Announcement

Collapse
No announcement yet.

GNOME Will Move Full-Speed With Wayland Support

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

  • #31
    Well it seems that when the power of bitchyness is greater than the power of collaboration to get things done. It really feels like a reactionary move done almost as much out of spite than for technical reasons. Seriously it took canonical saying "your solution isn't good enough, we're rolling our own" for gnome to get off it's arse and actually get wayland support under way. How long would it have taken if canonical had instead depended on gnome doing it in their own time? Certainly too long to be of use for a mobile platform that's launching in October. And from what i've seen, canonical's contributions would be turned down only for the equivalent functionality to be re-implemented in an incompatible way about 2 years after they'd be of use to anyone (notifications anyone?)

    Comment


    • #32
      Originally posted by mangecoeur View Post
      Well it seems that when the power of bitchyness is greater than the power of collaboration to get things done. It really feels like a reactionary move done almost as much out of spite than for technical reasons. Seriously it took canonical saying "your solution isn't good enough, we're rolling our own" for gnome to get off it's arse and actually get wayland support under way. How long would it have taken if canonical had instead depended on gnome doing it in their own time? Certainly too long to be of use for a mobile platform that's launching in October. And from what i've seen, canonical's contributions would be turned down only for the equivalent functionality to be re-implemented in an incompatible way about 2 years after they'd be of use to anyone (notifications anyone?)
      Great to distort history, but that is not how things happened. In case you want to learn and not just troll, read my previous messages.

      In brief: "Canonical: We go to Wayland". GNOME makes several trials. Then Canonical does not communicate and changes direction without informing.

      Canonical is the one not communicating here and keeping things private. Wtf with calling GNOME reactionary? Note that we already have done loads of work for Wayland.

      PS: Wayland != GNOME.

      Comment


      • #33
        Originally posted by agd5f View Post
        Regardless of whether you prefer wayland or mir, both are currently lacking a certain amount of what some would call essential functionality in comparison to X. Off the top of my head:
        - Multi-GPU support
        - Hybrid laptop (PowerXpress, Optimus) support
        - modeswitching API
        - multi-display API
        These are obviously not insurmountable, but still a lot of work. Just something to keep in mind.
        I don't have any Idea how this works and how it is implemented but Weston works with two displays for me.

        Comment


        • #34
          Originally posted by agd5f View Post
          - Hybrid laptop (PowerXpress, Optimus) support
          I am pretty sure even X doesn't have that, or did Torvald cursing Nvidia actually had an effect I never heard of ? Cause if yes, I want the solution, my thinkpad would appreciate it. =p

          Comment


          • #35
            Originally posted by bkor View Post
            Wtf with calling GNOME reactionary? Note that we already have done loads of work for Wayland.

            PS: Wayland != GNOME.
            I thought it was well established that GNOME just can't win. A strawman is always convenient.

            I've tried to follow most of the discussions regardign the move but here's what I still don't get:
            Which single component requires the most work for GNOME to default to Wayland in something not overly conservative but aimed at a large number of users, like Fedora? Is it the protocol? Weston (even relevant?)? Mutter? Gtk? or something else?
            Last edited by Kostas; 03-13-2013, 10:44 AM.

            Comment


            • #36
              Originally posted by iniudan View Post
              I am pretty sure even X doesn't have that, or did Torvald cursing Nvidia actually had an effect I never heard of ? Cause if yes, I want the solution, my thinkpad would appreciate it. =p
              It's not perfect, but things like bumblebee make it workable.

              X gets lots of flack (some of it justified, much of it not), but it has loads of functionality which is extremely useful in corner cases, and which some of us need. Wayland is aiming at covering the most common use cases first, before tackling such stuff. So I expect X to be a factor for a long time to come.

              I don't know how exactly this is handled, but I can not use a system without working multi-seat. This requires multi-seat aware login manager and multi-seat aware input handling. I don't think that this works in Wayland right now, or that it is much of a priority. Just one example.

              Comment


              • #37
                Originally posted by Kostas View Post
                I've tried to follow most of the discussions regardign the move but here's what I still don't get:
                Which single component requires the most work for GNOME to default to Wayland in something not overly conservative but aimed at a large number of users, like Fedora? Is it the protocol? Weston (even relevant?)? Mutter? Gtk? or something else?
                There is not just a single component that can make GNOME work under Wayland. If one is adjusted and the rest is not, the result is failure. That is why we release things together. At the moment the full plan is still being made. This is logical as the email only went to release-team, there actually is no decision at this time as we (as release team) did not announce anything. We generally do not dictate things amongst the GNOME developers and try to get consensus.

