacks, and hardcoded workaround for edge cases in X or features that are unsupported today. If you use a Wayland library like you use X libraries it’s not that much harder to make a simple Wayland compositor.
This is one of the most cited “wayland issues” and it’s IMO completely and utterly wrong. Go look at how convoluted existing XOrg codebases (mutter, metacity, compton, compiz, etc.) are. It’s only “simple” if you wish to draw a rectangle upon the screen ( and even then, wlroots exist)