Thanks, Ponce. Awesome community! I'm a Wayland guy, but X11 runs everything since the Stone Ages. A true, dependable workhorse.
On an up-to-date Gnome or KDE Desktop, Wayland is better because of way more Progress. So, there should be the Experience better. Depending on what Gnome Version will be used in Zorin 18, it could be a better technical Experience then.
I don't know how experienced you are with Linux systems, but I've used Wayland with multiple distros and it's incredibly silky smooth especially with ArchLinux for gaming.
As a compositor it's the future gold standard and successor of X11. Has smoother animations and tear free visuals, crispy scaling for high DPI screens, better security, reduced legacy overhead, leaner, meaner, etc..
We just use X11 because of legacy apps and their compatibility layers and lazy developers who don't properly develop drivers on time like NVIDIA for Linux systems.
It's kind of how Systemd replaced SysV init in many Linux distributions for faster, more modular system initialization. As the saying goes, out with the old, in with the new.
Yes, because Arch (and others) uses (more) up-to-date Desktop Environments. So, it has the (more) up-to-date Wayland Implementation. But on LTS Distro's it depends what Desktop Environments in which Version are used.
I personally use Wayland here on Zorin 17, too and don't have Issues - but I use an AMD APU. But we have here more than enough related Issues to that - especially with Nvidia Graphics. When You have the right DE Version and the right Driver Version, You can have a good Wayland Exeprience with Wayland. But this is here on Zorin at the Moment not yet the Case.
More modular? It is less modular. Systemd is a Deamon - everything is integrated. That is the opposite of modular.
What SystemD does, however, is break down the init processes into chunks that can run in parallel creating a bit faster boot time.
There is high demand for X11 because advanced or niche workflows still need X11-specific capabilities such as screen sharing, window managers with fine-grained control, or color calibration tools. Wayland lacks the ability to manage these since it shifts complexity to the compositor and toolkits. Since Wayland is founded on this, it cannot be "fixed". The only workaround is to continue using the X Server: XWayland.
So, no - we do not use X11 just for "legacy apps."
Wow. Well.
Nvidia wanted to focus on EGLStreams and Wayland wanted focus on GBM. Each side can accuse the other of not doing what the other wanted them to do.
Nvidia did make the change to GBM, however. It required a large rewrite of the driver architecture in order to accommodate this.
That's not laziness.
@Aravisian I couldn't have said it better myself. Even though my computer is still somewhat new, (5-years old) it still requires X11 on Zorin OS, cause were not using the latest and greatest of bases and DE's.
Wayland breaks most peoples experiences, who use Zorin OS. This is why we recommend most people to use X11 instead. Its not rocket science, its simply a reality.
And yep, not all dev's are lazy, some have to work within constraints of what they do. This is why Zorin OS will continue to support X11, in at least the next distro, Zorin OS 18.
Going fast and break things is not the way people! NASA doesn't operate that way, neither do I. Slow and steady wins the race, which is why NASA isn't even considering a human based Mars mission, till like 2035.
Thanks for the clarity, Aravisian. You're correct. I misspoke about systemd. Not more modular, but superior to SysV for faster parallel boots, robust dependcy management, reliable services, and unified logging. It's less modular but more streamlined. I don't want to go back to clunky, sequential boots.
Some folks are stuck in the dark ages. X11 is a creaky, insecure, overcomplicated mess. Some tweaks in Wayland can be tricky, but XWayland bridges the gap. Nvidia's EGLStreams push was a selfish power grab to maintain their proprietary grip on Linux graphics, delaying progress. Still, Wayland's the future for most use cases. X11's days are numbered.
Framing your argument this way is unproductive. Historically, the dark ages were not dark.
And making a personal attack against disagreement by accusing those that disagree with your claims as "stuck in Darkness" implies that they could see the light, but refuse to do so - assuming motives and intent, downplaying any valid and rational arguments that they make.
XWayland creates an Xserver to handle applications that Wayand cannot. In essence, this is running Xorg as a backend on Wayland.
How is this not a complicated mess?
It undoes the promise of Wayland : to simplify and reduce. It does this because Wayland is unready for release and use. And importantly, it is misleading.
It suggests that Wayland is doing the job when in reality, it was unable to do the job and needed X-server - which it claims to have replaced.
