Kbunutu Users are getting Wayland :) Kubuntu Won't be Switching to Mir or XMir

[UnkendTech]

Kbunutu Users are getting Wayland Kubuntu Won't be Switching to Mir or XMir
i Lol'ed so hard at Jonathan Riddell's Blog today as we know the Kbunutu Developers been trying to get Support From Canonical for Kbunutu KDE setup and now Canonical has pissed off the Kbunutu Developer's, by ****ing them around its sad -_-

it really make me want to help them Develop on wayland

"A few months ago Canonical announced their new graphics system for Ubuntu, Mir. It's a shame the Linux desktop market hasn't taken off as we all hoped at the turn of the millennium and they feel the need to follow a more Apple or Android style of approach making an OS which works in isolation rather than as part of a community development method.

Here at Kubuntu we still want to work as part of the community development, taking the fine software from KDE and other upstreams and putting it on computers worldwide. So when Ubuntu Desktop gets switched to Mir we won't be following. We'll be staying with X on the images for our 13.10 release now in development and the 14.04LTS release next year. After that we hope to switch to Wayland which is what KDE and every other Linux distro hopes to do"

http://blogs.kde.org/2013/06/26/kubuntu-wont-be-switching-mir-or-xmir

[Dolus]

Ursprünglich geschrieben von Gordon Frohman:

Isn't that the whole point of open source, though? Anyone can take other people's work and make it into something different? Most of the complaining I hear about Ubuntu, including early complaints about Mir, is that it's full of developed-in-house stuff that makes it tricky to maintain compatibility with other distros and goes against the community spirit of Teh Linux. So isn't Mir being almost identical to Wayland better than the alternative?

Not if Wayland apps don't run in Mir or vice versa....also Canonicals EULA along with the GPLv3 cause a host of other problems for Mir

[blackout24]

Ursprünglich geschrieben von Nothing:

THE PHILOSOPHIES OMG LINUX ISN'T SUPPOSED TO DO DIS OR DO DAT

Seriously, just get over yourselves and wish BOTH Canonical and the Wayland team the best of luck, they'll need it, especially considering how slowly Wayland's progress has been and how fast Canonical's promise to a stable release (14.04LTS) will have in the next 3/4 year. The one thing that got me a little ticked is when there was a contributor from Wayland that basically said "Oh great, more work for us to do". Really? So, you just do whatever you want which is why Linux lags so badly. Competition should be motivation to innovate and stuff like that. Go Ubuntu, Canonical, Linux, Wayland, whatever that makes MS lose its monopoly! I'm NOT speaking as a dev, I'm speaking as an impatient consumer waiting for whatever the bigger, better "thing" is!

Great another Ubuntard that thinks Wayland devs where sucking their thumbs all the time until great Canonical stepped in and developed something in 6 months. :D Yeah pushing something out the door half assed really should be how software is developed. So where is the toolkit support for your beloved Mir? Ohh there isn't any. Nevermind. Ever thought about why Linux always had 100+ alternatives of evereything except for the display server? Huh? Because low level fragmentation is ♥♥♥♥♥♥♥♥ and doesn't help anyone. Ever wondered why Canonical can do demos like the ones they have shown in the past days? (Apart from forking XWayland) Because of the very thing they are trying to get rid off. A common graphics stack for Linux. Everyone prefers to have just one solid base. And how is Mir going to end MSs monopoly? Everyone was using X just fine with previous Ubuntu releases but now everyone is like OMG MIR NEED IT NOW!!! OMG!!! :D

I'll just quote a post from Phoronix forum that pretty much sums it up.

Originally Posted by r_a_trip View Post
Sometimes these forever noob Windows refugees really show their annoying ignorance. When Wayland was recognised to be a worthy approach to succeed X11 (in the guise of X.org), the Linux graphics stack was not in shape to support the technological requirements of an architecture like Wayland.

