Intel Reverts Plans, Will Not Support Ubuntu's XMir

[chrisb]

Without telling us how you come to this conclusion this is a worthless statement.

Mint's primary claim to fame is being a niche for people who want to run Gnome forks Cinnamon/Mate. I think in the long term Gnome will get it's act together, fix the problems, maybe adopt some of the popular bits of Cinnamon, and people (Fedora) will choose it over the forks. RedHat at least have motivation to spend money and make this happen.

Funny that people don't complain about fragmentation when it comes to Gnome forks...

[bkor]

Funny that people don't complain about fragmentation when it comes to Gnome forks...

Most GNOME developers are totally ok with the forks. MATE is basically continuing GNOME 2. They wasted a lot of effort in that (NOT reused the port to dconf + gtk+3.x), but who cares. Then Cinnamon tries something different as well. Some of the MATE developers now maintain GNOME modules that we don't need/develop anymore using the GNOME infrastructure (git, ftp). Within GNOME we asked if anyone wanted to maintain the fallback-mode modules (partly due to getting requests from developers wanting to do this).

For Upstart, systemd, OpenRC and sysvinit I can quite easily tell the differences among these. Different goals.

For Mir, my main issue is not being able to understand why it exists (I can easily understand MATE+Cinnamon) plus the secret change of direction. That together with the way that Mir was announced (misinformation about Wayland) results in one thought: "urgh!"

[oyvind]

Restricting their driver's upstream support for a display server on political grounds is a surprisingly childish move by Intel ! Wayland still QQ over some competition.

[TheBlackCat]

Restricting their driver's upstream support for a display server on political grounds is a surprisingly childish move by Intel ! Wayland still QQ over some competition.

You haven't read any of the discussion here? It has been pointed out over and over that it isn't Intel's responsibility to maintain Canonical's in-house project for them. If Canonical really backs Mir they should be willing to support it themselves. If they don't want that responsibility, why should Intel?

[dh04000]

For Mir, my main issue is not being able to understand why it exists

You see how getting <100 lines of code into the driver went for them? Could you imagine trying to get features they needed in Wayland would have gone? What happens when a product gets pushed back because some jerk living in his parent's basement in Romania doesn't like an essential <100 line patch from them? Investors can't depended on the whim of basement dwellers in Romania when investing. A company needs some control of their products as well. This is why opensource and business have a hardtime getting along.

[Vim_User]

Facts and realism is a bit scarce in these parts.

Correct. I see neither side back up their claims with facts, so a discussion about the size of userbases is pretty pointless.

[Vim_User]

Funny that people don't complain about fragmentation when it comes to Gnome forks...

Because it is irrelevant. Any software should run on any DE without extra maintenance.

[Vim_User]

You see how getting <100 lines of code into the driver went for them?

Yes, we see how it worked out trying to shift the burden of maintenance for their in-house solution to upstream. Oh, wait, that is not what you wanted to say, or?

Could you imagine trying to get features they needed in Wayland would have gone? What happens when a product gets pushed back because some jerk living in his parent's basement in Romania doesn't like an essential <100 line patch from them? Investors can't depended on the whim of basement dwellers in Romania when investing. A company needs some control of their products as well. This is why opensource and business have a hardtime getting along.

Business and open source only have a hard time to get along when your business mostly consists out of dick moves and wanting to be the leader of a movement you don't contribute to. If you do it the proper way, like for example Red Hat does, their is no problem with business and open source.

[matzipan]

Intel has clearly chosen Linux over Ubuntu.

Canonical, you are welcome to come back to the Linux fold any time. But if you want to continue to build yourselves into a walled garden, then you need to take responsibility and care of every single one of your own plants and trees and shrubs within your own walled off garden. Don't expect us to come in and be your slave laborers.

FFS, Mir vs Wayland happens at toolkit level, anyway. Nobody has to ever care about Wayland vs Mir other than GTK and Qt. They are the only ones doing the extra friggin work for it.

To be expected since Intel's main enemy is ARM and Mir will promote mobile devices.

I would argue that Ubuntu is probably the only thing closest to Intel entering the mobile market. Tizen? never cared about it and never will. Ubuntu Phone + Desktop Dock? It's probably the only chance where Intel is a lot superior than ARM.

The patch was a mere buffer redirect from the video driver to mir, it just gave a major boost.

[F i L]

Funny that people don't complain about fragmentation when it comes to Gnome forks...

Because it's just another X11 DE.. meaning any app targeting X11 can run on it as-is. Compare that to Mir, which will require application devs to target it's API and compile separate (from Wayland & X11) builds to run... that applies to every linux application (luckily GTK/Qt will take much of that burden away from devs which target those APIs, but not all of it). Two different DS also complicates the life of AMD, Nvidia, and Valve (right when they're pushing for Linux as a gaming platform).

In short, the 'fragmentation' of a DE is no where near the same as fragmentation of a DS.