As you can see from this One Point: There is nothing so simple as "X11 is in the Dark Ages" here. X11 is necessarily relevant due to Waylands inability to handle display; only to smooth and polish this fact, we rename it "XWayland."
Moved to new topic so as to not disrupt a Tutorial Thread.
@hex1337 You clearly have not used PCLOS Debian KDE Plasma, no systemd, no Pulse Audio, and very slick boot time with X11 and Liquorix kernel. I for one would be interested in XLibre, even if it is a one man band he was lead dev on maintaining Xorg/X11.
Red Hat & FreeDesktop Go Into Mass Censorship Mode over Xorg Fork, Bans XLibre Dev, it really seems these halfwits want to bury X11/Org
Not surprising since GNOME is dropping X11.
It is also another example of Gnome seeking Microsoft like dominance and centralization on GnuLinux.
We left that behind with Microsoft only to repeat the futility with Gnome.
I know it creates work, but I really, REALLY like it when distributions have several DEs to choose from. Garuda is not for me, but props to them for offering Plasma, GNOME, Cinnamon, XFCE, Sway, i3, Hyprland, and Cosmic. (i3 and Hyprland are not, as I understand it, DEs, but for the purpose of alternatives to GNOME, fill the same role.)
That is why I prefer PCLOS Debian - comes in budgie, Cinnamon, KDE (Plasma), Plasma Mini, MATE (Think Gnome 2), and xfce. No systemd, no pulse audio (Pipewire takes its place) and Synaptic Package Manager - Discover is present in the Plasma release but I prefer Synaptic any day, possibly Apper as an alternative to Discover.
Dark age was just a playful jab, sorry for the shade. What I meant was X11's like that beat up station wagon your uncle still drives. It's got duct tape windows, a wheezing engine, and a cassette player stuck on repeat. It guzzles gas and barely rolls. It technically gets you where you need to be. Wayland's the sleek electric car: smooth, secure, and built for now, though it needs XWayland's rusty trailer hitch to tow those old X11 apps along. It's messy, sure, but it's not Wayland pretending to be X11. It's just keeping the old jalopy running until everyone upgrades to the future.
X11 is a hot mess under the hood. Apps can spy on each other, malicious ones can hijack your screen, and its outdated setup has zero guardrails, leaving your system wide open. There is literally no client isolation. It's monolithic all in one architecture mixes display, input and window management, making maintenance a nightmare and unsafe. There is also no love for modern GPU features like hardware accelerated compositing and graphics are choppy and games are laggy.
Love that. Sounds like a scrappy garage band of Linux distros. Maybe punk rock cousin to Mandriva Linux. Keep rocking PCLOS.
So Xorg got forked into X Libre, but Red Hat and Free Desktop banned him and torched his repo? Harsh move, maybe another win for Wayland. They're like grumpy tow truck drivers who would rather scrap the whole thing than let anyone tinker. Reminds me of a video with Linus Torvalds roasting Debian for being a pain to install, praising distros that just work so he can get on with his life. I guess Linux is shifting gears from a tinkerer's paradise for scientists and wizards to a smooth, point and click ride for regular folks who only want a Windows like experience.
No, PCLOS is still PCLOS, with TexStar using .rpm and based on Mandriva 9.2 base updated. PCLOS Debian is a separate beast altogether, with Upgreyed pumping quite a bit into this distribution, with the different DE's I quoted. PCLOS original flavour comes with KDE Plasma, MATE (think Gnome 2) and xfce. The community releases of PCLOS produce the forerunner to Plasma, Trinity DE, LXDE, LXQt, and OpenBox. And for clarity, PCLOS Debian has no connection to Debian, it just uses Debian as it's base. It is clear that Red Hat and their ilk are all part of the 'Anti Fix-It Yourself' brigade. No one has yet raised the issue of Accessibility - one of the great pioneers of Accessible software such as Orca screen reader and magnification, plus a plethora of accessibility tools always overlooked that came with Compiz.
The levels of exaggeration are astounding.
Wayland is younger than X11, but not by a whole lot. It is not new, being over seventeen years old. That's your Upgrade Into the Future? Your post reads like propaganda.
To say X11 is "barely rolling" is absurd. It's maintained, mature, widely deployed, and highly configurable. Modern desktop environments run very well on X11.
