• Digit@lemmy.wtfEnglish
    3·
    17 hours ago

    When I was new here, it was XFree86 vs Xorg. (Showing my age).

    There will be new technologies yet. And not all of them will successfully supplant/obsolete the older. Some of the older may even come back, despite (/ because of) the efforts made to eliminate them.

    what the issues with Wayland are.

    LMLLMTFY…

    Here’s a concise summary of the main issues people have with Wayland:

    • Compatibility: Many legacy X11 applications and workflows (e.g., screen recording, remote desktop, and some games) don’t work seamlessly under Wayland, requiring patches or workarounds.
    • Screen Sharing/Recording: Wayland’s security model restricts direct screen access, making tools like OBS, Discord, or Zoom less reliable for screen sharing or recording.
    • Multi-GPU/NVIDIA Support: Historically, NVIDIA’s proprietary drivers had poor Wayland support, though this has improved recently. Multi-GPU setups can still be problematic.
    • Customization: Wayland compositors (like Weston, Sway, or GNOME’s Mutter) are less flexible than X11 window managers, limiting advanced user customization.
    • Remote Desktop: X11’s network transparency (e.g., ssh -X) is missing in Wayland, requiring alternatives like RDP or VNC.
    • Input Handling: Some users report issues with input latency, touchpad gestures, or keyboard layouts, especially in niche setups.
    • Maturity: While Wayland is stable for most users, edge cases (e.g., hybrid graphics, VR, or professional audio/video tools) may still face bugs or missing features.
    • Why it matters: Wayland aims to modernize Linux graphics with better security and performance, but the transition has been slow due to these pain points. Most major distros now default to Wayland, but power users and professionals often still rely on X11 (or XWayland for compatibility).

    *shrug*. I guess that’ll do.

    I’m >9000% more enthused for XLibre, and I utterly disregard the gaslighting associations some insist we believe about it, like community developed XLibre’s the fascist one, not Wayland spearheaded by a monopolistic corporation with nazi roots. XD

    Edit:

    PS, then I asked...

    wayland’s more bloaty, and complicated, right? what with its extra compositor layer atop the server layer before the window manager layer, or whatever it is, right? (I dont really understand it… the complicatedness is more than i’ve got my head around, since not all methods of using it seem to have the same ontos/topos).

    Your frustration is understandable—Wayland’s architecture can feel more complex than X11’s, but the “bloat” and “complication” are often misunderstood. Let’s clarify the layers and why it feels this way:

    X11 vs. Wayland: Layer Comparison X11 (Simpler, but Monolithic)

    • Single Layer: X11 combines the display server, window management, input handling, and compositing into one monolithic system.
    • Pros: Easy to understand (one thing does everything).
    • Cons: Security risks, poor performance, and inflexibility (e.g., compositing was bolted on later). Wayland (Modular, but Distributed) Wayland splits these responsibilities:
    1. Compositor (e.g., Weston, Mutter, KWin, Sway):
    • Replaces X11’s “server” role.
    • Handles rendering, input, and compositing directly.
    • Acts as the “display server” and the “window manager” in one.
    1. Clients (Applications):
    • Talk directly to the compositor via the Wayland protocol (no middleman like X11’s server).
    1. Protocols (e.g., wlroots, xdg-shell):
    • Define how clients/compositors communicate (e.g., for window management, screensharing).

    Key Difference:

    • In X11, the “window manager” is a separate program (e.g., i3, Openbox) that talks to the X server.
    • In Wayland, the compositor is the window manager (e.g., Sway = compositor + window manager).

    Why It Feels More Complicated

    1. No Single “Wayland Server”:
    • Unlike X11’s Xorg, there’s no one “Wayland server.” Each compositor (Mutter, KWin, Sway) implements Wayland differently.
    • This can lead to inconsistencies (e.g., GNOME’s Wayland vs. Sway’s Wayland).
    1. Protocols Everywhere:
    • Wayland uses protocols (like xdg-shell, viewporter, screencopy) for features X11 handled implicitly.
    • Example: Screen sharing requires explicit protocol support (e.g., pipewire), while X11 just “let apps grab the screen.”
    1. Less “Magic”:
    • X11 allowed hacks (e.g., xrandr, xinput) to work around limitations.
    • Wayland enforces stricter rules (e.g., no direct GPU access for clients), which feels restrictive.
    1. Transition Pain:
    • Tools like xrandr, xinput, or Xephyr don’t exist in Wayland. Replacements (e.g., wlr-randr, wayland-debug) are less standardized.

    Is It More Bloaty?

    • No, but it’s more modular:

      • Wayland itself is a protocol, not a program. The “bloat” comes from compositors adding features (e.g., GNOME’s Mutter vs. minimal Sway).
      • X11’s “simplicity” was deceptive—it was a single binary, but it did everything poorly. Wayland distributes responsibilities more cleanly.

    • Example:

      • X11: One process (Xorg) handles everything, often inefficiently.
      • Wayland: The compositor (e.g., weston) does rendering, while pipewire handles screensharing, and libinput handles input.

    TL;DR Wayland feels more complicated because:

    • It replaces X11’s “one big blob” with explicit, modular components.
    • The lack of a single “Wayland server” means behavior varies by compositor.
    • Protocols replace X11’s implicit behaviors, requiring more setup for edge cases. But: This modularity is why Wayland is faster, more secure, and more maintainable long-term. The complexity is a tradeoff for fixing X11’s fundamental flaws.

    ___

    I wonder how much of its training data was feeding on all the pro-wayland advertising spiel.