What are the Implications of Kubuntu with Wayland and Ubuntu with Mir

[SteveRiley]

Has anybody checked what Mint is going to do?

Muktware interviewed Clement Lefebvre back in March. The money quote is great:

Mir is irrelevant. Nobody ever heard of it a week ago and plans don't change based on wild speculations. If Ubuntu isn't clear on what they want Ubuntu to be, that's their problem. It has nothing to do with Linux Mint.

xenopeak, a Mint forum admin, has more:

As previously shared by Clem; Linux Mint isn't in the business of gambling on winners. Linux Mint 16 will use Xorg. Like all other distros and desktop environments, except for Ubuntu with Unity, Linux Mint will most likely eventually move to Wayland. Mir is only developed for Ubuntu with Unity currently. It may eventually also be made suitable for other desktop environments and other distros, but that is unlikely. Mir is a one distro + one desktop environment project.

Mir is being developed for Unity specifically, and Unity was developed for Ubuntu specifically. The lack of Unity on other distros is not because everybody doesn't like it (there are many users that do like it), but due to it having been developed for Ubuntu specifically and being hard to port to other distros.

xenopeak confirms what we've mentioned here before: Mir is by Canonical, for Canonical. Not for anyone else. Non-Canoncial entities who wish to use it must realize they're accruing major risk.

The (unknown) writer of Ubuntube, a blog I recently discovered, makes a rather strong accusation:

I am afraid that the decision to develop Mir instead of supporting Wayland is nothing else than a strategical decision to get rid of competitors like Mint which are using Ubuntu as a basis for their distros and to avoid that in the future some bloodsuckers will try to benefit from developments like Ubuntu Touch etc.

Far be it from me to presume the ability to read minds, but...hmmmm...

[vw72]

Has anybody checked what Mint is going to do? One would think they are in the same boat as the various *buntus. Maybe it would be possible to partner with them because I can't see them porting their flagship product which is still Gnome 3 underneath to run on Mir, when Gnome 3 itself is going Wayland.

Just a thought.

[SteveRiley]

Found this twist:

ricegf • 2 days ago Ricegf's Law: Diversity is a feature, not a bug.

Here:
http://www.itwire.com/opinion-and-an...se-is-dividing

Beginning a blog comment with a law named after one's self is possibly not the best way to win folks to one's side of any argument.

ricegf appears eager to orient the universe about himself and thus act dismissive of that which doesn't matter to him. And indeed, as a user of "apps written to development libraries," the choice of display servers matters little -- to him. But the issue here really isn't about ricegf or his ilk, is it? For it is critically important that the *buntu flavors undertake the very discussion we're all observing. Canonical's move matters deeply to those who create (not merely use) the derivatives we know and love. Furthermore, ricegf shows little understanding that, really, the issue ought to matter to him, if he uses one of the flavors.

Oh, and the major example of Android diversity is all the malware that's beginning to appear. Hardly a strong position from which to mount an argument.

[kubicle]

@Fintan:

I'm all for healthy diversity. I like the fact that we have multiple distributions, multiple DEs and multiple applications for particular tasks. However, the difference in these is that I can build and run software that was written by someone using a different distribution and for another DE (let's say someone creates software on Fedora for the Gnome desktop using the Gtk toolkit, I can still use that software in my Kubuntu if I wish)...that kind of diversity creates healthy competition where everyone wins.

Also, healthy diversification happens for good reasons. For example, DEs have their own strengths and weaknesses, and people can have choice whether they go for feature-richness or light-weightness, configurability or simplicity, traditional or modern etc.

Mir, OTOH, is diversity by isolation. If Unity will only run on Mir and Mir will only run Unity, we lose some of the healthy competition where people can pick and choose what they wish to run. Also, the reasons for Mir are muddled at best. Despite the initial FUD on Wayland by Canonical, Mir offers no technical benefits over Wayland....none. It could, in theory, be just as good as Wayland and do just the same thing, but that's not really a valid justification for it.

[Fintan]

Found this twist:

ricegf • 2 days ago Ricegf's Law: Diversity is a feature, not a bug.
This is a *small* molehill around which to build a mountain. Apps are written to development libraries - Qt, GTK+, JVM - rather than a display server. Most desktops can run on multiple display servers; and Mir will still run X applications, just like Wayland. Why should I care until I see a feature that differentiates one display server from another for my own use cases?
And you could make the opposite argument as well. Canonical is porting Unity to Qt, the same library used by KDE, for 14.04. Clearly the "fractures" are *healing* in the *buntu world, no?
From the historical perspective, unification leads to evolutionary dead ends like OSF/1 and CDE (or Windows 8); diversity leads to a thousand diverse mammals like the herd of Androids that breed like rabbits and eat the eggs of the proprietary dinosaurs.
In the end, Linux is dominating virtually all computing market segments
precisely because of the explosive innovation that is fed and watered by
diversity.