So all the years that "Wayland was never finished" the Wayland/X.org devs were busy clearing the ground and building the foundations on which Wayland will run. KMS, graphics memory management and scheduling in the kernel, input improvements, Mesa (OGL) improvements, modularization of X.org with an eye towards Wayland, all of that to create an environment were Wayland can operate.

Somewhere along the way comes Canonical and the Self-Appointed BS-ing^H^H^H^H^H^H Benevolent Dictator For Life and they proclaim Wayland support (with tremendous fanfare) on Ubuntu at such an early stage that it raises eyebrows. Then for the following years, Canonical doesn't do a thing to advance Wayland in any shape or form. Then right at the time that Wayland, the toolkits and the desktop environments get ready to build the new grand display edifice, Canonical swoops in and drops their prefabbed little shack on the corner of the Wayland foundations. They make a lot of noise over their little hut, talk trash over the Wayland project (while largely ignoring wo built the foundations under their shed) and then expect the world to support their little hovel.

Ever since Canonical said "Ooh shiny, graphics, Mir", we have a lot of know-nothing Ubuntu users screaming "Mir, Mir, Mir, Now, Now, Now!" They bought the Wayland FUD hook, line and sinker and never thought about where Canonical is going and if it is where they want to follow. The most funny thing about all this is that before the Mir announcement, we never really heard these people about the display stack. And why would we? Although X11 is old and creaky, the X.org dev team has worked around the deficiencies and has given us a display server with X.org, which serves our needs to this day and while not perfect, it is more than adequate. It has certainly supported Canonical through Warty Warthog to Raring Ringtail.

[aliceif]

I still don't really understand what exactly is "wrong" with X11 and why we need a new thing like Wayland or Mir. I don't recall ever having seen a good explanation about that apart from X11 being old.

[UnkendTech]

Ursprünglich geschrieben von aliceif:

I still don't really understand what exactly is "wrong" with X11 and why we need a new thing like Wayland or Mir. I don't recall ever having seen a good explanation about that apart from X11 being old.

X11 needs root to work, X11 make's why to many call's, you cant get high res out of X11, X11 can't tell an pop up from an Screensaver, X11 use to be most like it's own OS more info in the link below

http://www.phoronix.com/scan.php?page=article&item=x_wayland_situation&num=1

[s.plisskin [Linux Decker]]

The story:

https://www.youtube.com/watch?v=RIctzAQOe44

fixed

[UraniumDeer]

X11:
- Old
- Easy to reach resolutioncap
- Wrapped in lots of compatibility layers
- Almost an entire OS in it self
- Makes lots of unnecessary 'roundtrips', causing stuff like screentearing, inputlag
- Causes instability

Wayland:
- Buffer protocol
- Only does what is absolutely necessary for a displayserver.
- Fast
- Responsive
- Very high resolution supported
- Pixel perfect frames
- Created by the very people that is maintaining X

XWayland:
- Runs X-applications with an X-server as a client of Wayland, resulting in less inputlag, nicer frames, while not creating a lot of overhead (unless running several X-apps I guess), as the X-server thinks that it's in fullscreen (iirc).

Mir:
- Wayland + [edit] LibHybris [/edit]

XMir:
- ~XWayland

Correct me if I'm wrong, or better yet, make a better/more correct summation than the above :)

[UnkendTech]

Ursprünglich geschrieben von UraniumDeer:

X11:
- Old
- Easy to reach resolutioncap
- Wrapped in lots of compatibility layers
- Almost an entire OS in it self
- Makes lots of unnecessary 'roundtrips', causing stuff like screentearing, inputlag
- Causes instability

Wayland:
- Buffer protocol
- Only does what is absolutely necessary for a displayserver, and does it well (in the Unix spirit)
- Fast
- Responsive
- Very high resolution supported
- Pixel perfect frames
- Created by the very people that is maintaining X

XWayland:
- Runs X-applications with an X-server as a client of Wayland, resulting in less inputlag, nicer frames, while not creating a lot of overhead (unless running several X-apps I guess), as the X-server thinks that it's in fullscreen (iirc).

