Ubuntu 14.04 Codename Revealed, Mir Haters Attacked

[LinuxGamer]

People who submit commits to Mir could possibly submit them to Wayland. Plain and simple.
Mir uses license that, as far as I understand doesn't allow forking. Meaning, Mir is owned by Ubuntu and not you. So why support THEM instead of community?

And frankly, there is nothing good about DEB VS RPM.

Mir vs Wayland as Wayland and Mir will not use any of the same code in a few years/year this will be catastrophe for Linux users and Linux developers

.deb vs .rpm is a joke you can even install RPM's on a Deb system
heres a little into on Rpm https://wiki.debian.org/RPM

[mrugiero]

People who submit commits to Mir could possibly submit them to Wayland. Plain and simple.
Mir uses license that, as far as I understand doesn't allow forking. Meaning, Mir is owned by Ubuntu and not you. So why support THEM instead of community?

And frankly, there is nothing good about DEB VS RPM.

The license allows for forking, though they stated they want to discourage it and might intentionally break compatibility if such forks start to exist. Also, you are the owner of your patches, you only give them a right to sublicense (it's unfair and asymmetric, but not nearly as bad as you painted it), but you are still free to commit your own patches anywhere you want to with any license you want.

[ua=42]

Can someone please explain to me what is so bad about MIR. I've looked at it and it looks like a nice improvement. I don't know why everyone bashes it. I'm not trying to defend it, I'm just looking to get educated about what the problem with it is. I mean, there isn't anything else to use righT? There's Weston but that's just a reference thing and not even fully operational display server.

Wayland is a protocol. Why is it bad to have MIR display server based on the protocol. Thanks.

A. Mir doesn't have a protocol. Meaning every change to mir would require rebuilding/installing new versions of every toolkit that uses it or things would break.
B. Performance wise, they should be near identical. Wayland was started a long time before MIR. Ergo, some people are annoyed they didn't help with developing it instead of spending even more money making their own duplicated system.
C. Technically, the way Wayland does things vs MIR is more elegant/simple which should make maintence lower.

There is also things Canonical has done PR wise to piss of everyone involved with wayland.
A. Canonical in their initial PR said allot of false negative statements about wayland.
B. Mir copied XWayland wholesale and released a big PR about their solution for compatibility on MIR.
C. This latest press conference.
D. The fact they didn't help the project that was further along and instead spent even more money coming up with their own solution.

[LinuxGamer]

A. Mir doesn't have a protocol. Meaning every change to mir would require rebuilding/installing new versions of every toolkit that uses it or things would break.
B. Performance wise, they should be near identical. Wayland was started a long time before MIR. Ergo, some people are annoyed they didn't help with developing it instead of spending even more money making their own duplicated system.
C. Technically, the way Wayland does things vs MIR is more elegant/simple which should make maintence lower.

There is also things Canonical has done PR wise to piss of everyone involved with wayland.
A. Canonical in their initial PR said allot of false negative statements about wayland.
B. Mir copied XWayland wholesale and released a big PR about their solution for compatibility on MIR.
C. This latest press conference.
D. The fact they didn't help the project that was further along and instead spent even more money coming up with their own solution.

and like i said catastrophe for Linux users and Linux developers

[Kostas]

Having a hard time accepting that MS may believe any of the things he wrote. This strikes me as something between a desperate PR move to shift the focus from the XMir debacle to the Mir "haters" and a rally to the Ubuntu faithful. Either way it's sad because it make the chances of Canonical reverting to Wayland (if &) after Mir's development falls flat on its face.

[LinuxGamer]

Having a hard time accepting that MS may believe any of the things he wrote. This strikes me as something between a desperate PR move to shift the focus from the XMir debacle to the Mir "haters" and a rally to the Ubuntu faithful. Either way it's sad because it make the chances of Canonical reverting to Wayland (if &) after Mir's development falls flat on its face.

