Reply to “All the faces of Ubuntu”

Dear Mark Shuttleworth,

so you “have absolutely no doubt that Kwin will work just fine on top of Mir”. This is great and I totally appreciate that you think Mir is a great system. But I’m wondering why you don’t use KWin then, after all it will work fine on top of Mir and is Qt based?

But I have doubt that KWin will work just fine on top of Mir and I have already stated so. You might have wanted to check the facts before stating such claims (somehow I get a feeling for a pattern here).

What makes me think that you cannot make such bold claims:

  • You don’t even know how to write KWin
  • Currently the number of commits to KWin by an Canonical employee is 0 (git log — kwin | grep @canonical)
  • No Canonical employee has so far contacted the KWin team on how we could integrate Mir and whether we are interested at all
  • I have to question the abilities of Canonical to judge what other software can do and cannot after Canonical argued with non existing issues in Wayland for Mir
  • We are still waiting for the Wayland adjustments for KDE done by Canonical. May I remind you:
    We’ll help GNOME and KDE with the transition, there’s no reason for them not to be there on day one either.

I have to ask you to keep KWin out of the pro-Mir campaign. I didn’t ask for Mir, I don’t want Mir and reading blog posts like the one which triggered this reply does lower my motivation to ever have anything to do with Mir. Mir is an answer to a question nobody asked. It’s a solution to problem which does not exist.

Your community manager recently posted on Google+ he had a frustrating day. Guess what my week has been and guess who I can blame. Guess what I great day I will have after reading your blog post this morning.

67 thoughts on “Reply to “All the faces of Ubuntu”

  1. Shuttleworth has become nothing more than a PR head. He’ll say whatever is necessary to get Ubuntu in the foreforth. And that said, I wouldn’t worry or concern myself with it at all.

    If people asked why it didn’t work, you can simply say, “Mark misspoke intentionally about the capabilities of Mir” or be more blunt and say, “Mark lied.”

  2. The they answered with mir question is “How can we copy Apple?”. New desktop environment, new display server, …, new toolkit ?

    1. New Kernel :)
      It would be interesting to see them after a while saying that Linux Kernel doesn’t satisfy their needs and they have to write a new Kernel.
      I doubt that will happen but you can expect anything from Canonical this days…

      1. Canonical write *very* little code and very little of the code they do write will be around in 5 or 10 years. They are a truly ridiculous marketing and branding company desperately masquerading as an engineering company. Anyone with any talent who has been through their doors has promptly left.

    2. Canonical’s moves are more similar to Google, really. Linux with a custom display server and a focus on web apps.

      The vision seems to be pushing ubuntu toward something between ChromeOS and Android, with maybe some legacy compatibility (so, ChromeOS + Android + Windows 8?).

      I guess my description applies to almost any of the big players these days, though.

      1. But that would reduce Canonical to a copy cat. No one can want that, as it would be a risky strategy that completely ignores the first mover advantage.

      2. I’d argue that Google has actual reasoning to make their own stuff that people like and use while Canonical is making Mir for no good reason.

      3. But at least Google are clear about their goals and don’t mislead their “community”. Nor do they make grand, false promises and then develop projects in secret for 6 months behind many contributors’ backs. If they want to go off and do their own thing, fine. But their latest controversies have either been divisive and misleading, wasting people’s time or just plain lying. They are using generous and well meaning people like cannon fodder. Google just do their thing under no false pretences. Canonical are fucking scum. Their “community managers” are sleazy and slimy liars.

  3. First let me say I agree with you.

    All I wanted to say is it is funny to read the end of your post when in your native language the word “mir” means “peace” :)

    1. If I remember my lessons in Russian correctly “mir” not only means “peace” but also “world” or? Its quite a duality .

      1. I don’t know about russian, but it means “peace” in slovenian, croatian and serbian language :P

        1. Well, Đuro – war is peace, no? :D
          Anyway I just want to say thanks to Martin for all his hard and goodf work, and I would not give this kind of chicanery the time of day if I was him.

    2. “mir” means “(to) me” in German (1. person, 3rd case, personal pronoun)

      Pretty much hits the nail regarding Canonical, btw.