Mir:
- Wayland + Surfaceflinger + a bit of canonical code

XMir:
- ~XWayland

Correct me if I'm wrong, or better yet, make a better/more correct summation than the above :)

the Unix part is a little off as unlike Xorg Wayland is not being developed the Unix way :)

[blackout24]

Mir doesn't share any code with Surfaceflinger which is Androids Display Server.
They use libhybris to use it on Android GPU drivers, which is not developed by Canonical but by the Mer Project and also allows you to use Wayland on Mobile devices: http://mer-project.blogspot.se/2013/04/wayland-utilizing-android-gpu-drivers.html

[UraniumDeer]

Ursprünglich geschrieben von Doc Holliday:

the Unix part is a little off as unlike Xorg Wayland is not being developed the Unix way :) :tradingcard:

I can't remember where I saw it, but someone said, that making something perform something specific, and do it well, is the Unix-way. One can't exactly say that Xorg does that :P
But then, I'm still rather noobish in all this, so I'm just going to trust you, and edit my summation :)

Ursprünglich geschrieben von AbartigerNorbert:

Mir doesn't share any code with Surfaceflinger which is Androids Display Server.
They use libhybris to use it on Android GPU drivers, which is not developed by Canonical but by the Mer Project and also allows you to use Wayland on Mobile devices: http://mer-project.blogspot.se/2013/04/wayland-utilizing-android-gpu-drivers.html

Now that's rather interesting. I've definitively read something that could be misunderstood as Canonical using code from Surfaceflinger, but I know I've read something about them using libhybris, so I've misunderstood that.

Thanks for the corrections :)

[blackout24]

Ursprünglich geschrieben von UraniumDeer:

Now that's rather interesting. I've definitively read something that could be misunderstood as Canonical using code from Surfaceflinger, but I know I've read something about them using libhybris, so I've misunderstood that.

Thanks for the corrections :)

Probably because early Ubuntu Touch images used Surface Flinger and still do, until they swap it out for Mir. Here is what they had to say about Surface Flinger.

We looked further and found Google’s SurfaceFlinger, a standalone compositor that fulfilled some but not all of our requirements. It benefits from its consistent driver model that is widely adopted and supported within the industry and it fulfills a clear set of requirements. It's rock-solid and stable, but we did not think that it would empower us to fulfill our mission of a tightly integrated user experience that scales across form-factors.

Source: http://fridge.ubuntu.com/2013/03/04/mir-an-outpost-envisioned-as-a-new-home/

I think the important word here is "empower". Pretty much saying that the only reason they rolled their own is to control it.

[allenmaher]

Both Mir and Wayland will one day be better than X-org, they are very similar. Mir will likely be out first on people's desktops (October), wayland will be on more distros other than Ubuntu. X-org will be the steam gaming choice for the next little while so no need to rush to judgement. I will likely not be gaming on Xmir in 13.10 unless the performance picks up which it may since a few features have yet to be implemented that will speed things up.

Pharonix's tests of Xwayland showed some promising results on very limited tests, however those same tests showed that xwayland on the test system still crashed in 3D applications. It is early days for both.

The whole ecosystem spliiter debate is a red herring. We have multiple desktop environments, multiple office suites, multiple audio servers and so on and so forth. Wayland is itself splitting the display server by competing with X-org. The whole argument is silly.

Mir's main focus right now is Unity and Ubuntu Touch. They have offered to submit patches to gnome and kde for mir support, however the maintainers have turned them down. Wayland while it looks great and I have installed it and played with the demos for years, is very very slow in getting out the door.

Which ever one you end up using it will take a long time to get all your applications using it natively. This will not be a quick transition. For those who use proprietary drivers (AMD and NVIDIA folks) the first sign of any support is Mir in Ubuntu 14.04. But for that to matter for steam games we need a Mir/Wayland aware SDL layer which it seems there are test releases of both available.

In the words of Douglas Adams, DON"T PANIC.