Here:
http://www.itwire.com/opinion-and-an...se-is-dividing

[NoWorries]

Kubuntu v Ubuntu: looks like the house is dividing

It is not surprising that the path ahead is now becoming better defined. As mentioned on this forum the future is as outlined in the ITWire article which gives the position of Jonathan Riddell and the KDE developers who will not be using Mir or XMir.

"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."

So SA(B)FL's "Olive Branch" regarding XMir is rejected and it is total commitment to Wayland.

I will be very interested to see how "Polished" Mir is after such a short development cycle.

[SteveRiley]

For you maybe, some of my clients are, and do read this site. Along with many other Linux sites as they are converts not only within their work but personally as well. They are asking me about this, and they don't like the answers.

Do remember that this forum is not official. While two of the admins are also part of the Kubuntu team, we are not the primary maintainers, and this forum is independent from Canoncial, Blue Systems, and the Kubuntu project. As regards KDE's and Kubuntu's futures, only the folks at KDE and those on the Kubuntu core team can speak for that.

Previous actions speak loud and clear. I am only a begrudingly acceptor of KDE4.x. ... No, and again past actions and words or more aptly the rhetoric from wayland speak loud and clear the vision wayland seeks... and it is NOT inline with mine.

And you paid how much for this software? Right. Therefore, who gets to set the vision?

the jist is that wayland devs want to fix X, BUT.. because they don't want to play by the X rules and structure they will take their toys and go to some other sandbox. I am all for improving/updating X... but NOT at the expense of X compatibilty... I am stopping there as this is close to breaking my rules for this.... I and my clients are not on the same page as wayland

And no I don't belive that XDMCP or X forwarding via SSH will be supported.. Politely, I am not interested in the solution to a problem that doesn't exist.

The only person injecting X vs. Wayland debate into this thread is you. Furthermore, impugning the motives of the Wayland developers is rather unbecoming.

KDE won't work with out wayland packages and I am not using any distro which has it installed.

Well, Martin has made his plans very clear: KWin will become a Wayland compositor. This is likely going to affect your long-term plans. Fortunately, no immediate decisions are required, which gives you an opportunity to ensure that the Wayland folks understand your specific needs. So long as Wayland enables new scenarios and also preserves compatibility through an abstraction layer, what's not to love? Certainly there's no need to be bitter.

[SteveRiley]

It seems to me that Ubuntu has a massive amount of stuff in the repositories. How much would it cost them to continue supporting X.org and add Weston/Wayland till we all die of old age?

For starters, take a look at the X-Swat and Xorg-Edgers PPAs. These contain newer X/Mesa/etc. stacks. One possible future is that all maintenance of X moves here after Canonical switches to Mir. Another possible future is that the flavors band together and jointly shoulder the responsibility. It's too soon to know, really. What we do know is that Canonical won't do their own work on X.

I doubt if Canonical will risk pissing off a large portion of FOSS to save a few cents.

I call to your attention Canonical's Contributor License Agreement, to which I linked earlier. Risk does not enter into Canonical's calculus.

BTW, Is it my imagination that Wayland has had a surge of activity sense Canonical started the Mir thing? When did they start that beasty anyway?

Wayland has been in the works for quite a while. Remember one of the goals is not to break API/ABI compatibility. That necessarily means the development of Wayland will take much longer than something new from scratch that can leave the past behind. Much like every other flashy but shallow pole dancer, Mir makes no promises now or in the future.

[lcorken]

@rec9140

12.04, 2017?

Perhaps the dust will settle to your satisfaction or needs by then. In either case at least some of us will likely be here in one form or another. We like to help but can't solve all problems.

Enjoy.

Ken.

[rec9140]

Solution seeking problem

Fortunately, as KFN is not a site that mediates interactions between consultants and clients, we can be less formal here and even encourage folks to relax a bit.

For you maybe, some of my clients are, and do read this site. Along with many other Linux sites as they are converts not only within their work but personally as well. They are asking me about this, and they don't like the answers.

You seem to discount the ability for people to learn from past mistakes and improve themselves.

Previous actions speak loud and clear. I am only a begrudingly acceptor of KDE4.x.

Really? So you don't think that future versions of Wayland might provide everything you need? Or that the effort to achive backwards compatibility will not work?

