Announcement

Collapse
No announcement yet.

Louvre Is A New C++ Library Helping To Build Wayland Compositors

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #41
    Originally posted by user1 View Post

    That's not a good analogy. Minix was an OS for educational purposes, so not something for general usage. When it comes to Wayland libraries, I see wlroots as something that at least reduces the fragmentation of the Wayland ecosystem, since there is no reference display server implementation like in the case of X.org. While Gnome and KDE already have their own Wayland display server implementation, I think it's good if everything else will target wlroots in order to reduce fragmentation. And this is an attempt to only make it worse. Because even if they claim to have some performance benefits, it doesn't seem like their main goal of easing the development of Wayland compositors is any different to that of wlroots.

    And this is what bothers me about open source in general. On one hand, you have so many complains about developer burnout, lack of resources and lack of compensation, but on the other hand, some developers find resources to fork stuff, or reinvent the wheel only for marginal benefit.
    I think the bigger explanation behind this chaotic mess is that when Wayland was initially released it was just a protocol with a reference implementation (Weston). What they should have done is also released a wl-roots like library from the beginning to make it easier for compositors to adopt/try out/prototype Wayland, even if a hypothetical wl-roots like library wouldn't be usable by some compositors at least the developers would have some idea on what to expect.

    Instead what we got was a protocol, which for anyone that has any experience in software development knows what you can actually ascertain from a protocol (aside from some properties) is very little, and the Weston (the reference implementation) was next to useless for this specific usecase. This state of affairs meant that for the large part of a decade, Wayland got almost no real world use (no one really took it that seriously) and to the point of this thread, we got this annoying fragmentation where every compositor ended up developing their own integrations into Wayland.

    Even with wl-roots, the DE/compositors that actually use it are a drop in the ocean compared to the jugernauts (KDE/Gnome/Mir etc etc)

    Comment


    • #42
      Originally posted by jacob View Post

      Never say never. GNOME and KDE were both heavily invested in their ORBit and DCOP IPC systems and happily threw them out when DBUS came along. They had their own audio servers and moved to Pulse Audio (and now PipeWire). They had their own session management logic and readily switched to systemd (at least GNOME did, not sure about KDE). So if there was a Wayland display manager foundation that does what they need, I don't see why they would eventually use it if ultimately it allows them to reduce their development effort.
      especially KDE which is written in C++ could sync up well with this new compositor, while gnome might better sync with wlroots.

      Comment


      • #43
        Originally posted by jacob View Post

        Never say never. GNOME and KDE were both heavily invested in their ORBit and DCOP IPC systems and happily threw them out when DBUS came along. They had their own audio servers and moved to Pulse Audio (and now PipeWire). They had their own session management logic and readily switched to systemd (at least GNOME did, not sure about KDE). So if there was a Wayland display manager foundation that does what they need, I don't see why they would eventually use it if ultimately it allows them to reduce their development effort.
        The key difference here was that ESD, aRts, ORBit and DCOP were born out of pure necessity, didn't aim to provide any particularly sophisticated functionality and at least the KDE-umbrella projects had numerous technical issues. Giving them up for something common and better maintained was welcomed by everyone. Furthermore, neither IPC nor audio handling is specific for graphical desktop environments.

        On the other hand, window management is specific to DEs and it is one of the most important parts of use experience. I'd argue that the one major difference between GNOME and Plasma is how they manage windows and workspaces. Major DEs need to develop their own WMs because they are key to the user experience they provide. In the Wayland world, the WM is also the compositor, so swapping out all the guts for something common and implementing all of custom feature on top of it would be a massive pile of work with dubious benefits.

        Comment


        • #44
          I compiled that project, but when I start louvre-weston-clone nothing happens, no messages and no output. I thought that file was a reference implementation, what am I missing here?

          Comment


          • #45
            Originally posted by user1 View Post

            Just curious, what do you not like about them?
            Mutter is off the table right away due to having to deal with typical gnome bullshit. you can actually work with KDE devs so that's a lot more promising, but in the end, kwin is still a massive, old project and because of that it's still hard to hack on. ofc thats not to say you can't do it, you absolutely can, and the code isn't a bad quality, its still just a bloody lot to deal with.

            in the end, doing something from scratch the way you know how to, and in a way you are comfortable with, is much better then trying to shove shit code into a pre-existing project.

            Comment


            • #46
              I wholeheartedly support projects like Louvre. We need more innovation and more new ideas, not just sticking to a few "standard" implementations.

              Trying out new things is the only way to real progress. Poly-centric development is an important piece of that.

              The people who are whining about fragmentation are those who just prefer the status quo, who like orders, rules and authority, and who would be the happiest if they never had to learn anything new.

              Btw, I currently prefer Xorg, but in the future, I might change my mind.

              Comment


              • #47
                Originally posted by darkonix View Post

                I have several different overlapping and/or contradictory opinions about that ;-)
                LOL...

                Comment


                • #48
                  Originally posted by ehopperdietzel View Post
                  Hello, I am the developer of Louvre, and I would like to express my gratitude to Michael Larabel for sharing information about the project.
                  Good work! It seems a very nice project.

                  Comment


                  • #49
                    I love Wayland!

                    Comment


                    • #50
                      Originally posted by geearf View Post
                      Isn't that the same for C++?
                      Because it starts with C too?
                      There is no content in your comment I could respond to.

                      Comment

                      Working...
                      X