This will be a long slow bumpy ride and games work on X-org now and for the next few years.

[UnkendTech]

Ursprünglich geschrieben von allenmaher:

Both Mir and Wayland will one day be better than X-org, they are very similar. Mir will likely be out first on people's desktops (October), wayland will be on more distros other than Ubuntu. X-org will be the steam gaming choice for the next little while so no need to rush to judgement. I will likely not be gaming on Xmir in 13.10 unless the performance picks up which it may since a few features have yet to be implemented that will speed things up.

Pharonix's tests of Xwayland showed some promising results on very limited tests, however those same tests showed that xwayland on the test system still crashed in 3D applications. It is early days for both.

The whole ecosystem spliiter debate is a red herring. We have multiple desktop environments, multiple office suites, multiple audio servers and so on and so forth. Wayland is itself splitting the display server by competing with X-org. The whole argument is silly.

Mir's main focus right now is Unity and Ubuntu Touch. They have offered to submit patches to gnome and kde for mir support, however the maintainers have turned them down. Wayland while it looks great and I have installed it and played with the demos for years, is very very slow in getting out the door.

Which ever one you end up using it will take a long time to get all your applications using it natively. This will not be a quick transition. For those who use proprietary drivers (AMD and NVIDIA folks) the first sign of any support is Mir in Ubuntu 14.04. But for that to matter for steam games we need a Mir/Wayland aware SDL layer which it seems there are test releases of both available.

In the words of Douglas Adams, DON"T PANIC.

This will be a long slow bumpy ride and games work on X-org now and for the next few years.

the guy who had the Xwayland installed did not know what he was doing and was using an Ironlake gpu so far out of date it was not funny

[blackout24]

Ursprünglich geschrieben von allenmaher:

The whole ecosystem spliiter debate is a red herring. We have multiple desktop environments, multiple office suites, multiple audio servers and so on and so forth. Wayland is itself splitting the display server by competing with X-org. The whole argument is silly.

Mir's main focus right now is Unity and Ubuntu Touch. They have offered to submit patches to gnome and kde for mir support, however the maintainers have turned them down. Wayland while it looks great and I have installed it and played with the demos for years, is very very slow in getting out the door.

Sorry but that is easily the stupidest comment I have ever read. The guys who develop wayland have been X.Org hackers for years. Everyone agreed that we need an X server replacement and everyone agreed that it should be wayland. And now you are comparing having multiple display servers to having multiple browsers or destop enviroments. OUCH!
Fragmentation is stupid on the low level. Why not fork or rewrite the kernel or glibc? Everyone rather prefers having one solid solution on which everyone can build there house on. Riddle me this: Why hasn't the Linux communtiy spawned dozens of forks and rewrites of X over the last 15 years? It's not like we don't liek to rewrite thing and fork. Because everyone agree that low level fragmentation is bullsh!t!
Also good job ignoring all the work that had to be done just to build an enviroment on which Wayland can run. You know toolkits, mesa improvments, modularizing X etc. etc.

[aliceif]

There IS a kernel fragmentation amoung current X11 (not necessarily Linux) users.
You forgot that FreeBSD and most other BSD UNIXes use the X server for UI, and those do not use the Linux kernel (actually Arch and Debian even have versions using the FreeBSD kernel). And there are multiple Linux kernels in use because every distro makes adjustments to the kernel they use, they use different versions (3.2, 3.5, 3.8 are all used in currently supported ubuntu versions, 3.9 in Arch and others. ) and there is also the Linux-libre kernel used in some distros.
In other words, there is more low-level fragmentation among the FOSS community than you think and X11 is the only thing that is used everywhere.
And X11 vs Wayland vs Mir vs whateverelsemightcome is not just a Linux issue, which means variety is not bad.

[Zyro]

Ursprünglich geschrieben von AbartigerNorbert:

Why not fork or rewrite the kernel or glibc?

You mean like egcs, eglibc, uClibc, dietlibc and hurd? ;)