  4. Martin if you want to be a Star of Linux, then start being innovative and start hitting the stages promoting the works and future, other wise get behind the leaders and move the cart forward.

    Linux is the future. Take your place.

      1. Just a Phoronix troll and Ubuntu fanboy that thinks that you have to be decapitated for not following Shuttleworth’s vision.
        Just ignore him, it is of no worth listening to that.

      2. i think that canonical leaders wants to do whats in their minds, and they think the world must accept any decision they made !
        my judgement may be biased because i just “don’t love” ubuntu (this is very polite talking :) ).
        my advice is: go ahead and leave canonical leaders say what ever thay want to say; at the end they can’t force you to be in their side while they doesn’t even contribute in other projects they uses for their benefit (linux kernel as a BIG example).

      3. Ignore him. He’s been trolling anyone who will listen. On phoronix, he posted that if you aren’t willing to accept Mir, you should be removed as Kwin mantainer. So yeah, e8hffff is an Ubuntu fanboy troll.

        1. ah it’s him – I read over the phoronix thread earlier and was wondering how people could not get that we don’t accept patches for distro issues. Well fanboyism is a huge issue in free software. It reminds me so much of religion…

          1. I suspect it reminds you so much of religion because it IS religion. Wherever you go, some people will insist upon replacing thought with blind faith. Or acting as though they do for the sole purpose of pissing people off. It’s hard to tell the difference between a dedicated troll and a legitimate religious zealot.

          2. So are you actually saying that if Mir takes off and makes Ubuntu even more popular you will work against it by not maintaining patches to fix distro issues? Is that what you’re saying? That your principles get in the way of helping linux become more popular?

            1. no, that’s not what I’m saying. You might want to read my first blog post to that topic.

  5. I see Mir as one of the most pointless projects ever started, since I don’t see anything in Mir which isn’t possible with Wayland. So it’s an interesting choice developing their own dm protocol which does nothing else than give Canonical the control. As I see it, Mir is the solution to nothing, and could be the cause of some more work – at least if you want to support Ubuntu and use the dm directly, since I doubt anybody else will adopt to Mir over Wayland.

    1. They just want it under that horrible CLA. The one that gives them the rights to turn every project it covers closed-source and proprietary. These guys are something else. It’s almost like they’re sponsored by Microsoft to cause chaos and confusion in the Linux ecosystem or something.

  6. To all those who are shilling for wayland, well, where is it ? Who’s using it ? How long has it been in development and what is its status ? What is its roadmap, support from vendors ( traditionally vendors have been dumped upon by these same people) ?

    I do not use ubuntu (never did) and I didn’t care much for M.S.’s vision until now. He does make sens when he points out the same issue I have raised here. It may also be that many people who worked on xorg have now landed in the wayland project, bringing the same mindset, ideas and the same problems to it.

    I don’t think anyone would have issues with wayland if it was in any form ready for all form factors that linux is relevant on. it isn’t even ready for desktop and if xorg’s history is anything to go by, it never will be. Its development will always trail the market demands.

    1. Wayland is at 1.0 spec. and Qt has bindings for creating compositors for it. I suggest you listen to the latest Luminosity of Free software by Aaron Seigo on youtube for more.

      This is really a cart before the horse problem. Before we can switch to wayland, work needs to be done to integrate/migrate to it. Martin along with many other KDE folks are doin that work right now and soon plasma active followed by the plasma desktop shell will be capable of running on Qt5 and Wayland.

      Once that work is complete, distros can decide if they want to default to Wayland or Xorg.

      Switching out such a core component takes a lot of work and time. Have patience. Mir would take just as much work to swtich to and from what I’ve seen isn’t anywhere close to as feature complete as Wayland.

    2. Also many other entries from Martin’s blog will show you the excellent progress he is making on refactoring Kwin to work with either Xorg or Wayland.

    3. Wayland is there. Toolkits are ported, and there are working WMs already. That the transition to a new dm takes some for huge projects as gnome shell or kwin is clear, and it was known that it will require time for a transition to wayland from the beginning, and I personally expect it to take to 2020 until everything is “wayland native” without X. Transition takes time – sometimes a lot of time, and if a technology has been around for 25 years as X11, it even takes longer.

