Intel rejection of Ubuntu’s Mir patch

[Sharp_Penguin]

The debate between Mir and Wayland continues

http://arstechnica.com/information-technology/2013/09/intel-rejection-of-ubuntus-mir-patch-forces-canonical-to-go-own-way/

[TheSniperFan]

Sorry, but what "debate"?
It's nothing more than "Canonical vs The rest of the Linux world".
Considering the reasons why they made their own display server, it's no wonder why the there is no one standing behind them.

[UnkendTech]

Sharp_Penguin님이 먼저 게시:

The debate between Mir and Wayland continues

http://arstechnica.com/information-technology/2013/09/intel-rejection-of-ubuntus-mir-patch-forces-canonical-to-go-own-way/

what debate Mir is a fish out of water

Go Intel The hero we deserve!

[arc| Gps]

I agree with intel on this,
Why should they mantain anything just for Ubuntu.
Thats not how Linux works.

And why cant the dev working on mir work on Weyland ?
I would like Canonical to explain that to us.

But I am not holding my breath.

[noam.mor]

You are a company with grand plans. You make a product. Your product's roadmap needs a display server which will run both on mobile and on a PC, be ready in less than 1 year, and you can develop your grand plans on it *right now*. Does Wayland meet all these requirements?

[shaola]

noam.mor님이 먼저 게시:

You are a company with grand plans. You make a product. Your product's roadmap needs a display server which will run both on mobile and on a PC, be ready in less than 1 year, and you can develop your grand plans on it *right now*. Does Wayland meet all these requirements?

yes, it does if you as a company invest half of the time and the money you are going to invest in your own project, but of course then you cannot license it as you like. (Read the license of mir to understand it)

[noam.mor]

All I know is that every website I looked at so far said that Wayland is still a few years from being production-ready. Ubuntu needs a working cross-platform display server 6 months from now for their ubuntu on mobile enterprise. I really dislike their move to Mir as an Ubuntu user - I am on Desktop, I don't really care about mobile devices, and I'm sure things will work worse on Mir. But on the other hand, if they want to let you connect your phone to a monitor and make it a laptop, what other choice did they have? They can't take over Wayland's development; They'd have to fork it, and that wouldn't have been met with any better response from the community. Wayland as an open source project *will not* be what Ubuntu needs it to be in 6 months.

[Cybertao]

Forks happen all the time. Projects split, diverge, re-merge, and combine. Most consider it to be a good thing.
It doesn't usually involve altering your fork underground for months, before announcing it to the world as a new project while slinging insults at the community who developed the original code-base. If Mir was just a fork I doubt anyone would have an issue with it.

As it is no one seems to have an issue with Mir itself. Canonical's arrogant behaviour and development path is the problem.

[Tweak]

iker_sal님이 먼저 게시:

yes, it does if you as a company invest half of the time and the money you are going to invest in your own project, but of course then you cannot license it as you like. (Read the license of mir to understand it)

Problem is investing in doesn't give you control over the direction of the project. This is really where the problem comes from.

Canonical has very specific goals they are trying to achieve with Mir (and Unity) and they didn't feel that Wayland was going in the direction they wanted (as Gnome wasn't).

we currently have 3 very distinct goals in the Linux world right now. Distros focused around the Community, Distros focused around Enterprise users and Distros focused on gaining end user acceptance.

For the most part Canonical is alone in their goal. There are other user friendly distros, however the goal of them is to take the Community and Enterprise tools and make them easier for end users, where Canonical is (somewhat taking from apple) wanting to make something the end user wants.

The former two camps are not generally interested in the average user as most of us who are here can fix our own problems and don't mind working through issues. Problem is we are unlikely to get the mainstream acceptance we want if we all stay on that path.

Can Canonical keep the delicate balance of doing there own thing and being compatible with what the community is doing? Only time will tell, but I don't think it's time to condemn the project till we actually see how it turns out.

[blackout24]

noam.mor님이 먼저 게시:

All I know is that every website I looked at so far said that Wayland is still a few years from being production-ready. Ubuntu needs a working cross-platform display server 6 months from now for their ubuntu on mobile enterprise. I really dislike their move to Mir as an Ubuntu user - I am on Desktop, I don't really care about mobile devices, and I'm sure things will work worse on Mir. But on the other hand, if they want to let you connect your phone to a monitor and make it a laptop, what other choice did they have? They can't take over Wayland's development; They'd have to fork it, and that wouldn't have been met with any better response from the community. Wayland as an open source project *will not* be what Ubuntu needs it to be in 6 months.

Well all your websites are obviously telling crap. The first native Wayland desktop will be released this month with Gnome 3.10. Native Mir desktop is targeted for 14.10, so over a year away! Jolla Sailfish Phone with Wayland will be shipped this year as opposed to Canonicals failed Ubuntu Phone campaign. Waylands reference compositor Weston worked nicely on mobile with multi touch input over 6 months before Canonical announced Mir. Google "Weston Galaxy Nexus".
There isn't a single technical reson why Canonical decided to develop Mir. It's all about creating an Android like ecosystem totally under control by Canonical.

[TheSniperFan]

^This.
But you forgot one important part: Canonicals CLA.
Probably the biggest difference between Wayland and Mir is that Mir stands under Canonicals CLA. I cannot understand why someone would support that.

[UraniumDeer]

Wayland is also fantastic for RPi.
AFAIK, Mir is basically Wayland with server-side decorations, with an Android library that I can't remember the name of right now. This allows them to use Android graphics drivers. It's an oversimplification, but the point is that Wayland is extremely flexible, and can be extended to do everything Canonical needs.

If Canonical had put the man-hours into developing Wayland further, imagine how far it'd be.
Sometimes it's necessary to reinvent the wheel - but it's just senseless to do it twice, when the first reinvention was better!

[noam.mor]

Well. When Wayland is in good shape, I will certainly prefer it to Mir. I doubt that will happen in the next year. This will not be Canonical's fault.

[R3450N]

Wayland isn't quite ready yet, and it will take time for Wayland to actually cement into Linux and fully replace X window system. However, Cannonical are making a head-start and simply taking Mir seriously because quite rightly so the future of computing kinda does lie in mobile computing. What Microsoft tried to achieve is what Cannonical are trying to achieve. Who knows if that's successful or not? I say we should wait until Mir is released, and just see what happens. One thing to take into account would be the differences in how Mir and Wayland are licensed.

[UraniumDeer]

Wayland is in pretty good shape now, apart from some input and some accessibility. Other than that, it's a matter of XWayland being completed and merged into the X-server, and for AMD and nVidia to support EGL. And then of course the DE's need to combine WM and compositor, and make it compatible with wayland, which GNOME soon has a tech-preview, and a release mid next year, and KDE had the same release-frame, but currently there's no official plan as to when they move to Wayland.
I think Enlightenment was planning for a March Wayland release - not at all sure though.

[R3450N]

UraniumDeer님이 먼저 게시:

Wayland is in pretty good shape now, apart from some input and some accessibility. Other than that, it's a matter of XWayland being completed and merged into the X-server, and for AMD and nVidia to support EGL. And then of course the DE's need to combine WM and compositor, and make it compatible with wayland, which GNOME soon has a tech-preview, and a release mid next year, and KDE had the same realease-frame, but currently there's no official plan as to when they move to Wayland.
I think Enlightenment was planning for a March Wayland release - not at all sure though.

Regardless, a split is imminent if Cannonical continue on this path.