Wayland is still incomplete in practice for many use cases. It’s not merely “smooth and secure”, many features (like remote desktop, screen recording, and input methods) are still fragmented across implementations and are often harder to configure than in X11.
X11 does lack client isolation but users aren't running untrusted graphical applications indiscriminately.
That said, your claim that your "system is wide open with zero guardrails" is completely false.
Take a look for your .Xauthority
file. That is one of many guardrails.
Others include AppArmor, xhost
and the many patches released for X11 by Xorg for passive grab limitations that your post pretends simply do not exist.
Presenting your case is fine as long as it is factual. Presenting falsehoods is unacceptable.
And yes, Wayland sandboxes the client, isolating applications - but...
Any code used in the compositor could allow a malicious client to exploit weaknesses and break isolation. These compositors are complex codebases with graphical stacks, protocol extensions, input management, and GPU drivers.
You can argue it puts a band aid on your "security concerns" for X11 - yet it does not solve the problem.
Because of Waylands isolation, now applications must rely on out of band API's (e.g. xdg-desktop-portal) or privileged API's, which adds complexity and a New Series of Clients that Need Access to Display.
It gets worse. I address this below but X11 is Modular (not monolithic) and because of this, the client, server and windowing are isolated from each other.
Not so with Wayland. Which makes Waylands security have a single point of failure: The Compositor. Hijack that - and you get access to everything...
This is a mischaracterization.
X11 can and does support hardware acceleration and compositing hrough extensions like GLX and using compositors like Compton or Mutter.
Performance issues are not inherent to X11, but arise from how it’s configured and extended. Many high-performance games run under X11 with low latency and good frame rates, especially with modern drivers.
Wayland aims to make GPU support more direct and cleaner, but this varies dramatically depending on compositor (e.g., Mutter, KWin, Sway, Wayfire) and graphics stack support.
So, while Wayland has a cleaner design, X11 is not inherently laggy, and many users still find it superior for gaming depending on their hardware and setup. Indeed, it is the most often recommended switch in many cases right here on the Zorin OS Forum.
Sigh...
Modular. It is Modular. Not Monolithic.
The X server, window managers, and compositors are all separate.
The word you mean is Complex, not Monolithic.
Wayland is simpler, by design. Which... is also its flaw.
Not all issues are resolved merely by protocol simplification and Waylands ecosystem is fragmented and incomplete without any clear path on how to employ Waylands simplicity to solve that problem.
Which is why XWayland is being used. Not because of Lazy Developers loving Jalopy's.
But because in the simplicity of Wayland, it was built from the ground up to lack the capability and now that it has been shoved onto the base, the maintainers have no idea how to fix that without re-writing Wayland.
And as I point out above - this also channels all of Waylands security through the bottleneck of One Single Failure point.
Given a choice: I'll take X11's security over Waylands.
It's got the buffers Wayland lacks.
Aravisian, you're knowledgeable I give you that. But Wayland's the real deal. X was the standard but Wayland is the upgrade. It's time to ditch that old clunker! There's a reason all the major distros are flooring it to Wayland for it's beefy security, juicy performance, active development and an actual unified protocol that's like a GPS keeping all the extensions from veering into Bugsville.
.Xauthority is a leaky bucket. Apps can yoink inputs or scrape screen (ex. XGetImage). AppArmor is external and not native to X11. It's not X11's ride or die. It's more like a rent-a-cop bouncer taped to the door, half asleep and checking IDs with a flashlight that's out of batteries. Also passive grabs mitigate but don't eliminate risks like keylogging. It's like slapping a sticker on a keylogging landmine. CVE data aligns: X11 has more vulnerabilities (12 critical CVEs since 2020 vs Wayland's 3, per NIST)
Wayland's protocol is a badass bouncer that enforces client isolation with compositor-mediated input, preventing direct client-to-client attacks. The compositor, while critical, operates in a constrained VIP environment. Wlroots patches in 2025 show active development, outpacing Xorg's moldy updates.
X11's modularity... HA! It's a chaotic flea market, window managers like Mutter and Compton are all over the place, causing complexity, glitches and inconsistent behavior. Wayland's slick unified protocol ensures consistency, with extensions like xdg-decoration simplifying implementation. Tools like xdg-desktop-portal standardize screen recording and input across compositors (ex. Sway, KWin), countering your fragmentation claim.
