Announcement

Collapse
No announcement yet.

Wayland, Weston 1.2 Release Candidate Are Out

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

  • #21
    Originally posted by schmidtbag View Post
    Hah too bad this is the internet because we could make this interesting by betting which will be the official replacement to X11. Personally, I'm 60% sure Mir is going to come out on top, though I would prefer if wayland did.
    I would say that is far from a safe bet because, as other people have stated, Canonical has very little pull in of itself. Sure, it has its vaunted market share, but it still does not have enough users to make it truly powerful (I would even wager that the majority of Linux users use another distro, even if Ubuntu probably does have a strong plurality of them). Market share is also very hard to quantify, and as such is rather hard to use to ones advantage without a clear majority like Microsoft has.

    Wayland has almost the entire FOSS apparatus behind it and a significant amount of commercial support. Most major distributions other than Ubuntu have committed to using it, as have almost all of the major desktop environments besides Unity. Almost no one in the business world uses Ubuntu and instead use RHEL and SUSE, meaning that the business sector is going to end up using Wayland, and that is where the money is. And, as Wayland is the approved successor to X backed by Xorg, it is also the safer choice for businesses worried about change.

    So I do not think the battle, if you will permit the term, is so heavily weighted towards Mir as you might think.
    Last edited by Hamish Wilson; 10 July 2013, 02:33 PM.

    Comment


    • #22
      Originally posted by LinuxGamer View Post
      multi-seat support is a WIN
      Multi-seat support was in X since forever and a day.

      Comment


      • #23
        Originally posted by jrch2k8 View Post
        well wayland will go public ever? when good enough and only when few glitches corner cases are expected

        Mir will reach users first if they dont hurry? well to be honest no even 1 developer give a rat ass about it, for wayland there no risk at all in waiting until is good enough, is just that simple

        why, OMG ubuntu the desktop OS for the world bla bla bla? Wayland is backed by FSF freedesktop.org and Xorg foundation to start with and is been backed and supported by Samsung, Digia, Intel, AMD, RedHat just to name some fortune 500 companies add to that the community and basically every toolkit and software with a GUI on linux, so unlike canonical they don't have to force alpha software + ugly hacks so ppl can say OMGz Mir can move windows!!!, if you wanna test wayland use Arch or Gentoo, if you are not an advanced user wait for fedora 20[native wayland support with gtk and gnome 3.10]

        XWayland, so ppl can see windows moving too? unlike canonical wayland devs don't get in the wild alpha software or incomplete projects, XWayland was a toy needed to test a point that revealed some weaknesses in the stack that are been fixed since then and some that will come with DRI3000, until then XWayland is on freeze but canonical in their desperation to show something and make some PR is pushing software that is not meant to be public + very ugly hacks[XServer fullscreen on top of Mir is horribly wrong]

        if ubuntu first mir win bla bla bla? well ask in Nvidia or AMD forums where their Linux money come from[hint: ubuntu is many many million dollars away] and check which projects that distro support, that should be answer enough and no Games are not massive enough yet to reduce the money income difference enough for their to care beyond give standard EGL support.

        why XMir is good enough for canonical then? well canonical jumped waters thinking their preach would be listened and it was not[regular uses don't count here, they don't port software or do anything technically useful] and launching a project with that much PR to find out 5m later they decided to make this "server-compositor-thingy" based on wayland myths that nobody in canonical was smart enough to crosscheck proving all their technical reason were absolutely wrong[and piss off to death every developer out there that could be useful for the project] killed all the probable support from the community and their credibility as a bussiness, so right now their trying to recover somehow making this PR monkey show with XMir[this code is so hacky that i doubt even microsoft will publish it] trying to win average joes users and see if they can put pressure enough[is not working tho, all their Mir patches or begs for help receive an appropiate fuck off, read mailing lists]

        anyway wayland should be native and available in most distros this year for Gnome and EFL, and Q1 2014 for first KDE SC 5 betas releases[including kubuntu]

        and fedora rawhide should have gnome 3.10 in repos, you can try that too
        I'm not sure why you and everyone else are talking so much about Canonical, I asked about wayland. You didn't get to my question til the ask sentence, and still didn't provide any sources.