Apple use to play the Victim or make there users feel like they was the Victim and now Canonical is trying the same thing but in reality the Linux user base is the Victim's of Canonical PR campaign and to Canonical its Do or DIE (suicide) and they're taking every one down with them they can

[Alex Sarmiento]

Well, aside from GNOME Shell benefiting Ubuntu GNOME users currently, and the relative fundamentality of display technologies vs. package managers, I'd say being open source isn't the most relevant thing to most computer users. I think what is most important is that your software empowers you, and it's no coincidence that the best general purpose software tends to be open source. Of course, that isn't to say that all open source software empowers you, as much of it is frustrating to use, despite the developers' intentions.

A sense of security is only as good as the rationale it's based on- delusion isn't a good thing. Familiarity is nice, but hardly indicates an objective quality. I don't find a problem with people doing what they're used to, of course, if it's not inherently 'bad'. But I can't say I advocate actions that would tear our community apart without a damn good reason. If you want to create a whole new display server, and you want to convince the developers of various frontend software to come along, you're going to need a lot better reasons than, "well, we're the best, and this is where the party's at, so you kinda' have to! We've never done anything like this before, but you can watch us try."

I really do wish them the best, since Ubuntu users deserve something good to come of all this. But if we're not all in this together anymore, Canonical should make it clear so we can do something more productive than whine at each other. I'm still hoping we can work together, though- if KDE and GNOME could collaborate on freedesktop.org, and if the various web vendors could collaborate with W3C, we should be able to come out of this with something gained.

None of that are valid complains. Just whining about Canonical not working anymore with other projects that are not part of their goals. All you are saying is that you , as a developer , do not see any benefits from MIR , so Canonical is the evil. Well, keep going. The whining is not productive and is nonsensical. Canonical made quite clear why decided to develop MIR: because it fit best their strategy, they do not want a protocol . I can not think in a better and more valid reason than that, to be honest

[dee.]

Can someone please explain to me what is so bad about MIR. I've looked at it and it looks like a nice improvement. I don't know why everyone bashes it. I'm not trying to defend it, I'm just looking to get educated about what the problem with it is. I mean, there isn't anything else to use righT? There's Weston but that's just a reference thing and not even fully operational display server.

Wayland is a protocol. Why is it bad to have MIR display server based on the protocol. Thanks.

Because Mir does not use the Wayland protocol and is incompatible with it. It would all be fine if Mir were a real Wayland compositor and used the same protocol, there'd be no problem then. But it's not.

It would even be acceptable if Mir just supported the client side protocol, and did whatever on the server side. They could eliminate all the problems, all the confusion and criticism with such a simple solution. But they just won't do it. It's the same thing with the shopping lenses - there too, the answer was simple and easy: just make it opt-in and most people would stop complaining. But no... This behaviour of Canonical where they're ignoring such easy and obvious ways to fix their problems and silence their critics is really baffling. Surely they can see these courses of action, if even I can see them? And yet they just stick to their guns and alienate a large part of the community on which they depend on for their very existence.

[Bathroom Humor]

Normal desktop GUI apps are written to the GTK or KDE API or one of the Java-toolkits like SWT/Swing/JavaFX and don't go down to the X/Wayland/Mir level, so they shouldn't need to be updated or face fragmentation issues.

I should be able to use the same GTK/Swing apps that I currently do regardless of whether my system uses X, Wayland, or Mir.

Am I missing something?

I'm not sure of the specifics, but even in that best-case scenario, where only the toolkits are needed to build for multiple display servers at once, people would need to depend on Canonical successfully and consistently developing and supporting a fork of every toolkit that not only has patched Mir support, but also retains perfect compatibility with the original toolkits so there aren't any nasty bugs or Dev facing building differences, then host the servers and hope devs don't mind compiling their programs two times for Mir or Wayland.
That would greatly ease my stress on the matter, but even then it seems like a big hassle for everyone to worry about, as well as put more strain on Canonical to support more and more projects. :/

[chrisb]

A. Mir doesn't have a protocol. Meaning every change to mir would require rebuilding/installing new versions of every toolkit that uses it or things would break.

I really don't understand all of the complaints about Mir not being a protocol. So what? It's an implementation of a display server with a client side API implemented as a shared library. It like screaming "GTK sucks! It isn't a protocol!" again and again.. no, it isn't a protocol, but who cares? You write your code and link to a shared library. The only time "it isn't a protocol" would matter is if you wanted to create a competing implementation.