      Wayland put out a great specification for a dm, a good implementation and a good reference implementation of clients. That’s a work you have to do first, and that wayland takes long to come out with a final specification and implementation is only a sign that they thought carefully about the choices they made – and the guys who develop wayland know the pitfalls of the X11 protocol.

      The short-sight shot of Canonical with Mir isn’t worth anything, if you ask me. It’s based upon wayland on “problems” and “limitations” which simply don’t exist.

      Even not being too much into dms I rather would trust the FOSS guys as Martin, Kristian and others about that than Canonical who want to push forward a company controlled solution.

    4. Go find the answers yourself. If you’re not capable of doing so, you’ll only fall victim to Canonical FUD.

      Wayland itself is basically done. GNOME are aiming at fully supporting it within a year (2 releases). XFCE supports it right now apparently.

      What the fuck were you looking for. Someone to come by your house and give you the full details? If you are a developer, you’d know where to look. If you’re not a developer, you shouldn’t even care about Wayland. The transition for X to Wayland are mostly for the benefit of maintainers not users. I really don’t understand why so many non-developers want to discuss it ad infinitum.

  7. Amazon that Shuttleworth picks the one piece of KDE software that is least likely to ever be on Mir. And the hubris of “And I’m pretty confident Mir will be on a lot more devices than Wayland” is galling. It’s obvious that Shuttleworth is going into a defensive posture and basically not talking sense anymore.

    1. It’s very likely he will go to offensive with mir on mobile devices. And he could be right, but I personally doubt it.

      1. Most of their work is already done for them by Debian and upstreams and they still have trouble meeting release targets and making them stable. I massively doubt they can achieve all the grandiose goals they’ve been dreaming up recently. It’s almost farcical to watch. They’re setting themselves up for a lot of embarrassment and failure in about 1 years time. Shitterworth is going to have multiple omelettes worth of egg on his face.

  8. KWin is big and nice software, but software which you are fixing from the beginning of it’s first release – to be usable – is not a software. Me and my team have never released something as crappy as first KWin, and truth to be told it isn’t better even now. This is the main reason I stopped using KDE after all. And I really like open source but not crappy software where excuse is ‘it’s free and opensource so what do you want’. Just look at the KWin bug list for christ’s sake. I really hope they never have to do anything with KWin or you, because as a leader who considers itself to be a real developer you suck at it. And where are you all now with ‘forks are great; if you don’t like it you can always make your own version; that’s the power of opensource, you can do what you wan’t; blah blah blah – nowhere to be found, is it?

    1. You are of course free to think that our software is buggy or that our bug list is too long. Both doesn’t make it true.

    2. @Peter, who are you? What’s your team?
      @Martin, I guess you allow comments from trolls just for fun, and this really is not affecting you. I’m sure even Shuttleworth would be ashamed for having some of these trolls on his side :) p
      Thank you very much, Martin.

    3. Well, its a long time ago since using KWin; so i couldn’t say anything about how stable or usable it is.

      But what i can say is that your statement about open-source software and forking is true.

  9. I must agree with you all the way. I think it was very prideful and ignorant of Shuttleworth to just assume Kwin will come to Mir, especially since his company does not contribute to KDE. Shuttleworth and his team are taking a blind eye to the needs of the Linux community; in fact, their actions make them seem delusional, a statement which is supported by Shuttleworth’s Kwin comments on his blog. Hopefully, Canonical will come to their senses and stop going ahead with these disruptive activities.

  10. I really couldn’t care if my KDE is running on X, Wayland or Mir. Why waste so much words on this? For me, Kubuntu is working nicely, in (small) part thanks to Mark. If it stops working, for what ever reason, like transitioning to Mir and I couldn’t opt to install X, then I would just move to another distribution.

    I feel free, and want to thank all that make this possible.

    I hope all agree that both Mark and Martin are free to choose what to spend their money and time on.

  11. I think what Shuttleworth is insinuating here is that Mir is so awesome, it’ll get adopted by awestruck hordes; and KWin maintainers would be forced to write their own Mir backend. LoL

    Shuttleworth is just delusional–just like how Steve Jobs believed the world revolves around himself.

