Announcement

Collapse
No announcement yet.

Canonical "Won't Fix" GTK+ Wayland For Ubuntu

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

  • phoronix
    started a topic Canonical "Won't Fix" GTK+ Wayland For Ubuntu

    Canonical "Won't Fix" GTK+ Wayland For Ubuntu

    Phoronix: Canonical "Won't Fix" GTK+ Wayland For Ubuntu

    While there has been a Wayland back-end within GTK+ 3.x, Canonical won't be enabling the Wayland support within their GTK+ tool-kit package anytime soon...

    http://www.phoronix.com/vr.php?view=MTI1NTI

  • newwen
    replied
    I believe they're going to enable it in the 13.04 release, because libegl1-mesa already depends on libwayland0

    https://bugs.launchpad.net/debian/+s....0/+bug/954352

    Leave a comment:


  • Darxus
    replied
    Originally posted by ворот93 View Post
    Someone in this thread mentioned that in case of GTK all backends are loaded at once and cannot be separated (unlike Qt).
    Yes. And then you argued with my post without reading it.

    I have no reason to believe the Wayland backend of GTK cannot be broken out into a dynamically loadable library, as in Qt. That just hasn't been done yet, so at the packaging level, it can't be split up. I've opened a bug requesting it in response to the "won't fix" this article is about: https://bugzilla.gnome.org/show_bug.cgi?id=690381

    Leave a comment:


  • ворот93
    replied
    Originally posted by Darxus View Post
    Not if they're broken out into dynamically loadable libraries not included in the iso.
    Someone in this thread mentioned that in case of GTK all backends are loaded at once and cannot be separated (unlike Qt).

    Leave a comment:


  • Darxus
    replied
    Originally posted by ворот93 View Post
    Adding more backends to GTK = increasing distribution size and potentially creating new bugs.
    Not if they're broken out into dynamically loadable libraries not included in the iso.

    Leave a comment:


  • ворот93
    replied
    Originally posted by Darxus View Post
    I think you must have misinterpreted what I wrote. I'm not suggesting increasing their iso size or doing anything I believe will sacrifice stability.
    Adding more backends to GTK = increasing distribution size and potentially creating new bugs.

    Leave a comment:


  • Darxus
    replied
    Originally posted by ворот93 View Post
    Ubuntu is not "other distros" and endless "testing" is not the goal of Canonical. They are not going to sacrifice stability or distributiion size for the sake of testers' convenience.
    I think you must have misinterpreted what I wrote. I'm not suggesting increasing their iso size or doing anything I believe will sacrifice stability.

    Leave a comment:


  • ворот93
    replied
    Because shipping GTK with Wayland support will make it a lot easier for a lot more people to test, making it easier to make Wayland more mature and ready for general consumption. I think the process has already provided very useful testing which has found problems I think other distros would agree should be fixed.
    Ubuntu is not "other distros" and endless "testing" is not the goal of Canonical. They are not going to sacrifice stability or distributiion size for the sake of testers' convenience.

    Leave a comment:


  • Darxus
    replied
    The title of this article is wrong: 'Canonical "Won't Fix" GTK+ Wayland For Ubuntu'

    It didn't get marked "Won't Fix" for all of Ubuntu, as the title says, it got marked "Won't Fix" only for the upcoming Raring release. If you look at the linked bug, its status is still "Confirmed" for Ubuntu.

    I opened the bug. Ubuntu's request that the Wayland backend for GTK to be split out to a dynamic library before being included in one of their releases doesn't seem terribly unreasonable to me. I just wish they mentioned it six months ago, last time they said they wouldn't include it in a release.

    Can't include this in Precise because it's a stable release....
    Can't include this in Quantal because it's too late in the release cycle, even though the bug was open before the Precise release....
    Can't include this in Raring because it would cause the GTK package to depend on a Wayland package....

    This is getting old.

    Originally posted by brent View Post
    I don't see any problem with the decision at this time. Wayland is not mature or ready for general consumption in any way, so why should they ship GTK with Wayland support? The very few users that like to experiment with Wayland just need to compile/install a Wayland-capable build of GTK. No big deal.
    Because shipping GTK with Wayland support will make it a lot easier for a lot more people to test, making it easier to make Wayland more mature and ready for general consumption. I think the process has already provided very useful testing which has found problems I think other distros would agree should be fixed.

    I wrote two different build scripts, and most of the Wayland build instructions, to make compiling from source as easy as possible, and increase testing. I still think it's a pain that far fewer people are willing to go through than those who will install a binary package from a default archive.

    [I switched from Ubuntu to Debian because of Ubuntu's Amazon ads.]

    Leave a comment:


  • GreatEmerald
    replied
    Originally posted by ninez View Post
    indeed, it has swung waaay off-topic....

    ...so the real question is does Magiea offer gtk+ with wayland support, unlike ubuntu?
    It's actually a good question. Mageia is usually more conservative when it comes to defaults, but they also tend to have a lot of options if one chooses to tweak them. For instance, I installed it on one netbook just because it's pretty much the only distro I know of that has good Razor-Qt support. So I wouldn't be surprised if they set the Wayland flag on as well.

    Originally posted by curaga View Post
    Yes, users could. It's merely the fact the kernel is not exactly an inviting environment to submit one's first patch, kind of like X. What would an user do if he can't code, and the patch comes back with "fix X, Y and Z"?

    The author of the patch is in the best position to submit it.
    Yea, that's true. Or the kernel maintainers can do that themselves, although I suppose that going through Ubuntu's patches is not the first idea that pops into mind when trying to deal with an issue...

    Leave a comment:

Working...
X