                If you want to see what needs to be done: https://live.gnome.org/Wayland

                It has a list of applications and if they work on Wayland or not. It also lists all the things that needs to be done.

                High level speaking the most critical components are: Mutter, GTK+. Everything is being investigated and written down on above link.

                Comment


                • #38
                  Originally posted by bkor View Post
                  That's odd, because Canonical did the same thing with Unity: Developed it internally, never communicated.

                  We had a design hackfest where Canonical did their own meetings.

                  So interesting that you seem to have more knowledge about this... care to cite your references?
                  First of all, thanks for all your great work on Gnome. It is highly appreciated! I have to admit that I did not follow the process around Canonical's relationship with Gnome prior to the start of Unity closely, but I did catch some reports and impressions around the Internet. Firstly, Ubuntu seemed to do wonders for the Gnome desktop with numerous patches, the most prominent Gnome version in this respect I believe is in Ubuntu 10.04. These patches seemed to have a hard time being accepted upstream in Gnome (again my perception, I did not follow this process closely). Could you shed some light on this? Did Canonical have a hard time getting their improvements accepted upstream in Gnome2?

                  Comment


                  • #39
                    Is it me or ever since Canonical announced the Mir display server, everyone else seems to be doubling down on implementing Wayland support?

                    Comment


                    • #40
                      Originally posted by Krysto View Post
                      Is it me or ever since Canonical announced the Mir display server, everyone else seems to be doubling down on implementing Wayland support?
                      Doubling down with words at least.

                      Comment


                      • #41
                        Originally posted by Del_ View Post
                        First of all, thanks for all your great work on Gnome. It is highly appreciated! I have to admit that I did not follow the process around Canonical's relationship with Gnome prior to the start of Unity closely, but I did catch some reports and impressions around the Internet. Firstly, Ubuntu seemed to do wonders for the Gnome desktop with numerous patches, the most prominent Gnome version in this respect I believe is in Ubuntu 10.04. These patches seemed to have a hard time being accepted upstream in Gnome (again my perception, I did not follow this process closely). Could you shed some light on this? Did Canonical have a hard time getting their improvements accepted upstream in Gnome2?
                        Depends on how the work was done: If you submit patches, then everyone is treated the same way. Big changes need to be discussed, etc. If you develop something for 6 months and then submit a patch, the change of it being accepted is lower. Likely the maintainer has a different idea, etc.

                        Canonical is and will not be treated any different than any other contributor. Meaning: Sometimes happens that people forget about patches, while at the same time respond more quickly to a patch from a familiar name. That is just the way things go. But if someone pretty much dumps some code without discussing things beforehand, then likely the maintainer will have a different idea on how things should've been implemented. Which increases the chance that the patch is either ignored (because it is massive) or rejected for needing more work (because maintainer doesn't want it that way).

                        Easy ways to ensure your patches are accepted: Discuss beforehand, become a "known" person, become a maintainer. Pretty much the same as in any other free software project.

                        Within GNOME, maintainers have a lot of decision power (which is logical IMO as usually the maintainer will end up supporting the code).

                        PS: There have been some huge controversies in the past regarding this, 2.x timeframe. At the same time, IIRC there also have various contributions that were accepted outright.

                        Comment


                        • #42
                          Originally posted by Alex Sarmiento View Post
                          What if Wayland moves its ass once for all? It is almost vaporware.
                          Its not vaporware, wayland/weston has been fully usable/testable for ages, download rebeccablackos and try it yourself. The things blocking it from adoption are this:

                          1. EGL drivers

                          2. DE's like gnome and kde porting to wayland

                          Until AMD/Nvidia release an EGL driver wayland will be OSS drivers only, and until gnome and KDE finish porting to wayland its rather pointless to adopt it into a distro which is why it hasn't happened yet... once gnome and kde support wayland we will see cutting edge distros support it optionally, we probably won't see it used by default in "major" distros until their is driver support from nvidia/amd.

                          Comment


                          • #43
                            I HAVE THIS FEELING

                            Comment


                            • #44
                              Originally posted by johnc View Post
                              Doubling down with words at least.
                              List of GNOME apps already working under Wayland before the Mir announcement:
                              https://live.gnome.org/Wayland/Applications

                              Obviously even if some application works, it does not mean it is the same experience, which is why the support is planned over 2 releases.

                              Comment


                              • #45
                                Originally posted by johnc View Post
                                Doubling down with words at least.
                                words is all canonical has too. at this point mir barely does anything and is far less functional/complete than wayland. yet canonical claims they can have mir finished, have backends for GTK/QT finished, have unity totally rewritten in QT5/QML and ported to mir by 2014? Don't make me laugh.

                                Comment

                                Working...
                                X