Skip Navigation

You're viewing a single thread.

213 comments
  • Wayland is still too immature. I couldn't get it to work on my Kubuntu distro.

    And then there's this list of problems with Wayland.

    https://gist.github.com/probonopd/9feb7c20257af5dd915e3a9f2d1f2277

    BEGIN RANT

    "Move fast and break things" may be fine for software gurus who love to experiment and have no problem hitting their head against the wall every few days while believing in the promise of a free-to-fix future, but this isn't true for poor or busy people who are NOT middle class folks living in their own house in a suburb with a garage full of computer parts. There are single parents, caregivers with disabled and/or elderly, folks who need a reliable computer for their studies, and in general people who simply need something that JUST WORKS.

    I'm a caregiver, and unfortunate I'm poor enough that I don't have money to buy a commercial OS. Heck, I wish Windows just worked instead of making old versions obsolete. I was perfectly fine with Windows 7 ten years ago until Microsoft started doing planned obsolescence bullshit with their forced updates. I had to switch to Linux because Windows became very unreliable and I needed a stable platform that wouldn't ruin my work.

    (So if you're one of the persons who reply to "Help my Linux is having problems" with "well you should know Linux is like that, you should have thought it twice before switching", then you're part of the problem because that's a very, very shitty answer to give to a non technical end user with limited time and resources)

    The year of the Linux desktop will never arrive if developers keep pushing incomplete and buggy software to the end users instead of actually fixing bugs and delivering their stuff ONLY when they're ready.

    Wayland is NOT ready for the end user.

    END RANT.

    • if developers keep pushing incomplete and buggy software to the end users instead of actually fixing bugs

      My understanding is, the issue is that fixing bugs in X has become too much of an issue due to bloat and bad historical architecture, so the developers working on it - and providing the software for free, if not working for free - instead worked together to develop a new standard aiming to fix the issues inherent to X's code and design.

      The "list of problems" is absolute bullshit right from the start. The first two sections are "It didn't used to work like this in X, Wayland is trash!" and "I had some screen recording software using X APIs and they don't work when not running in X!". In fact, a lot of them follow this pattern, blaming Wayland because it doesn't have 100% backwards compatibility. It's not an X rewrite, it's meant to be a new, better piece of software.

      I will not deny that Wayland has problems, of course - but those mostly come down to NVidia refusing to support open protocols, missing features that are yet to be implemented, and missing software support for Wayland.

      I will also say that on Arch, which doesn't assume I'm using X, Wayland does work completely fine for me when following instructions. It might be an issue with the distro you're using not having good support, or one of those edge cases like problematic hardware. I definitely agree that you should stick with X for now if you have problems, but I'll also say that you're getting it for free, and if you don't report problems, they might also not know about them, for example because it only occurs on specific hardware.

    • Move fast and break things

      Bro, wayland is 15 years old

      CC BY-NC-SA 4.0

    • There are single parents, caregivers with disabled and/or elderly, folks who need a reliable computer for their studies, and in general people who simply need something that JUST WORKS.

      This is also one of the many reasons why Linux as an OS fails to establish a bigger user base. Of course, this is one of the smallest problems, but it still is.

      Like Linus said, every tool every dev made was usually because they wanted to fix problems in their workflow, not because users needed something that they can provide. Sure, I'd also look at things from this perspective. After all, god made his beard first, not everyone else's, but the trouble is, things don't really move fast enough in that direction. Don't get me wrong, there has been progress in GUI tools, but not enough IMO. Most tools are terminal based, and while that is not a problem for most UNIX type OS enjoyers, that is a problem for your averige Joe. That might not be the crowd we're trying to get off of MS and Apple products, but they still play an important role IMO, more of a guide as how a UI should look and feel to the average user. Linux and other UNIX based OSes kinda messed up with this one. Things are getting better though, have to say. I don't use some of the UIs for stuff people usually do, but I have tried a few and I have to say that things are moving in the right direction the past few years. Just not fast enough...

    • You're not wrong but also misinterpreting this. Yes, it's bad to push incomplete software on end users, and it's even apart of the entire development ideals of Linux: never break userspace. There's even small bits of code (see: egrep and fgrep) in the core commands that has been on the chopping block for removal for 2 decades but hasn't because removing them would break apps.

      The choice of PUSHING Wayland on end users is not up to the developers making wayland, it's up to the distro maintainers, and this image honestly doesn't even make sense. Most distros right now are either so nothing, and the ones that do are disabling Wayland until it's more feature complete. The only big distro I remember that's specifically is pushing for it is Fedora, and Fedora is specifically known for pushing for new initiatives.

      X11 works just fine, and will work just fine for a long time, and if there's ever a point where a majority of apps start dropping X11 support for Wayland, it's going to be because Wayland just works by that point and has for long enough for devs to care.

      That article itself against has been a pain point for years because it over-dramaticises a lot of the pain points about Wayland and a lot of the issues it touts don't... exist anymore. I've used lots of software like OBS on Wayland just fine a long time ago even though the article says it's been broken for years. Nvidia on Wayland has also just gotten to a good state on proprietary drivers while the article implies you need the crappy open source drivers to use Wayland at all, which hasn't been true for a very long time. I could go on about this article, but Brodie Robertson has already talked it to death on YouTube.

      Wayland does "just work" (no bugs, no configuration, just switch to it and nothing breaks) for a lot of users at this point, and I'm tired of this article ignoring that and trying to make it seem like Wayland is this buggy slop everyone's being forcefed when it's not.

213 comments