Announcement

Collapse
No announcement yet.

Development Pace Of X.Org Is On The Decline

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

  • Development Pace Of X.Org Is On The Decline

    Phoronix: Development Pace Of X.Org Is On The Decline

    Donnie Berkholz presented at FOSDEM 2013 with various X.Org statistics and a look at the health of the development community. Not counting just the X.Org Server but also related components within the X.Org umbrella, the pace of development appears to be on the decline...

    http://www.phoronix.com/vr.php?view=MTI5NDg

  • #2
    X.Org is old and high maintainance. I suppose most are waiting for alternatives like Wayland. There is no need to provide much work for such a project that will become irrelevant after a while...

    Comment


    • #3
      Originally posted by TemplarGR View Post
      X.Org is old and high maintainance. I suppose most are waiting for alternatives like Wayland. There is no need to provide much work for such a project that will become irrelevant after a while...
      Wayland is not trying to be a new Xorg-Server, it just is an alternative in user-space. There may be efforts some day to make a cleaner kernel-space-X-environment (which I would love to see, as currently things like keyboards and mice are redundantly set up), but Wayland isn't it.

      Comment


      • #4
        I would love to see some general input framework.
        Mice, keyboard, joystick, steering wheel, touch screen, others...

        Guess most work now is in remaking things with a clean, well functioning design.
        (Which is by the way a very good reason to remake things.)

        Comment


        • #5
          Michael got one thing wrong. Stats only account X.org server development, and activity on x.org infrastructure around x server.

          MESA, kernel DRM, Wayland, where not measured in that way.

          Those are areas which could account for devs who migrated away from "just" xserver.

          My picks are:

          1) Android. Tons of GPU drivers that somebody need to write. And since those usually are closed source, devs will not be able to join X.org community.
          2) Android/Linux app development. X.org is one among many FLOSS projects that want to have new devs. Joining others is easier.

          Comment


          • #6
            Unprivilegied

            I would like to see XWayland, and unprivileged X server that runs without superuser privileges.

            Comment


            • #7
              Originally posted by frign View Post
              Wayland is not trying to be a new Xorg-Server, it just is an alternative in user-space. There may be efforts some day to make a cleaner kernel-space-X-environment (which I would love to see, as currently things like keyboards and mice are redundantly set up), but Wayland isn't it.
              I hope not. Shoveling everything into kernel is not the solution to provide singleton services. There is no reason to run every componenet at PL0 and unrestricted memory access, I hope we'll see drivers and their respective stacks executed with lower privileges and in their separate adress space in the future.

              Comment


              • #8
                I think it's time to release x12 and just start cleaning it out. I'm sure a 25 year old project like this is full of useless code. It seems like X has been cleaned up a bit the past year but I'm thinking clean it up to the point where its 40% smaller. the reason i say go for x12 is because there NEEDS to be broken compatibility. I feel like one of the things that holds back X is how it keeps trying to remain compatible with hardware nobody uses and hasn't used for years.

                Comment


                • #9
                  Great news!

                  This is excellent news, and I don't mean this sarcastically but forward looking!

                  The X devs are probably the ones with the deepest, most intimate knowledge of X and they are -INVALUABLE- in building the bridge to the future called X on Wayland (XWayland?) and eventually Wayland-only, once the most popular apps are migrated/rewritten to Wayland via cool toolkits like Qt, GTK, etc.

                  I love X, have used it since 1993 on IBM AIX stations back in the days of Mosaic on X... but its time has come and passed, reasserted by the awkward way it's been developed in the last several years where for the sake of performance, the X server has been bypassed in any worthwhile attempt.

                  People go off spouting things like, well Wayland's just a redo of X - yeah? And OS X was "just a redo of OS 9", or in other words, repayment for all technical debt incurred by a platform that died (whether because of Jobs' departure or not, is debatable), and look what OS X did for Apple? Propelled them to the top of the heap and made them the world's MVP.

                  X is currently best labelled as technical debt, which will be repaid by creating a "new X" (I hate to use that analogy, but it serves the purpose) which is all that X wasn't and then some....

                  I implore the X devs to communicate, swallow their egos and unite - all for the purpose of setting the world free - by giving it the next best thing since Linux - Wayland on Linux along with the avalanche of WMs and apps that are soon to follow, all the while not abandoning X-based apps! I still want my 'xeyes' .

                  Comment


                  • #10
                    I do not understand why people think seem to think that open source projects' health is determined by development activity. At some point, everything that needs to be done has been done and there is no point to making further changes. Can anyone who thinks that changes should be made indefinitely name what those changes should accomplish?

                    Comment


                    • #11
                      Originally posted by ryao View Post
                      At some point, everything that needs to be done has been done and there is no point to making further changes.
                      Change & transformation never stops - this is just reality of existence. I'm not sure what your point or question is - can u be more specific?

                      Comment


                      • #12
                        Originally posted by MartinN View Post
                        Change & transformation never stops - this is just reality of existence. I'm not sure what your point or question is - can u be more specific?
                        How much change and transformation does the `echo` command need? As for my question, what changes need to be made to Xorg? I suspect that there is not much that needs revision for most people, which is why development activity is declining.

                        Comment


                        • #13
                          Originally posted by ryao View Post
                          How much change and transformation does the `echo` command need? As for my question, what changes need to be made to Xorg? I suspect that there is not much that needs revision for most people, which is why development activity is declining.
                          Echo vs X? Really?? Come on man.

                          You are right - X needs no change. It needs to die. Wayland is the next generation display server. If u want to find out why X needs to change, read the Wayland rationale from the FAQ.

                          Comment


                          • #14
                            Originally posted by MartinN View Post
                            Echo vs X? Really?? Come on man.

                            You are right - X needs no change. It needs to die. Wayland is the next generation display server. If u want to find out why X needs to change, read the Wayland rationale from the FAQ.
                            My point is that development of any software should eventually taper off once it matures. More complex software takes longer to mature. X works fine on many systems and needs few changes. If Wayland matures, it will be in the same position.

                            As for rationale, I can see applications for Wayland on embedded systems where memory is a premium, but I don't think it makes much difference for most systems people consider to be computers. SGI workstations from the 1980s ran X windows without problems. If computers from the 1980s could run it, then ours should have no problems running it. After all, our computers are many times faster.

                            Comment

                            Working...
                            X