  12. We are opensource developers because we choose to do things fun, interesting and differant, yet we scorn each other for that which defines us.

    If your going to dislike someone for something, in particular Mark Shuttleworth do it because he is a “plagiarist” and not because he want’s to do something differant and forge his own path.

    I dislike Ubuntu for the same reason I dislike Kde, Qt, Gtk, Gnome, Martin Graesslin!! ;) they all breed extream fanboyism, and hinder my ability to do things that I find interesting and creative without the severe backlash of those who don’t approve and think I should focuse my work on not re-inviting the wheel.

    Rockets will carry man/woman to the un-explored depth of space not wheels.

    1. In general i agree about “do your own, i do mine” but this one’s special – Canonical is pointing a threat against the move away from X11.
      That’s not like “i like this distro or that DE” – changing the Display Server is nearly like changing the kernel (except for GUIless servers, of course ;-)

      It cannot be hurried or done by a single force (this is why Mir will for sure fall back behind the roadmap on either quality or time) and any disturbance is but one more reason to “stay with X” as has been the last 10 years (and X11 challengers have even tried before)

      This will either introduce a major fraction (on the GUI level) between Ubuntu and everybody else – or we simply stay with X11.
      If this is not Canonicals sheer intention (ie. they’re at war) they should come to their minds, get a reasonable idea about Wayland (protocol) and weston (reference implementation) and check what they consider to be defects and whether it actually is and *especially* whether it’s in the Protocol or the reference compositor.

      If they truely spot issues in the protocol, the sane approach is to point them and propose an alteration.
      If the just believe that weston is shit and they can do by far better: they shall go ahead, much fun.

      But as of now, it looks like they intend to take from android and wayland/weston/… whatever they might need and create an incompatible deviation for the sake of their CLA.

      1. I understand your concern regarding canonical’s actions.

        My previous comment was only for the sake of reminding others that some of us do our own thing for the sake of creativity, enjoyment, scratching an itch and not to be disruptive.

        Personaly I think there intentionaly making ubuntu incompatble(Pulling an Android) for the sake of profit.

        But not to worry there is nothing they can do that we can not do better. Canonical is out gunned in every sence.

  13. I’m with you Martin but I think it is time now to get back to work for the Linux community and gain some distance from the whole Ubuntu/Cannonical mess. There is little we can do about the way Canonical behaves so let’s just focus on the important stuff.

  14. Dear Martin
    I am not an expert in graphical software and in software at all, just a long time user.
    I love what you have been doing with KWin and, this very software.
    Also I love what the Ubuntu were doing with Unity, also very interesting, exciting, inspiring.
    But I can’t stand the “Mir is an answer to a question nobody asked. It’s a solution to problem which does not exist.” reception. Excuse me if my opinion would be proven false in future by I can’t even imagine that these people at Canonical were doing this graphical software Mir to disassemble the Linux distros community at the price of separate market. That is all the possible reason I could see now and it doesn’t hold any criticism. This Mir would cost them big money and time without your and others contribution, you know. And going through this hell for any reason than the solution for the really big technical problem is absurd and very apocalyptic.

    Dear Experts
    I would like to ask the experts here, woudn’t like to bother Martin,

    would the Mir environment mean that common hardware acceleration dependent programs would need some rewriting to work with Mir if they were written for Wayland?
    The same question for X?

    I am sorry if these are silly noobs questions but then please explain where is the actual point of failure for the accelerated programs interoperability between the distros?

    The developers just need to support Qt or some commonly used *GL to Nvidia+AMD bindings and should not worry about that new Wayland or Mir and KWin on something except that these servers both are EGL oriented?

    And now I will ellabouratte!

    As I understood before now, the popular WM and DM and LoginM are kind of the same teer of software that at worst are restricted to be used with the Qt or GTK or else lib and is following some UI conceptions (like popup here or there) and at best are just another software in a RAM that could possibly (if the graphic server able to) work in that Ctrl+F Linux environments or even in the same environment side by side kind of “If Qt and GTK could work side by side, why not”? In Linux community world this teer is “just for looks and feels” and not for the mandatory ruling over the programming environment for accelerated software more than the Composite manager is needed to provide for themes developers?