XWayland is just a compatibility bridge, not a crutch, allowing X11 apps to run on Wayland while its ecosystem matures.
Wayland often outperforms X11, especially on modern hardware. Phoronix benchmarks (2024) show +20% FPS gains in games like Cyberpunk 2077 on Wayland. I guess X11's GLX enables hardware acceleration but is less efficient than Wayland's direct GPU integration.
Disagreement is a good thing.
Evaluating disagreement allows people to critically examine the merit of ideas. To look at details through a lens and achieve greater accuracy.
When it comes to a complex subject, such as this one, then the best presentable case is one in which openness and honesty take precedent.
As an example:
Many Wayland proponents cite X11 as insecure, but withhold, do not divulge, neglect to mention Xorgs patches that were immediately applied to the examples Wayland proponents refer to.
Whereas... when I discuss it - as seen above, I openly and honesty discuss the flaws of each. X11 and Wayland alike. When Wayland does something well, I say so and I applaud it.
Using independently verifiable information - that any reader can research and verify, is critical to this kind of open and honest dialog.
However...
The army of analogies that you deploy onto your battlefield are unverifiable. I cannot reasonably check the code for sleepy guards with dead flashlights.
I cannot look into Wayland and verify whether it has a Beefy bouncer sporting the fastest E.V. with a golden suitcase and a pet unicorn.
I have explored the X11 system, I have built window structures for it but I never once, found a flea market. If I had, I probably would have spent some hours there, looking for good classic vintage items at lowball prices.
I cannot point to how a protocol functions in order to refute a claims of fleas and bouncers.
And again, we really need honesty.
Wayland and X11 both are Active.
Both are.
X11 is actively maintained - including security patches - as of this very day. It is foundational, matured and established.
Wayland... is incomplete. So, you can say it is in active development; because it really needs to be.
Not just necessitating XWayland for x11 applications, either.
It is simply Untrue that Wayland is ready and that XWayland only is needed for applications not yet "caught up to Wayland."
Extensions:
Wayland is built to be lean and minimal. Simple. But, that simplicity creates complexity.
Wayland then must relay on Protocol extensions in order to handle Clipboard and drag-and-drop, Window management hints (e.g., tiling, maximize, decorations), Input methods (on-screen keyboards, complex input languages),Screen capture and sharing, Multi-monitor and fractional scaling, Idle detection and inhibition (e.g., for video playback), Gamma control, color calibration, and HDR support.
None of this is "bullheaded" application issues. These are all compositor and desktop - Unready and non standardized, fragmented and unready for proper Wayland implementation.
Even on this Zorin OS Forum, we often discuss how Wayland implementation in Gnome... that is just one D.E. is improved in other cutting edge distros and still not yet working or ready.
Because compositors must opt in to each extension, even broadly useful protocols may take years to propagate across the ecosystem. Developers must maintain fallbacks, and features often feel experimental or unstable for long periods. Again, on the Desktop. Not the applications.
API's:
And where is our progress? Even on the cutting edge, Global Hotkeys: Not fully supported. Color management: Limited or unavailable. Fractional scaling: Remains incomplete and varies based on desktop environment. Screen Sharing and remote desktop: a variable that is prone to crashing, errors if it works.
Not to mention the many configurations that have no Wayland equivalent because none have even been developed, yet. Think xrandr
.
Yeah, despite having Incomplete or compositor-specific tooling, incomplete or fragmented protocol extensions, it can be used by the casual user at this time.
For system administrators, power users, accessibility developers, remote workers, or niche software stack developers; Wayland is unready, incomplete and broken. Disregarding any X11 applications using XWayland.
Can you please form your arguments as verifiable statements that describe what Wayaland and X11 do, instead of blood drinking insects, bouncers, cars, steam engines, Circus animals and the like?
I should probably just stay out of this as I'm not well versed in the topic, however it's my understanding that a significant portion of Gamescope's use cases, if not its reason for its existence, is because Valve needed a compositor that provided features Wayland didn't yet--and even then, it uses XWayland for games. Apologies if any of that is incorrect; I've done my due diligence, but authoritative sources (as opposed to Reddit threads) are hard to find.