No, and again past actions and words or more aptly the rhetoric from wayland speak loud and clear the vision wayland seeks... and it is NOT inline with mine. A recently phonorix article enlighted me to the whole wayland mess in a way that few others have till now... in summary the jist is that wayland devs want to fix X, BUT.. because they don't want to play by the X rules and structure they will take their toys and go to some other sandbox. I am all for improving/updating X... but NOT at the expense of X compatibilty... I am stopping there as this is close to breaking my rules for this.... I and my clients are not on the same page as wayland,

To further assuage your angst - read this: http://blog.martin-graesslin.com/blo...-trancparency/

It seems clear to me that X11 network transparancy is being considered and supported. If Martin gets his way - we'll have a better toolkit to replace it.

I am trying to keep this within my own rules and politely... we are on diametrically opposite worlds apart on this. And no I don't belive that XDMCP or X forwarding via SSH will be supported.. it has been very clearly stated by wayland devs that, paraphrasig, "no one uses that!" Well let me get my 4x4 out, now say that again! :mad: I am trying to keep this polite, but do the severe discord between my world and the world of wayland supporters this is going to quickly escalate beyond polite. Politely, I am not interested in the solution to a problem that doesn't exist. Again, you can hold the posts about the X. v. wayland will save the desktop...I've read them all...and I am still not interested. Period.

I know where this is headed KDE won't work with out wayland packages and I am not using any distro which has it installed. I purge enough... and no trotting out Linux Answer #1.1.a "compile or fix etc. yourself !" is not the solution.


I and my clients are going to be at a point I was at not long ago... If KDE is not the path then what is?

As for too soon to jump ship.. My clients and I will upgrade all systems to 12.04 and then all upgrades will cease, till 2017 and either we are no longer on Linux or there will be a KDE like DE on X that is based off *butunu and its wide base of software or we will be moving back to that other OS. :mad:

[NoWorries]

Following this thread is intreguing and informative.

Just found this:
http://www.thepowerbase.com/2013/06/...-ubuntu-13-10/

Just wondering

I was indeed surprised to see this clip. If KDE works in all aspects and not just for those in this demonstration, is Canonical putting out an Olive Branch to the other ubuntu flavours?

The announcement does however indicate that flavors such as Kubuntu, Xubuntu, Lubuntu and others run amicably on XMir, and even goes so far as to assume that those will use it.

As far as I am concerned, the Jury is still out on this issue and I wait to see if the progress can be substantiated and the code open sourced for the experts to see. Please note that I am not one of these!

[lcorken]

It seems to me that Ubuntu has a massive amount of stuff in the repositories. How much would it cost them to continue supporting X.org and add Weston/Wayland till we all die of old age? If Mir and Ubuntu Phone or whatever fails then we may all be going back to Debian/Fedora/Arch...etc. I doubt if Canonical will risk pissing off a large portion of FOSS to save a few cents.

Yes I have been experimenting with other distros (Fedora,Debian and even Arch). They are all viable options. None quite measure up to Kubuntu for me but I am only 66 years old so I still have some adaptability left in these old bones.

BTW, Is it my imagination that Wayland has had a surge of activity sense Canonical started the Mir thing? When did they start that beasty anyway?

Oh well, back to my rum and... What WAS that?

Ken.

[SteveRiley]

That kind of response to my clients would not be, and is not funny.

Fortunately, as KFN is not a site that mediates interactions between consultants and clients, we can be less formal here and even encourage folks to relax a bit.

I've spent a lot A LOOOOOOOT of effort to dump crap OS's, and get them on Linux. Now after only a few years Linux wants to upset the cart and ditch it all!? ? ?

No one is asking you to "ditch" anything. Canonical has a business objective in mind; Mir is an avenue for meeting that objective. One cannot rightfully criticize them for that. But one can critize them for abandoning the flavors; alternately, perhaps it's time for the flavors to envision new strategies for themselves. Such change is the only constant in our universe, I'm afraid.

This is going to turn into the fiasco I expected when this whole wayland situation started.

You seem to discount the ability for people to learn from past mistakes and improve themselves.

Again... let me reiterate, wayland has been weighed and measured and found wanting for my use, and my clients... I and my clients need to know where were headed... and that looks like Linux will not be the future... again save your wayland is the saviour to the world, I am not interested.

Really? So you don't think that future versions of Wayland might provide everything you need? Or that the effort to achive backwards compatibility will not work?

[oshunluvr]

@rec9140: To further assuage your angst - read this: http://blog.martin-graesslin.com/blo...-trancparency/

It seems clear to me that X11 network transparancy is being considered and supported. If Martin gets his way - we'll have a better toolkit to replace it.

[Fintan]

Following this thread is intreguing and informative.

Just found this:
http://www.thepowerbase.com/2013/06/...-ubuntu-13-10/

Just wondering