    I think under that teer are the very complex teer of X server or Wayland or Mir that are doing the heavy job that even Kwin could not manage (by the security design or something else). These are talking to graphic drivers (proprietary, open source or even best from the Linux kernel) and back to these from previous teer, like KWin) for what is needed. So the KWin or not Wayland or not and Qt or not the applications that need better graphical performance and more direct access to the hardware resources Are able to run smoothly on any Linux distro thanks to some *GL foundation bindings and drivers that expose these bindings and that is how it is or this is how it should be for the sake of interoperability?

    I just though that maybe the programs for their interoperability were also dependent on X somehow for any historical reason and now there is a chance to develop the new server that are the new architecture that just work with these *GL bindings in some easy way and that easy way is the only purpose of the server, that is the story?

    If it is right then there would be the AMD’s/Nvidia’s/xxx problem to make another driver for these new Wayland and another for Mir and I imagine that these most proven open source drivers that are embedded in the Kernel itself would not even have this problem. That is pretty much all the problem with another Server around?

    If that was true then I think another Server at this time (when Wayland is not yet the new Standard server) is just another amazingly good FLOSS project that could help to determine the fastest, coolest and most capable new King of the Linux Graphic Servers for everybody to use everywhere?

    Honestly I think I am just a messed user and just too positive and now there would be an ugly truth of One or Another server for that graphic programmers to only capable to work with and so the War but I just hope that I am pretty much right and the problem is not a problem but a big misunderstanding and Martin was just angry about the arrogant hype around his work and about why he should answer these people if he have the work he love and do good and why another server matter so much if we haven’t even used the Wayland (that I was somehow watching all around from the first announce so long ago even so I am just a user) to start with.

    In (the same again, don’t read if you have bothered with my previous words and already have the answers) other words my questions and assumptions boiling down to If the programs are/should be dependent on the *GL bindings that graphical driver provides OR some lib of choice like Qt or GTK that talks with these hard *GL things and then helps the programmer that doesn’t want to talk with the driver (even so there are just standardised *GL functions) and that is all (except the packaging/versioning difficulties) the user need to have for interoperability of his software on any Linux distro ? Does it mean that programmer of let’s say Doom 4 could completely avoid the Server (like that X or better ones) and just program the thing with QT(or else like this) or directly to avoid QT (if QT is overhead itself or because it also runs on that easened stuff from Server I don’t know I am just a user) he (programmer) could code for these driver exposed *GL standardized binding and be happy and effective with this?

    Some sort of table what is under what and where is the interoperability problems would be helpful, if it is not possible and all is in some complex unobvious 43D hierarchy than ok I would do good with simple words explanation. Thank you for your time and attention!

    1. What a mess !

      So Wayland, X and Mir are all providing independent non compatible way to create and manipulate video buffer and input system. So OpenGL doesn’t solve anything as it start to handle things after your window has been created, nor does it handle input also. There is no standard API there. Every toolkit and every application that don’t use a toolkit (Firefox, LibreOffice, Chrome, …) need to be ported.

      Porting an optimized besion of anything take a huge chunk of work, people want to avoid that as much as possible. Also, it is not only that they need to support Mir client, but if KWin want to work nicely on an Ubuntu where all Ubuntu application rely on Mir to get displayed. Then KWin need to handle the Mir protocol to. That’s where things start to get ugly as from now on, if KWin (and so KDE) want to be useful on Ubuntu, they are forced to do a work that Ubuntu don’t and is not going to do.

      Creating it’s own compositing protocol, means only one things, they are parting away from the GNU/Linux stack and they don’t want Free Software to be involved in that anymore. So be it, let them go. There is enough replacement out there.

      1. Thank you! It is becoming a bits clearer for me.
        So even Doom 4 would need to be hardly ported to X, Mir and Wayland to support the major distros if they would go for only one server of choose possibility? What a mess could it be!
        But these are Linux distros, in this world all is customizable, isn’t it possible to pause the main server for the game time, and just run the separate server for the Doom 4, whatever it will be?