        Comment


        • #24
          Originally posted by dh04000 View Post
          I'm not sure why you and everyone else are talking so much about Canonical, I asked about wayland. You didn't get to my question til the ask sentence, and still didn't provide any sources.
          i linked all this in at least 70 different post already, learn to google and use mailing lists and you will always be informed.

          ok, in easier english XWayland is not ready for any sort of production use yet until dri3000 and Xorg 1.15 gets here and XWayland[as designed] is rootless, aka XWayland can't use a fullscreen Xserver without the same dirty ugly hacks that XMir uses[heavily rejected from upstream for ovbious technical reasons]

          Comment


          • #25
            Originally posted by dh04000 View Post
            I'm not sure why you and everyone else are talking so much about Canonical, I asked about wayland. You didn't get to my question til the ask sentence, and still didn't provide any sources.
            Well, keep ignoring that I already gave the ultimate answer to that question: there is no point in running a desktop on top of a nested X server over some other display system. That's why no distro plans or does that. It's a bad design practice to increase bug surface for no win.

            Comment


            • #26
              Originally posted by mrugiero View Post
              Well, keep ignoring that I already gave the ultimate answer to that question: there is no point in running a desktop on top of a nested X server over some other display system. That's why no distro plans or does that. It's a bad design practice to increase bug surface for no win.
              for real a nested XServer/Layer no make sense you can't be more right

              Comment


              • #27
                Originally posted by AJenbo View Post
                We already know that enough prices are missing that it would not be ready to be pushed on the users of a major distribution. Proprietary driver still do not have support, no major de has been adapted for it, xwayland has some issues and is not merged yet. Probably the first major distribution to use it will be fedora, but until all the pieces are in place your probably only going to find it in distributions meant for testing. Having everything run inside a xwayland window is of no benefit, and problematic as long as there isn't good driver coverage.
                dont bring in propriatary drivers as reason for not using wayland thats maybe a reason for a distro for retards... but normaly opnesource devs dont wait for propriatary support before they release their shit... if all works good enough on the free drivers but nvidia and co needs another 10 years to support it... nobody will wait for this retards.

                Comment


                • #28
                  Originally posted by blackiwid View Post
                  dont bring in propriatary drivers as reason for not using wayland thats maybe a reason for a distro for retards... but normaly opnesource devs dont wait for propriatary support before they release their shit... if all works good enough on the free drivers but nvidia and co needs another 10 years to support it... nobody will wait for this retards.
                  But isn't that simply the definition of major distros to you any way?

                  Comment


                  • #29
                    Originally posted by GabrielYYZ
                    I've been meaning to ask a couple questions for a really long time but never got around to it, if anyone can answer or, at least, give an educated guess it'll be much appreciated.

                    Is it a bad idea to adapt weston to become the "standard" system compositor and use kwin/openbox/GNOME's/EFL's compositor as a session compositor? wouldn't that simplify much of the work the various DE's need to do? If not, why?
                    I don't think two compositors can be a good thing. If you run a desktop, it shouldn't be any surface on the same level, so running that on top of a compositor that will only composite that surface is nonsense. As if it could be done, I have no idea. But I don't think there's any reason to. What can be done and there are reasons to do so (no hard ones, but it would make life easier for a not so experienced dev) is using weston codebase as a reference, which is what weston is for. You can see the code and how a compositor should behave, and then improve on top of that to achieve what you need.
                    I fear a little bit for Openbox and the like. There wasn't too much activity on their repos lately, so I don't think they'll have the manpower to port to wayland anytime soon.
                    One could say wayland is thought for higher end devices and the like, but one could as well use the simplest desktop possible while leveraging wayland.

                    Comment


                    • #30
                      I don't think weston is a system compositor after reading the Wayland documentation. Gnome Shell and KWin will be wayland session compositors. A system compositor isn't even a requirement as it looks. Projects like DWM, Awesome, Openbox would only have to modify the "shell.c" in the weston git repo. It implements the desktop shell and window handeling with around 4000 LOC (DWM is about 2000 LOC afaik). So they'd use weston as session compositor and implement their window handeling in a module like the weston-desktop-shell (shell.c). You can load these modules in weston.ini. https://github.com/krh/weston/blob/master/weston.ini
                      Last edited by blackout23; 10 July 2013, 04:12 PM.

                      Comment

                      Working...
                      X