        “So Wayland, X and Mir are all providing independent non compatible way to create and manipulate video buffer and input system.”
        that still means that at worst there could be two servers running at the same time (X or Wayland and Mir for Ubuntu) or it is completely impossible by the Wayland or others protocol design? Maybe there is a chance for some protocol parts consensus that would provide possibility for parallel environment?

        How the Wayland provides support for X programs? Wait, I see https://en.wikipedia.org/wiki/File:XWayland.png and https://en.wikipedia.org/wiki/Wayland_(display_server_protocol) Compatibility with X.

        If it is like the X11 was on OSX that is pretty bad as X11 programs lagged there last time I’v seen, there was clear performance hit though maybe due to the poor implementation or poor hardware, but maybe if these servers could run in parallel or at least pause each other for every other window, it could be better?

        I think diversity is a benefit of Linux Community and that “do one thing and do it well” doesn’t mean to avoid variants, just the Rule of Modularity nowadays.

        “forced to do a work that Ubuntu don’t and is not going to do.” this is bad ignorance if so but we don’t know yet, maybe they would need some popular and strictly Wayland programs like some video software or else and then there could be compatibility suit. Besides there are other spin-offs https://upload.wikimedia.org/wikipedia/commons/7/79/UbuntuFamilyTree1210.svg that would need the Mir or Wayland support, that is just a bad part being Ubuntu spin-off. Are there an expert that could say what actually could need that Mir from Ubuntu within Kubuntu, what programs? http://www.kubuntuforums.net/showthread.php?62082-What-the-Mir-is-going-on!!! looks like the only problem is maintaining by own funds and last time I checked Kubuntu there are all parts of KDE, including the KDE Software Manager with all KDE analogs for most popular categories of programs and no Unity or other graphical parts at all maybe except the X11 from Ubuntu repositories. So the clear problem is with the repository programs that all could support just Mir but given the massive I think there indeed would be X programs support for Mir and may well be Wayland.

        “Creating it’s own compositing protocol, means only one things, they are parting away from the GNU/Linux stack and they don’t want Free Software to be involved in that anymore. ”
        If you mean GNU that is FSF main license I agree that they are gone away from the FSF ethic bases that are the bases for the GNU license and more to the Android that has done all this and worse for someone. Still it is a considerable part of community, the most popular and weighty distribution in Linux users world. So Ubuntu acting like from that camp but since they are following the GNU licenses for their programs and kernel they ARE GNU/Linux community more than Android and alienating this company isn’t solving the problem. Maybe the problem resolution could be in finding the better sources of money than Ubuntu do. However Mint is partially adware even so they ran on donations so yes, I think FSF ethics need some better support schemes to be popular, maybe just a FLOSS Kickstarter every year to run (the common place of people gathering to do funding for common good) not that on the FSF.org, that is failing in it. Maybe FSF is ok on the most important roles and there is just people and PR that are making the world spin around the projects like Ubuntu, if so the PR of projects like KDE is failing and project managers are PR so project manager should show some skill and create better waves than aggression against another GNU/Linux project. I need to mention that Kubuntu are benefiting from the Canonical repository work so I can say that Ubuntu is committed to the Kubuntu.
        Since Canonical are going with Launchpad, Software Center, Unity, Semantic desktop under it and other unique stuff for a long time already I conclude that they are just not that company to unite others under the wings and not trying to be. It is sad given their popularity and aura of all-humanity but they keep their work on and on with the Free Software Licenses and this alone deserves real respect.

        Anyway trying to weight this open source world on the wings of some common sense is and would be constantly deceptive for some other so not worth our nerves and just worth the anecdote, dear fellows.

  15. Thank you Martin, I totally agree with you. (http://ikhaya.ubuntuusers.de/2013/03/07/kommentar-wayland-vs-mir/)

    I alwyas thought it’s all about “free for all” & “capture the flag(s)”
    If he will go on trying to force all to play “follow the leader”… this will just end up in a f*#%&§g “team deathmatch” no one could be interested in…

    unity?
    -Zero commits
    -CLA
    -”shopping lens” / affiliate link”
    -Banshee “revenu” http://gburt.blogspot.de/2011/02/banshee-supporting-gnome-on-ubuntu.html
    -upstream
    etc

Comments are closed.